Refine
Document Type
- Conference Proceeding (194) (remove)
Conference Type
- Konferenzartikel (169)
- Konferenz-Abstract (19)
- Sonstiges (5)
- Konferenz-Poster (1)
Language
- English (194) (remove)
Is part of the Bibliography
- yes (194)
Keywords
- RoboCup (12)
- Machine Learning (9)
- Deep Leaning (7)
- Heart rhythm model (5)
- Modeling and simulation (5)
- Robustness (4)
- machine learning (4)
- Generative Adversarial Network (3)
- Radar (3)
- cryptography (3)
Institute
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (194) (remove)
Open Access
- Closed Access (79)
- Open Access (63)
- Closed (51)
- Bronze (22)
- Diamond (9)
- Grün (3)
- Gold (1)
RETIS – Real-Time Sensitive Wireless Communication Solution for Industrial Control Applications
(2020)
Ultra-Reliable Low Latency Communications (URLLC) has been always a vital component of many industrial applications. The paper proposes a new wireless URLLC solution called RETIS, which is suitable for factory automation and fast process control applications, where low latency, low jitter, and high data exchange rates are mandatory. In the paper, we describe the communication protocol as well as the hardware structure of the network nodes for implementing the required functionality. Many techniques enabling fast, reliable wireless transmissions are used – short Transmission Time Interval (TTI), Time-Division Multiple Access (TDMA), MIMO, optional duplicated data transfer, Forward Error Correction (FEC), ACK mechanism. Preliminary tests show that reliable end-to-end latency down to 350 μs and packet exchange rate up to 4 kHz can be reached (using quadruple MIMO and standard IEEE 802.15.4 PHY at 250 kbit/s).
In bimodal cochlear implant (CI) / hearing aid (HA) users a constant interaural time delay in the order of several milliseconds occurs due to differences in signal processing of the devices. For MED-EL CI systems in combination with different HA types, we have quantified the respective device delay mismatch (Zirn et al. 2015). In the current study, we investigate the effect of the device delay mismatch in simulated and actual bimodal listeners on sound localization accuracy.
To deal with the device delay mismatch in actual bimodal listeners we delayed the CI stimulation according to the measured HA processing delay and two other values. With all delay values highly significant improvements of the rms error in the localization task were observed compared to the test without the delay. The results help to narrow down the optimal patient-specific delay value.
Novel manufacturing technologies, such as printed electronics, may enable future applications for the Internet of Everything like large-area sensor devices, disposable security, and identification tags. Printed physically unclonable functions (PUFs) are promising candidates to be embedded as hardware security keys into lightweight identification devices. We investigate hybrid PUFs based on a printed PUF core. The statistics on the intra- and inter-hamming distance distributions indicate a performance suitable for identification purposes. Our evaluations are based on statistical simulations of the PUF core circuit and the thereof generated challenge-response pairs. The analysis shows that hardware-intrinsic security features can be realized with printed lightweight devices.
Narrowband Internet-of-Things (NB-IoT) is a 3rd generation partnership project (3GPP) standardized cellular technology, adopted for 5G and optimized for massive Machine Type Communication (mMTC). Applications are anticipated around infrastructure monitoring, asset management, smart city and smart energy applications. In this paper, we evaluate the suitability of NB-IoT for private (campus) networks in industrial environments, including complex cloud-based applications around process automation. An end-to-end system has been developed, comprising of a sensor unit connected to a NB-IoT modem, a base station (gNodeB) equipped with a beamforming array and a local (private) network architecture comprising a sensor management system in the edge cloud. The experimental study includes field tests in realistic industrial environments with latency, reliability and coverage measurements. The results show a good suitability of NB-IoT for process automation with high scalability, low-power requirements and moderate latency requirements.
The development of Internet of Things (IoT) embedded devices is proliferating, especially in the smart home automation system. However, the devices unfortunately are imposing overhead on the IoT network. Thus, the Internet Engineering Task Force (IETF) have introduced the IPv6 Low-Power Wireless Personal Area Network (6LoWPAN) to provide a solution to this constraint. 6LoWPAN is an Internet Protocol (IP) based communication where it allows each device to connect to the Internet directly. As a result, the power consumption is reduced. However, the limitation of data transmission frame size of the IPv6 Routing Protocol for Low-power and Lossy Network’s (RPL’s) had made it to be the running overhead, and thus consequently degrades the performance of the network in terms of Quality of Service (QoS), especially in a large network. Therefore, HRPL was developed to enhance the RPL protocol to minimize redundant retransmission that causes the routing overhead. We introduced the T-Cut Off Delay to set the limit of the delay and the H field to respond to actions taken within the T-Cut Off Delay. Thus, this paper presents the comparison performance assessment of HRPL between simulation and real-world scenarios (6LoWPAN Smart Home System (6LoSH) testbed) in validating the HRPL functionalities. Our results show that HRPL had successfully reduced the routing overhead when implemented in 6LoSH. The observed Control Traffic Overhead (CTO) packet difference between each experiment is 7.1%, and the convergence time is 9.3%. Further research is recommended to be conducted for these metrics: latency, Packet Delivery Ratio (PDR), and throughput.
The Internet of Things (IoT) application has becoming progressively in-demand, most notably for the embedded devices (ED). However, each device has its own difference in computational capabilities, memory usage, and energy resources in connecting to the Internet by using Wireless Sensor Networks (WSNs). In order for this to be achievable, the WSNs that form the bulk of the IoT implementation requires a new set of technologies and protocol that would have a defined area, in which it addresses. Thus, IPv6 Low Power Area Network (6LoWPAN) was designed by the Internet Engineering Task Force (IETF) as a standard network for ED. Nevertheless, the communication between ED and 6LoWPAN requires appropriate routing protocols for it to achieve the efficient Quality of Service (QoS). Among the protocols of 6LoWPAN network, RPL is considered to be the best protocol, however its Energy Consumption (EC) and Routing Overhead (RO) is considerably high when it is implemented in a large network. Therefore, this paper would propose the HRPL to enchance the RPL protocol in reducing the EC and RO. In this study, the researchers would present the performance of RPL and HRPL in terms of EC, Control traffic Overhead (CTO) and latency based on the simulation of the 6LoWPAN network in fixed environment using COOJA simulator. The results show HRPL protocol achieves better performance in all the tested topology in terms of EC and CTO. However, the latency of HRPL only improves in chain topology compared with RPL. We found that further research is required to study the relationship between the latency and the load of packet transmission in order to optimize the EC usage.
Neuromorphic computing systems have demonstrated many advantages for popular classification problems with significantly less computational resources. We present in this paper the design, fabrication and training of a programmable neuromorphic circuit, which is based on printed electrolytegated field-effect transistor (EGFET). Based on printable neuron architecture involving several resistors and one transistor, the proposed circuit can realize multiply-add and activation functions. The functionality of the circuit, i.e. the weights of the neural network, can be set during a post-fabrication step in form of printing resistors to the crossbar. Besides the fabrication of a programmable neuron, we also provide a learning algorithm, tailored to the requirements of the technology and the proposed programmable neuron design, which is verified through simulations. The proposed neuromorphic circuit operates at 5V and occupies 385mm 2 of area.
Printed electronics (PE) offers flexible, extremely low-cost, and on-demand hardware due to its additive manufacturing process, enabling emerging ultra-low-cost applications, including machine learning applications. However, large feature sizes in PE limit the complexity of a machine learning classifier (e.g., a neural network (NN)) in PE. Stochastic computing Neural Networks (SC-NNs) can reduce area in silicon technologies, but still require complex designs due to unique implementation tradeoffs in PE. In this paper, we propose a printed mixed-signal system, which substitutes complex and power-hungry conventional stochastic computing (SC) components by printed analog designs. The printed mixed-signal SC consumes only 35% of power consumption and requires only 25% of area compared to a conventional 4-bit NN implementation. We also show that the proposed mixed-signal SC-NN provides good accuracy for popular neural network classification problems. We consider this work as an important step towards the realization of printed SC-NN hardware for near-sensor-processing.
The visualization of heart rhythm disturbance and atrial fibrillation therapy allow the optimization of new cardiac catheter ablations. With the simulation software CST (Computer Simulation Technology, Darmstadt) electromagnetic and thermal simulations can be carried out to analyze and optimize different heart rhythm disturbance and cardiac catheters for pulmonary vein isolation. Another form of visualization is provided by haptic, three-dimensional print models. These models can be produced using an additive manufacturing method, such as a 3D printer. The aim of the study was to produce a 3D print of the Offenburg heart rhythm model with a representation of an atrial fibrillation ablation procedure to improve the visualization of simulation of cardiac catheter ablation.
The basis of 3D printing was the Offenburg heart rhythm model and the associated simulation of cryoablation of the pulmonary vein. The thermal simulation shows the pulmonary vein isolation of the left inferior pulmonary vein with the cryoballoon catheter Arctic Front AdvanceTM from Medtronic. After running through the simulation, the thermal propagation during the procedure was shown in the form of different colors. The three-dimensional print models were constructed on the base of the described simulation in a CAD program. Four different 3D printers are available for this purpose in a rapid prototyping laboratory at the University of Applied Science Offenburg. Two different printing processes were used: 1. a binder jetting printer with polymer gypsum and 2. a multi-material printer with photopolymer. A final print model with additional representation of the esophagus and internal esophagus catheter was also prepared for printing.
With the help of the thermal simulation results and the subsequent evaluation, it was possible to make a conclusion about the propagation of the cold emanating from the catheter in the myocardium and the surrounding tissue. It could be measured that already 3 mm from the balloon surface into the myocardium the temperature drops to 25 °C. The simulation model was printed using two 3D printing methods. Both methods as well as the different printing materials offer different advantages and disadvantages. While the first model made of polymer gypsum can be produced quickly and cheaply, the second model made of photopolymer takes five times longer and was twice as expensive. On the other hand, the second model offers significantly better properties and was more durable overall. All relevant parts, especially the balloon catheter and the conduction, are realistically represented. Only the thermal propagation in the form of different colors is not shown on this model.
Three-dimensional heart rhythm models as well as virtual simulations allow a very good visualization of complex cardiac rhythm therapy and atrial fibrillation treatment methods. The printed models can be used for optimization and demonstration of cryoballoon catheter ablation in patients with atrial fibrillation.
Ensuring that software applications present their users the most recent version of data is not trivial. Self-adjusting computations are a technique for automatically and efficiently recomputing output data whenever some input changes.
This article describes the software architecture of a large, commercial software system built around a framework for coarse-grained self-adjusting computations in Haskell. It discusses advantages and disadvantages based on longtime experience. The article also presents a demo of the system and explains the API of the framework.
PROFINET Security: A Look on Selected Concepts for Secure Communication in the Automation Domain
(2023)
We provide a brief overview of the cryptographic security extensions for PROFINET, as defined and specified by PROFIBUS & PROFINET International (PI). These come in three hierarchically defined Security Classes, called Security Class 1,2 and 3. Security Class 1 provides basic security improvements with moderate implementation impact on PROFINET components. Security Classes 2 and 3, in contrast, introduce an integrated cryptographic protection of PROFINET communication. We first highlight and discuss the security features that the PROFINET specification offers for future PROFINET products. Then, as our main focus, we take a closer look at some of the technical challenges that were faced during the conceptualization and design of Security Class 2 and 3 features. In particular, we elaborate on how secure application relations between PROFINET components are established and how a disruption-free availability of a secure communication channel is guaranteed despite the need to refresh cryptographic keys regularly. The authors are members of the PI Working Group CB/PG10 Security.
Complex tourism products with intangible service components are difficult to explain to potential customers. This research elaborates the use of virtual reality (VR) in the field of shore excursions. A theoretical research model based on the technology acceptance model was developed, and hypotheses were proposed. Cruise passengers were invited to test 360° excursion images on a landing page. Data was collected using an online questionnaire. Finally, data was analyzed using the PLS-SEM method. The results provide theoretical implications on technology acceptance model (TAM) research in the field of cruise tourism. Furthermore, the results and implications indicate the potential of virtual 360° shore excursion presentations for the cruise industry.
One of the challenges for autonomous driving in general is to detect objects in the car's camera images. In the Audi Autonomous Driving Cup (AADC), among those objects are other cars, adult and child pedestrians and emergency vehicle lighting. We show that with recent deep learning networks we are able to detect these objects reliably on the limited Hardware of the model cars. Also, the same deep network is used to detect road features like mid lines, stop lines and even complete crossings. Best results are achieved using Faster R-CNN with Inception v2 showing an overall accuracy of 0.84 at 7 Hz.
When designing and installing Indoor Positioning Systems, several interrelated tasks have to be solved to find an optimum placement of the Access Points. For this purpose, a mathematical model for a predefined number of access points indoors is presented. Two iterative algorithms for the minimization of localization error of a mobile object are described. Both algorithms use local search technique and signal level probabilities. Previously registered signal strengths maps were used in computer simulation.
The conversion of space heating for private households to climate-neutral energy sources is an essential component of the energy transition, as this sector as of 2018 was responsible for 9.4 % of Germany’s carbon dioxide emissions. In addition to reducing demand through better insulation, the use of heat pumps fed with electricity from renewable energy sources, such as on-site photovoltaics (PV) systems, is an important solution approach.
Advanced energy management and control can help to make optimal use of such heating systems. Optimal here can e.g. refer to maximizing self-consumption of self-generated PV power, extended component lifetime or a grid-friendly behavior that avoids load peaks. A powerful method for this is model predictive control (MPC), which calculates optimal schedules for the controllable influence variables based on models of the system dynamics, current measurements of system states and predictions of future external influence parameters.
In this paper, we will discuss three different use cases that show how artificial intelligence can contribute to the realization of such an MPC-based energy management and control system. This will be done using the example of a real inhabited single family home that has provided the necessary data for this purpose and where the methods are implemented and tested. The heating system consists of an air-water heat pump with direct condensation, a thermal stratified storage tank, a pellet burner and a heating rod and provides both heating and hot water. The house generates a significant portion of its electricity needs through a rooftop PV system.
Skin cancer detection proves to be complicated and highly dependent on the examiner’s skills. Millimeter-wave technologies seem to be a promising aid for the detection of skin cancer. The different water content of the skin area affected by cancer compared to healthy skin changes its reflective property. Due to limited available resources on the dielectric properties of skin cancer, especially in comparison to surrounding healthy skin, accurate simulations and evaluations are quite challenging. Therefore, comparing different results for different approaches and starting points can be difficult. In this paper, the Effective Medium Theory is applied to model skin cancer, which provides permittivity values dependent on the water content.
A method for evaluating skin cancer detection based on millimeter-wave technologies is presented. For this purpose, the relative permittivities are calculated using the effective medium theory for the benign and cancerous lesion, considering the change in water content between them. These calculated relative permittivities are further used for the simulation and evaluation of skin cancer detection using a substrate-integrated waveguide probe. A difference in the simulated scattering parameters S 11 of up to 13dB between healthy and cancerous skin can be determined in the best-case.
The Go programming language is an increasingly popular language but some of its features lack a formal investigation. This article explains Go's resolution mechanism for overloaded methods and its support for structural subtyping by means of translation from Featherweight Go to a simple target language. The translation employs a form of dictionary passing known from type classes in Haskell and preserves the dynamic behavior of Featherweight Go programs.
Featherweight Go (FG) is a minimal core calculus that includes essential Go features such as overloaded methods and interface types. The most straightforward semantic description of the dynamic behavior of FG programs is to resolve method calls based on run-time type information. A more efficient approach is to apply a type-directed translation scheme where interface-values are replaced by dictionaries that contain concrete method definitions. Thus, method calls can be resolved by a simple lookup of the method definition in the dictionary. Establishing that the target program obtained via the type-directed translation scheme preserves the semantics of the original FG program is an important task.
To establish this property we employ logical relations that are indexed by types to relate source and target programs. We provide rigorous proofs and give a detailed discussion of the many subtle corners that we have encountered including the need for a step index due to recursive inter- faces and method definitions.
Formal Description of Use Cases for Industry 4.0 Maintenance Processes Using Blockchain Technology
(2019)
Maintenance processes in Industry 4.0 applications try to achieve a high degree of quality to reduce the downtime of machinery. The monitoring of executed maintenance activities is challenging as in complex production setups, multiple stakeholders are involved. So, full transparency of the different activities and of the state of the machine can only be supported, if these stakeholders trust each other. Therefore, distributed ledger technologies, like Blockchain, can be promising candidates for supporting such applications. The goal of this paper is a formal description of business and technical interactions between non-trustful stakeholders in the context of Industry 4.0 maintenance processes using distributed ledger technologies. It also covers the integration of smart contracts for automated triggering of activities.
As industrial networks continue to expand and connect more devices and users, they face growing security challenges such as unauthorized access and data breaches. This paper delves into the crucial role of security and trust in industrial networks and how trust management systems (TMS) can mitigate malicious access to these networks.The TMS presented in this paper leverages distributed ledger technology (blockchain) to evaluate the trustworthiness of blockchain nodes, including devices and users, and make access decisions accordingly. While this approach is applicable to blockchain, it can also be extended to other areas. This approach can help prevent malicious actors from penetrating industrial networks and causing harm. The paper also presents the results of a simulation to demonstrate the behavior of the TMS and provide insights into its effectiveness.
This paper describes a thorough analysis of using PPO to learn kick behaviors with simulated NAO robots in the simspark environment. The analysis includes an investigation of the influence of PPO hyperparameters, network size, training setups and performance in real games. We believe to improve the state of the art mainly in four points: first, the kicks are learned with a toed version of the NAO robot, second, we improve the reliability with respect to kickable area and avoidance of falls, third, the kick can be parameterized with desired distance and direction as input to the deep network and fourth, the approach allows to integrate the learned behavior seamlessly into soccer games. The result is a significant improvement of the general level of play.
It seems to be a widespread impression that the use of strong cryptography inevitably imposes a prohibitive burden on industrial communication systems, at least inasmuch as real-time requirements in cyclic fieldbus communications are concerned. AES-GCM is a leading cryptographic algorithm for authenticated encryption, which protects data against disclosure and manipulations. We study the use of both hardware and software-based implementations of AES-GCM. By simulations as well as measurements on an FPGA-based prototype setup we gain and substantiate an important insight: for devices with a 100 Mbps full-duplex link, a single low-footprint AES-GCM hardware engine can deterministically cope with the worst-case computational load, i.e., even if the device maintains a maximum number of cyclic communication relations with individual cryptographic keys. Our results show that hardware support for AES-GCM in industrial fieldbus components may actually be very lightweight.
With the increasing degree of interconnectivity in industrial factories, security becomes more and more the most important stepping-stone towards wide adoption of the Industrial Internet of Things (IIoT). This paper summarizes the most important aspects of one keynote of DESSERT2020 conference. It highlights the ongoing and open research activities on the different levels, from novel cryptographic algorithms over security protocol integration and testing to security architectures for the full lifetime of devices and systems. It includes an overview of the research activities at the authors' institute.
Wireless communication networks are crucial for enabling megatrends like the Internet of Things (IoT) and Industry 4.0. However, testing these networks can be challenging due to the complex network topology and RF characteristics, requiring a multitude of scenarios to be tested. To address this challenge, the authors developed and extended an automated testbed called Automated Physical TestBed (APTB). This testbed provides the means to conduct controlled tests, analyze coexistence, emulate multiple propagation paths, and model dependable channel conditions. Additionally, the platform supports test automation to facilitate efficient and systematic experimentation. This paper describes the extended architecture, implementation, and performance evaluation of the APTB testbed. The APTB testbed provides a reliable and efficient solution for testing wireless communication networks under various scenarios. The implementation and performance verification of the testbed demonstrate its effectiveness and usefulness for researchers and industry practitioners.
Wireless sensor networks have found their way into a wide range of applications, among which environmental monitoring systems have attracted increasing interests of researchers. Main challenges for these applications are scalability of the network size and energy efficiency of the spatially distributed nodes. Nodes are mostly battery-powered and spend most of their energy budget on the radio transceiver module. In normal operation modes most energy is spent waiting for incoming frames. A so-called Wake-On-Radio (WOR) technology helps to optimize trade-offs between energy consumption, communication range, complexity of the implementation and response time. We already proposed a new protocol called SmartMAC that makes use of such WOR technology. Furthermore, it gives the possibility to balance the energy consumption between sender and receiver nodes depending on the use case. Based on several calculations and simulations, it was predicted that the SmartMAC protocol was significantly more efficient than other schemes being proposed in recent publications, while preserving a certain backward compatibility with standard IEEE802.15.4 transceivers. To verify this prediction, we implemented the SmartMAC protocol for a given hardware platform. This paper compares the realtime performance of the SmartMAC protocol against simulation results, and proves the measured values are very close to the estimated values. Thus we believe that the proposed MAC algorithms outperforms all other Wake-on-Radio MACs.
The authentication method of electronic devices, based on individual forms of correlograms of their internal electric noises, is well-known. Specific physical differences in the components – for example, caused by variations in production quality – cause specific electrical signals, i.e. electric noise, in the electronic device. It is possible to obtain this information and to identify the specific differences of the individual devices using an embedded analog-to-digital converter (ADC). These investigations confirm the possibility to identify and authenticate electronic devices using bit templates, calculated from the sequence of values of the normalized autocorrelation function of noise. Experiments have been performed using personal computers. The probability of correct identification and authentication increases with increasing noise recording duration. As a result of these experiments, an accuracy of 98.1% was achieved for a 1 second-long registration of EM for a set of investigated computers.
In recent times, 5G has found applications in several public as well as private networks. There is a growing need to make it compatible with diverse services without compromising security. Current security options for authenticating devices into a home network are 5G Authentication and Key Agreement (5G-AKA) and Extensible Authentication Protocol (EAP)-AKA'. However, for specific use cases such as private networks, more customizable and convenient authentication mechanisms are required. The current mobile networks use authentication based only on SIM cards, but as 5G is being applied in fields like IIoT and automation, even in Non-Public-Networks (NPNs), there is a need for a simpler method of authentication. Certificate-based authentication is one such mechanism that is passwordless and works solely on the information present in the digital certificate that the user holds. The paper suggests an authentication mechanism that performs certificate-based mutual authentication between the UE and the Home network. The proposed concept identifies both the user and network with digital certificates and intends to carry out primary authentication with the help of it. In this work we conduct a study on presently available authentication protocols for 5G networks, both theoretically and experimentally in hardware as well as virtual environments. On the basis of the analysis a series of proposed steps for certificate primary authentication are presented.
As cyber-attacks and functional safety requirements increase in Operational Technology (OT), implementing security measures becomes crucial. The IEC/IEEE 60802 draft standard addresses the security convergence in Time-Sensitive Networks (TSN) for industrial automation.We present the standard’s security architecture and its goals to establish end-to-end security with resource access authorization in OT systems. We compare the standard to our abstract technology-independent model for the management of cryptographic credentials during the lifecycles of OT systems. Additionally, we implemented the processes, mechanisms, and protocols needed for IEC/IEEE 60802 and extended the architecture with public key infrastructure (PKI) functionalities to support complete security management processes.
To demonstrate how deep learning can be applied to industrial applications with limited training data, deep learning methodologies are used in three different applications. In this paper, we perform unsupervised deep learning utilizing variational autoencoders and demonstrate that federated learning is a communication efficient concept for machine learning that protects data privacy. As an example, variational autoencoders are utilized to cluster and visualize data from a microelectromechanical systems foundry. Federated learning is used in a predictive maintenance scenario using the C-MAPSS dataset.
In recent years, predictive maintenance tasks, especially for bearings, have become increasingly important. Solutions for these use cases concentrate on the classification of faults and the estimation of the Remaining Useful Life (RUL). As of today, these solutions suffer from a lack of training samples. In addition, these solutions often require high-frequency accelerometers, incurring significant costs. To overcome these challenges, this research proposes a combined classification and RUL estimation solution based on a Convolutional Neural Network (CNN) and a Long Short-Term Memory (LSTM) network. This solution relies on a hybrid feature extraction approach, making it especially appropriate for low-cost accelerometers with low sampling frequencies. In addition, it uses transfer learning to be suitable for applications with only a few training samples.
The often-occurring short-term orders of manufactured products require a high machine availability. This requirement increases the importance of predictive maintenance solutions for bearings used in machines. There are, among others, hybrid solutions that rely on a physical model. For their usage, knowing the different degradation stages of bearings is essential. This research analyzes the underlying failure mechanisms of these stages theoretically and in a practical example of the well-known FEMTO dataset used for the IEEE PHM 2012 Data Challenge to provide this knowledge. In addition, it shows for which use cases the usage of low-frequency accelerometers is sufficient. The analysis provides that the degradation stages toward the end of the bearing life can also be detected with low-frequency accelerometers. Further, the importance of high-frequency accelerometers to detect bearing faults in early degradation stages is pointed out. These aspects have not been paid attention to by industry and research until now, despite providing a considerable cost-saving potential.
In this contribution, we propose an system setup for the detection andclassification of objects in autonomous driving applications. The recognition algo-rithm is based upon deep neural networks, operating in the 2D image domain. Theresults are combined with data of a stereo camera system to finally incorporatethe 3D object information into our mapping framework. The detection systemis locally running upon the onboard CPU of the vehicle. Several network archi-tectures are implemented and evaluated with respect to accuracy and run-timedemands for the given camera and hardware setup.
Physically Unclonable Functions (PUFs) are hardware-based security primitives, which allow for inherent device fingerprinting. Therefore, intrinsic variation of imperfect manufactured systems is exploited to generate device-specific, unique identifiers. With printed electronics (PE) joining the internet of things (IoT), hardware-based security for novel PE-based systems is of increasing importance. Furthermore, PE offers the possibility for split-manufacturing, which mitigates the risk of PUF response readout by third parties, before commissioning. In this paper, we investigate a printed PUF core as intrinsic variation source for the generation of unique identifiers from a crossbar architecture. The printed crossbar PUF is verified by simulation of a 8×8-cells crossbar, which can be utilized to generate 32-bit wide identifiers. Further focus is on limiting factors regarding printed devices, such as increased parasitics, due to novel materials and required control logic specifications. The simulation results highlight, that the printed crossbar PUF is capable to generate close-to-ideal unique identifiers at the investigated feature size. As proof of concept a 2×2-cells printed crossbar PUF core is fabricated and electrically characterized.
Printed electronics can add value to existing products by providing new smart functionalities, such as sensing elements over large-areas on flexible or non-conformal surfaces. Here we present a hardware concept and prototype for a thinned ASIC integrated with an inkjet-printed temperature sensor alongside in-built additional security and unique identification features. The hybrid system exploits the advantages of inkjet-printable platinum-based sensors, physically unclonable function circuits and a fluorescent particle-based coating as a tamper protection layer.
Spinal cord stimulation (SCS) is the most commonly used technique of neurostimulation. It involves the stimulation of the spinal cord and is therefore used to treat chronic pain. The existing esophageal catheters are used for temperature monitoring during an electrophysiology study with ablation and transesophageal echocardiography. The aim of the study was to model the spine and new esophageal electrodes for the transesophageal electrical pacing of the spinal cord, and to integrate them in the Offenburg heart rhythm model for the static and dynamic simulation of transesophageal neurostimulation. The modeling and simulation were both performed with the electromagnetic and thermal simulation software CST (Computer Simulation Technology, Darmstadt). Two new esophageal catheters were modelled as well as a thoracic spine based on the dimensions of a human skeleton. The simulation of directed transesophageal neurostimulation is performed using the esophageal balloon catheter with an electric pacing potential of 5 V and a trapezoidal signal. A potential of 4.33 V can be measured directly at the electrode, 3.71 V in the myocardium at a depth of 2 mm, 2.68 V in the thoracic vertebra at a depth of 10 mm, 2.1 V in the thoracic vertebra at a depth of 50 mm and 2.09 V in the spinal cord at a depth of 70 mm. The relation between the voltage delivered to the electrodes and the voltage applied to the spinal cord is linear. Virtual heart rhythm and catheter models as well as the simulation of electrical pacing fields and electrical sensing fields allow the static and dynamic simulation of directed transesophageal electrical pacing of the spinal cord. The 3D simulation of the electrical sensing and pacing fields may be used to optimize transesophageal neurostimulation.
An Empirical Study of Explainable AI Techniques on Deep Learning Models For Time Series Tasks
(2021)
Decision explanations of machine learning black-box models are often generated by applying Explainable AI (XAI) techniques. However, many proposed XAI methods produce unverified outputs. Evaluation and verification are usually achieved with a visual interpretation by humans on individual images or text. In this preregistration, we propose an empirical study and benchmark framework to apply attribution methods for neural networks developed for images and text data on time series. We present a methodology to automatically evaluate and rank attribution techniques on time series using perturbation methods to identify reliable approaches.
Erlang is a functional programming language with dynamic typing. The language offers great flexibility for destructing values through pattern matching and dynamic type tests. Erlang also comes with a type language supporting parametric polymorphism, equi-recursive types, as well as union and a limited form of intersection types. However, type signatures only serve as documentation; there is no check that a function body conforms to its signature.
Set-theoretic types and semantic subtyping fit Erlang’s feature set very well. They allow expressing nearly all constructs of its type language and provide means for statically checking type signatures. This article brings set-theoretic types to Erlang and demonstrates how existing Erlang code can be statically type checked without or with only minor modifications to the code. Further, the article formalizes the main ingredients of the type system in a small core calculus, reports on an implementation of the system, and compares it with other static type checkers for Erlang.
For the past few years Low Power Wide Area Networks (LPWAN) have emerged as key technologies for the connectivity of many applications in the Internet of Things (IoT) combining low-data rates with strict cost and energy restrictions. Especially LoRa/LoRaWAN enjoys a high visibility on today’s markets, because of its good performance and its open community. Originally LoRa was designed for operation within the Sub-GHz ISM bands for Industrial, Scientific and Medical applications. However, at the end of 2018, a LoRa-based solution in the 2.4GHz ISM-band was presented promising higher bandwidths and higher data rates. Furthermore, it overcomes the limited duty-cycle prescribed by the regulations in the ISM-bands and therefore also opens doors to many novel application fields. Also, due to higher bandwidths and shorter transmission times, the use of alternative MAC layer protocols becomes very interesting, i.e. for TDMA based-approaches. Within this paper, we propose a system architecture with 2.4GHz LoRa components combining two aspects. On the one hand, we present a design and an implementation of a 2.4GHz based LoRaWAN solution that can be seamlessly integrated into existing LoRaWAN back-hauls. On the other hand, we describe deterministic setup using a Time Slotted Channel Hopping (TSCH) approach as defined in the IEEE802.15.4-2015 standard for industrial applications. Finally, measurements show the performance of the system.
In recent years, the topic of embedded machine learning has become very popular in AI research. With the help of various compression techniques such as pruning, quantization and others compression techniques, it became possible to run neural networks on embedded devices. These techniques have opened up a whole new application area for machine learning. They range from smart products such as voice assistants to smart sensors that are needed in robotics. Despite the achievements in embedded machine learning, efficient algorithms for training neural networks in constrained domains are still lacking. Training on embedded devices will open up further fields of applications. Efficient training algorithms would enable federated learning on embedded devices, in which the data remains where it was collected, or retraining of neural networks in different domains. In this paper, we summarize techniques that make training on embedded devices possible. We first describe the need and requirements for such algorithms. Then we examine existing techniques that address training in resource-constrained environments as well as techniques that are also suitable for training on embedded devices, such as incremental learning. At the end, we also discuss which problems and open questions still need to be solved in these areas.
Training deep neural networks using backpropagation is very memory and computationally intensive. This makes it difficult to run on-device learning or fine-tune neural networks on tiny, embedded devices such as low-power micro-controller units (MCUs). Sparse backpropagation algorithms try to reduce the computational load of on-device learning by training only a subset of the weights and biases. Existing approaches use a static number of weights to train. A poor choice of this so-called backpropagation ratio limits either the computational gain or can lead to severe accuracy losses. In this paper we present TinyProp, the first sparse backpropagation method that dynamically adapts the back-propagation ratio during on-device training for each training step. TinyProp induces a small calculation overhead to sort the elements of the gradient, which does not significantly impact the computational gains. TinyProp works particularly well on fine-tuning trained networks on MCUs, which is a typical use case for embedded applications. For typical datasets from three datasets MNIST, DCASE2020 and CIFAR10, we are 5 times faster compared to non-sparse training with an accuracy loss of on average 1%. On average, TinyProp is 2.9 times faster than existing, static sparse backpropagation algorithms and the accuracy loss is reduced on average by 6 % compared to a typical static setting of the back-propagation ratio.
This paper presents a novel low-jitter interface between a low-cost integrated IEEE802.11 chip and a FPGA. It is designed to be part of system hardware for ultra-precise synchronization between wireless stations. On physical level, it uses Wi-Fi chip coexistence signal lines and UART frame encoding. On its basis, we propose an efficient communication protocol providing precise timestamping of incoming frames and internal diagnostic mechanisms for detecting communication faults. Meanwhile it is simple enough to be implemented both in low-cost FPGA and commodity IEEE802.11 chip firmware. The results of computer simulation shows that developed FPGA implementation of the proposed protocol can precisely timestamp incoming frames as well as detect most of communication errors even in conditions of high interference. The probability of undetected errors was investigated. The results of this analysis are significant for the development of novel wireless synchronization hardware.
Wireless synchronization of industrial controllers is a challenging task in environments where wired solutions are not practical. The best solutions proposed so far to solve this problem require pretty expensive and highly specialized FPGA-based devices. With this work we counter the trend by introducing a straightforward approach to synchronize a fairly cheap IEEE 802.11 integrated wireless chip (IWC) with external devices. More specifically we demonstrate how we can reprogram the software running in the 802.11 IWC of the Raspberry Pi 3B and transform the receiver input potential of the wireless transceiver into a triggering signal for an external inexpensive FPGA. Experimental results show a mean-square synchronization error of less than 496 ns, while the absolute synchronization error does not exceed 6 μs. The jitter of the output signal that we obtain after synchronizing the clock of the external device did not exceed 5.2 μs throughout the whole measurement campaign. Even though we do not score new records in term of accuracy, we do in terms of complexity, cost, and availability of the required components: all these factors make the proposed technique a very promising of the deployment of large-scale low-cost automation solutions.
Printed Electronics is perceived to have a major impact in the fields of smart sensors, Internet of Things and wearables. Especially low power printed technologies such as electrolyte gated field effect transistors (EGFETs) using solution-processed inorganic materials and inkjet printing are very promising in such application domains. In this paper, we discuss a modeling approach to describe the variations of printed devices. Incorporating these models and design flows into our previously developed printed design system allows for robust circuit design. Additionally, we propose a reliability-aware routing solution for printed electronics technology based on the technology constraints in printing crossovers. The proposed methodology was validated on multiple benchmark circuits and can be easily integrated with the design automation tools-set.
Most machine learning methods require careful selection of hyper-parameters in order to train a high performing model with good generalization abilities. Hence, several automatic selection algorithms have been introduced to overcome tedious manual (try and error) tuning of these parameters. Due to its very high sample efficiency, Bayesian Optimization over a Gaussian Processes modeling of the parameter space has become the method of choice. Unfortunately, this approach suffers from a cubic compute complexity due to underlying Cholesky factorization, which makes it very hard to be scaled beyond a small number of sampling steps. In this paper, we present a novel, highly accurate approximation of the underlying Gaussian Process. Reducing its computational complexity from cubic to quadratic allows an efficient strong scaling of Bayesian Optimization while outperforming the previous approach regarding optimization accuracy. First experiments show speedups of a factor of 162 in single node and further speed up by a factor of 5 in a parallel environment.
Patients with focal ventricular tachycardia are at risk of hemodynamic failure and if no treatment is provided the mortality rate can exceed 30%. Therefore, medical professionals must be adequately trained in the management of these conditions. To achieve the best treatment, the origin of the abnormality should be known, as well as the course of the disease. This study provides an opportunity to visualize various focal ventricular tachycardias using the Offenburg cardiac rhythm model.
Encapsulant-free N.I.C.E. modules have strong ecological advantages compared to conventional laminated modules but suffer generally from lower electrical performance. Via long-term outdoor monitoring of fullsize industrial modules of both types with identical solar cells, we investigated if the performance difference remains constant over time and which parameters influence its value. After assessing about a full year’s data, two obvious levers for N.I.C.E. optimization are identified: The usage of textured glass and transparent adhesives on the module rear side. Also, the performance loss could be alleviated using tracking systems due to lower AOI values. Our measurements show additionally that N.I.C.E. module surfaces are in average about 2.5°C cooler compared to laminated modules. With these findings, we lay out a roadmap to reduce today’s LIV gap of about 5%rel by different optimizations.
Biodegradable metals have entered the implant market in recent years, but still do not show fully satisfactory degradation behaviour and mechanical properties. In contrast, it has been shown that pure molybdenum has an excellent combination of the required properties in this respect. We report on PM based screen printing of thin-walled molybdenum tubes as a processing step for medical stent manufacture. We also present data on the in vivo degradation and biocompatibility of molybdenum. The degradation of molybdenum wires implanted in the aorta of rats was evaluated by SEM and EDX. Biocompatibility was assessed by histological investigation of organs and analysis of molybdenum levels in tissue extracts and body fluids. Degradation rates of up to 13.5 μm/y were observed after 12 months. No histological changes or elevated molybdenum levels in organ tissues were observed. In summary, the results further underline that molybdenum is a highly promising biodegradable metallic material.
Fused Filament Fabrication (FFF) is a widespread additive manufacturing technology, mostly in the field of printable polymers. The use of filaments filled with metal particles for the manufacture of metallic parts by FFF presents specific challenges regarding debinding and sintering. For aluminium and its alloys, the sintering temperature range overlaps with the temperature range of thermal decomposition of many commonly used “backbone” polymers, which provide stability to the green parts. Moreover, the high oxygen affinity of aluminium necessitates the use of special sintering regimes and alloying strategies. Therefore, it is challenging to achieve both low porosity and low levels of oxygen and carbon impurities at the same time. Feedstocks compatible with the special requirements of aluminium alloys were developed. We present results on the investigation of debinding/sintering regimes by Fourier Transform Infrared spectroscopy (FTIR) based In-Situ Process Gas Analysis and discuss optimized thermal treatment strategies for Al-based FFF.
We describe a prototype for power line communi- cation for grid monitoring. The PLC receiver is used to gain information about the PLC channel and the current state of the power grid. The PLC receiver uses the communication signal to obtain an accurate estimate of the current channel and provides information which can be used as a basis for further processing with the aim to detect partial discharges and other anomalies in the grid. This monitoring of the power grid takes advantage of existing PLC infrastructure and uses the data signals, which are transmitted anyway to obtain a real-time measurement of the channel transfer function and the received noise signal. Since this signal is sampled at a high sampling rate compared to simpler measurement sensors, it contains valuable information about possible degradations in the grid which need to be addressed. While channel measurements are based on a received PLC signal, information about partial discharges or other sources of interference can be gathered by a PLC receiver in the absence of a transmit signal. A prototype based on Software Defined Radio has been developed, which implements the simultaneous communication and sensing for a power grid.
The high peak power in comparison to the average transmit power is one of the major long-standing problems in multicarrier modulation and is known as the PAPR (peak to average power ratio) problem. Many PAPR reduction methods have been devised and their comparison is usually based on the complementary cumulative distribution function (CCDF) of the PAPR. While this comparison is straightforward and easy to compute, its relationship with system performance metrics like the (uncoded) BER or the word error rate (WER) for coded systems is considerably more involved. We evaluate the impact of the PAPR on performance metrics like uncoded BER, EVM (error vector magnitude), mutual information and the WER for soft decoding. In this context, we find that system performance is not necessarily degraded by an increasing PAPR. We show that a high number of subcarriers, despite the corresponding high PAPR, is actually not a problem for the system performance and provide a simple explanation for this seemingly counter-intuitive fact.
Background: A disturbed synchronization of the ventricular contraction can cause a highly developed systolic heart failure in affected patients, which can often be explained by a diseased left bundle branch block (LBBB). If medication remains unresponsive, the concerned patients will be treated with a cardiac resynchronization therapy (CRT) system. The aim of this study was to integrate His bundle pacing into the Offenburg heart rhythm model in order to visualize the electrical pacing field generated by His bundle pacing.
Methods: Modelling and electrical field simulation activities were performed with the software CST (Computer Simulation Technology) from Dessault Systèms. CRT with biventricular pacing is to be achieved by an apical right ventricular electrode and an additional left ventricular electrode, which is floated into the coronary vein sinus. This conventional type of biventricular pacing leads to a reduction of the left ventricular ejection fraction. Furthermore, the non-responder rate of the CRT therapy is about one third of the CRT patients.
Results: His bundle pacing represents a physiological alternative to conventional cardiac pacing and cardiac resynchronization. An electrode implanted in the His bundle emits a stronger electrical pacing field than the electrical pacing field of conventional cardiac pacemakers. The pacing of the His bundle was performed by the Medtronic Select Secure 3830 electrode with pacing voltage amplitudes of 3 V, 2 V and 1.5 V in combination with a pacing pulse duration of 1 ms.
Conclusions: Compared to conventional cardiac pacemaker pacing, His bundle pacing is capable of bridging LBBB conduction disorders in the left ventricle. The His bundle pacing electrical field is able to spread via the physiological pathway in the right and left ventricles for CRT with a narrow QRS-complex in the surface ECG.
Artificial intelligence (AI), and in particular machine learning algorithms, are of increasing importance in many application areas but interpretability and understandability as well as responsibility, accountability, and fairness of the algorithms' results, all crucial for increasing the humans' trust into the systems, are still largely missing. Big industrial players, including Google, Microsoft, and Apple, have become aware of this gap and recently published their own guidelines for the use of AI in order to promote fairness, trust, interpretability, and other goals. Interactive visualization is one of the technologies that may help to increase trust in AI systems. During the seminar, we discussed the requirements for trustworthy AI systems as well as the technological possibilities provided by interactive visualizations to increase human trust in AI.
The paper compares different anti-windup strategies for the current control of inverter-fed permanent magnet synchronous machines (PMSM) controlled by pulse-width modulation. In this respect, the focus is on the drive behavior with a relatively large product of stator frequency and sampling time. A requirement for dynamically high-quality anti-windup measures is, among other things, a sufficiently accurate decoupling of the stator current direct axis and quadrature axis components even at high stator frequencies. Discrete-time models of the electrical subsystem of the PMSM are well suited for this purpose, of which the method found to be the most accurate in a preliminary investigation is used as the basis for all anti-windup methods examined. Simulation studies and measurement results document the performance of the compared methods.
We consider the local group of agents for exchanging the time-series data value and computing the approximation of the mean value of all agents. An agent represented by a node knows all local neighbor nodes in the same group. The node has the contact information of other nodes in other groups. The nodes interact with each other in synchronous rounds to exchange the updated time-series data value using the random call communication model. The amount of data exchanged between agent-based sensors in the local group network affects the accuracy of the aggregation function results. At each time step, the agent-based sensor can update the input data value and send the updated data value to the group head node. The group head node sends the updated data value to all group members in the same group. Grouping nodes in peer-to-peer networks show an improvement in Mean Squared Error (MSE).
Enabling ultra-low latency is one of the major drivers for the development of future cellular networks to support delay sensitive applications including factory automation, autonomous vehicles and tactile internet. Narrowband Internet of Things (NB-IoT) is a 3 rd Generation Partnership Project (3GPP) Release 13 standardized cellular network currently optimized for massive Machine Type Communication (mMTC). To reduce the latency in cellular networks, 3GPP has proposed some latency reduction techniques that include Semi Persistent Scheduling (SPS) and short Transmission Time Interval (sTTI). In this paper, we investigate the potential of adopting both techniques in NB-IoT networks and provide a comprehensive performance evaluation. We firstly analyze these techniques and then implement them in an open-source network simulator (NS3). Simulations are performed with a focus on Cat-NB1 User Equipment (UE) category to evaluate the uplink user-plane latency. Our results show that SPS and sTTI have the potential to greatly reduce the latency in NB-IoT systems. We believe that both techniques can be integrated into NB-IoT systems to position NB-IoT as a preferred technology for low data rate Ultra-Reliable Low-Latency Communication (URLLC) applications before 5G has been fully rolled out.
Low latency communication is essential to enable mission-critical machine-type communication (mMTC) use cases in cellular networks. Factory and process automation are major areas that require such low latency communication. In this paper, we investigate the potential of adopting the semi-persistent scheduling (SPS) latency reduction technique in narrowband LTE (NB-LTE) networks and provide a comprehensive performance evaluation. First, we investigate and implement SPS in an open-source network simulator (NS3). We perform simulations with a focus on LTE-M and Narrowband IoT (NB-IoT) systems and evaluate the impact of the SPS technique on the uplink latency of these narrowband systems in real industrial automation scenarios. The performance gain of adopting SPS is analyzed and the results is compared with the legacy dynamic scheduling. Our results show that SPS has the potential to reduce the latency of cellular Internet of Things (cIoT) networks. We believe that SPS can be integrated into LTE-M and NB-IoT systems to support low-latency industrial applications.
Time Sensitive Networking (TSN) provides mechanisms to enable deterministic and real-time networking in industrial networks. Configuration of these mechanisms is key to fully deploy and integrate TSN in the networks. The IEEE 802.1 Qcc standard has proposed different configuration models to implement a TSN configuration. Up until now, TSN and its configuration have been explored mostly for Ethernet-based industrial networks. However, they are still considered “work-in-progress” for wireless networks. This work focuses on the fully centralized model and describes a generic concept to enable the configuration of TSN mechanisms in wireless industrial networks. To this end, a configuration entity is implemented to conFigure the wireless end stations to satisfy their requirements. The proposed solution is then validated with the Digital Enhanced Cordless Telecommunication ultra-low energy (DECT ULE) wireless communication protocol.
TSN, or Time Sensitive Networking, is becoming an essential technology for integrated networks, enabling deterministic and best effort traffic to coexist on the same infrastructure. In order to properly configure, run and secure such TSN, monitoring functionality is a must. The TSN standard already has some preparations to provide such functionality and there are different methods to choose from. We implemented different methods to measure the time synchronisation accuracy between devices as a C library and compared the measurement results. Furthermore, the library has been integrated into the ControlTSN engineering framework.
Background: Pulmonary vein isolation (PVI) using cryoballoon catheters are a recognized method for the treatment of atrial fibrillation (AF). This method offers shorter treatment duration in contrast to the classical therapy with high-frequency (HF) ablation.
Purpose: The aim of this study was to integrate different cryoballoon catheters and a HF catheter into a heart rhythm model and to compare them by means of static and dynamic electromagnetic and thermal simulation in use under AF.
Methods: The cryoballoon catheters from Medtronic and the HF ablation catheter from Osypka were modelled virtually with the aid of manufacturer specifications and the CST (Computer Simulation Technology, Darmstadt) simulation program. The cryoballoon catheter was located in the lower left pulmonary vein of the virtual heart rhythm model for the realization of pulmonary vein isolation (PVI) by cryoenergy. The simulated temperature at the balloon surface was -50°C during the simulation.
Results: During a simulated 20 second application of a cryoballoon catheter at -50°C, a temperature of -24°C was measured at a depth of 0.5 mm in the myocardium. At a depth of 1 mm the temperature was -3°C, at 2 mm depth 18°C and at 3 mm depth 29°C. Under the 15 second application of a RF catheter with a 8 mm electrode and a power of 5 W at 420 kHz, the temperature at the tip of the electrode was 110°C. At a depth of 0.5 mm in the myocardium, the temperature was 75°C, at a depth of 1 mm 58°C, at 2 mm depth 45°C and at 3 mm depth 38°C.
Conclusions: The simulation of temperature profiles during the virtual application of several catheter models in the heart rhythm model allows the static and dynamic simulation of PVI by cryoballoon ablation and RF ablation. The three-dimensional simulation can be used to improve ablation applications by creating a model in personalized cardiac rhythm therapy from MRI or CT data of a heart and finding a favourable position for ablation of AF.
Apache Hadoop is a well-known open-source framework for storing and processing huge amounts of data. This paper shows the usage of the framework within a project of the university in cooperation with a semiconductor company. The goal of this project was to supplement the existing data landscape by the facilities of storing and analyzing the data on a new Apache Hadoop based platform.
In automotive parking scenario, where the curb shall be detected and classified to be traversable or not, radars play an important role. There are different approaches already proposed in other works to estimate the target height. This paper assesses and compares two methods. The first is based on Angle of Arrival (AoA) estimation of input signals of multiple antennas using the Multiple-Input-Multiple-Output (MIMO) principle. The second method uses the geometry in multipath propagation of the radar echo signal for one antenna input. In this work a modified method of calculation of the curb height based on the second method is proposed. The theory of approach is mathematically proved and effectiveness is demonstrated by evaluation of measurements with a 77 GHz Frequency Modulated Continuous Wave (FMCW) radar. In order to evaluate the performance of the introduced method the mean square error (MSE) is used in the proposed scenario. This method, using only one antenna input, produced up to 3.4 times better results for curb height detection in comparison with former methods.
Investigation of the Angle Dependency of Self-Calibration in Multiple-Input-Multiple-Output Radars
(2021)
Multiple-Input-Multiple-Output (MIMO) is a key technology in improving the angular resolution (spatial resolution) of radars. In MIMO radars the amplitude and phase errors in antenna elements lead to increase in the sidelobe level and a misalignment of the mainlobe. As the result the performance of the antenna channels will be affected. Firstly, this paper presents analysis of effect of the amplitude and phase errors on angular spectrum using Monte-Carlo simulations. Then, the results are compared with performed measurements. Finally, the error correction with a self-calibration method is proposed and its angle dependency is evaluated. It is shown that the values of the errors change with an incident angle, which leads to a required angle-dependent calibration.
IoT networks are increasingly used as entry points for cyberattacks, as often they offer low-security levels, as they may allow the control of physical systems and as they potentially also open the access to other IT networks and infrastructures. Existing intrusion detection systems (IDS) and intrusion prevention systems (IPS) mostly concentrate on legacy IT networks. Nowadays, they come with a high degree of complexity and adaptivity, including the use of artificial intelligence. It is only recently that these techniques are also applied to IoT networks. In this paper, we present a survey of machine learning and deep learning methods for intrusion detection, and we investigate how previous works used federated learning for IoT cybersecurity. For this, we present an overview of IoT protocols and potential security risks. We also report the techniques and the datasets used in the studied works, discuss the challenges of using ML, DL and FL for IoT cybersecurity and provide future insights.
Harnessing the overall benefits of the latest advancements in artificial intelligence (AI) requires the extensive collaboration of academia and industry. These collaborations promote innovation and growth while enforcing the practical usefulness of newer technologies in real life. The purpose of this article is to outline the challenges faced during cross-collaboration between academia and industry. These challenges are also inspected with the help of an ongoing project titled “Quality Assurance of Machine Learning Applications” (Q-AMeLiA), in which three universities cooperate with five industry partners to make the product risk of AI-based products visible. Further, we discuss the hurdles and the key challenges in machine learning (ML) technology transformation from academia to industry based on robustness, simplicity, and safety. These challenges are an outcome of the lack of common standards, metrics, and missing regulatory considerations when state-of-the-art (SOTA) technology is developed in academia. The use of biased datasets involves ethical concerns that might lead to unfair outcomes when the ML model is deployed in production. The advancement of AI in small and medium sized enterprises (SMEs) requires more in terms of common tandardization of concepts rather than algorithm breakthroughs. In this paper, in addition to the general challenges, we also discuss domain specific barriers for five different domains i.e., object detection, hardware benchmarking, continual learning, action recognition, and industrial process automation, and highlight the steps necessary for successfully managing the cross-sectoral collaborations between academia and industry.
Sustainable chemical processes should be designed to combine the technological advantages and progress with lower safety risks and minimization of environmental impact such as, for example, reduction of raw materials, energy and water consumption, and avoidance of hazardous waste and pollution with toxic chemical agents. A number of novel eco-friendly chemical technologies have been developed in the recent decades with the help of the eco-innovations approaches and methods such as Life Cycle Analysis, Green Process Engineering, Process Intensification, Process Design for Sustainability, and others. An emerging approach to the sustainable process design in process engineering builds on the innovative solutions inspired from nature. However, the implementation of the eco-friendly technologies often faces secondary ecological problems. The study postulates that the eco-inventive principles identified in natural systems allow to avoid secondary eco-problems and proposes to apply these principles for sustainable design in chemical process engineering. The research work critically examines how this approach differs from the biomimetics, as it is commonly used for copying natural systems. The application of nature-inspired eco-design principles is illustrated with an example of a sustainable technology for extraction of nickel from pyrophyllite.
Amongst all the major hazard aspects for the health of people in big conglomerates is the increase of the particulate matter concentration. Traditional systems for particulate matter (PM) monitoring have a great number of drawbacks but the main issues are economical and are related to the installation costs and never ending periodical maintenance expenses. After all there are installations of such systems but their number is limited and having in mind the growth of population, cities and industry areas, there is even a bigger need for more information on air quality because PM changes non-linearly, has a wide range and different sources. In this paper, we propose an approach, based on low-cost sensor nodes, for real-time measuring and obtaining information about the PM concentration. The adoption of that approach allows for a detailed study of the intensities of pollution and its sources. The system power supply is powered by a PV module. The power supply unit is designed using a model-based design that is a new approach to prototyping power-operated electronic devices with guaranteed performance.
Smart Home or Smart Building applications are a growing market. An increasing challenge is to design energy efficient Smart Home applications to achieve sustainable and green homes. Using the example of the development of an Indoor Smart Gardening system with wireless monitoring and automated watering this paper is discussing in particular the design issue of energy autonomous working sensors and actuators for home automation. Most important part of the presented Smart Gardening system is a 3D printed smart flower pot for single plants. The smart flower pot has integrated a water reservoir for automated plant irrigation and an electronic for monitoring important plant parameters and the water level of the water reservoir. Energy harvesting with solar cells enables energy autonomous working of the flower pot. A low-power wireless interface also integrated in the flowerpot and an external gateway based on a Raspberry Pi 3 enables wireless networking of multiple of those flower pots. The gateway is used for evaluating the plant parameters and as a user interface. Particularly the architecture of the energy autonomous wireless flower pot will be considered, because fully energy autonomous sensors and actuators for home automation could not be implemented without special concepts for the energy supply and the overall electronic.
Recently, RobustBench (Croce et al. 2020) has become a widely recognized benchmark for the adversarial robustness of image
classification networks. In it’s most commonly reported sub-task, RobustBench evaluates and ranks the adversarial robustness of trained neural networks on CIFAR10 under AutoAttack (Croce and Hein 2020b) with l∞ perturbations limited to ϵ = 8/255. With leading scores of the currently best performing models of around 60% of the baseline, it is fair to characterize this benchmark to be quite challenging. Despite it’s general acceptance in recent literature, we aim to foster discussion about the suitability of RobustBench as a key indicator for robustness which could be generalized to practical applications. Our line of argumentation against this is two-fold and supported by excessive experiments presented in this paper: We argue that I) the alternation of data by AutoAttack with l∞, ϵ = 8/255 is unrealistically strong, resulting in close to perfect detection rates of adversarial samples even by simple detection algorithms and human observers.
We also show that other attack methods are much harder to detect while achieving similar success rates. II) That results on low resolution data sets like CIFAR10 do not generalize well to higher resolution images as gradient based attacks appear to become even more detectable with increasing resolutions.
Convolutional neural networks (CNN) define the state-of-the-art solution on many perceptual tasks. However, current CNN approaches largely remain vulnerable against adversarial perturbations of the input that have been crafted specifically to fool the system while being quasi-imperceptible to the human eye. In recent years, various approaches have been proposed to defend CNNs against such attacks, for example by model hardening or by adding explicit defence mechanisms. Thereby, a small “detector” is included in the network and trained on the binary classification task of distinguishing genuine data from data containing adversarial perturbations. In this work, we propose a simple and light-weight detector, which leverages recent findings on the relation between networks’ local intrinsic dimensionality (LID) and adversarial attacks. Based on a re-interpretation of the LID measure and several simple adaptations, we surpass the state-of-the-art on adversarial detection by a significant m argin and reach almost perfect results in terms of F1-score for several networks and datasets. Sources available at: https://github.com/adverML/multiLID
Recently, adversarial attacks on image classification networks by the AutoAttack (Croce and Hein, 2020b) framework have drawn a lot of attention. While AutoAttack has shown a very high attack success rate, most defense approaches are focusing on network hardening and robustness enhancements, like adversarial training. This way, the currently best-reported method can withstand about 66% of adversarial examples on CIFAR10. In this paper, we investigate the spatial and frequency domain properties of AutoAttack and propose an alternative defense. Instead of hardening a network, we detect adversarial attacks during inference, rejecting manipulated inputs. Based on a rather simple and fast analysis in the frequency domain, we introduce two different detection algorithms. First, a black box detector that only operates on the input images and achieves a detection accuracy of 100% on the AutoAttack CIFAR10 benchmark and 99.3% on ImageNet, for epsilon = 8/255 in both cases. Second, a whitebox detector using an analysis of CNN feature maps, leading to a detection rate of also 100% and 98.7% on the same benchmarks.
Detecting Images Generated by Deep Diffusion Models using their Local Intrinsic Dimensionality
(2023)
Diffusion models recently have been successfully applied for the visual synthesis of strikingly realistic appearing images. This raises strong concerns about their potential for malicious purposes. In this paper, we propose using the lightweight multi Local Intrinsic Dimensionality (multiLID), which has been originally developed in context of the detection of adversarial examples, for the automatic detection of synthetic images and the identification of the according generator networks. In contrast to many existing detection approaches, which often only work for GAN-generated images, the proposed method provides close to perfect detection results in many realistic use cases. Extensive experiments on known and newly created datasets demonstrate that the proposed multiLID approach exhibits superiority in diffusion detection and model identification.Since the empirical evaluations of recent publications on the detection of generated images are often mainly focused on the "LSUN-Bedroom" dataset, we further establish a comprehensive benchmark for the detection of diffusion-generated images, including samples from several diffusion models with different image sizes.The code for our experiments is provided at https://github.com/deepfake-study/deepfake-multiLID.
The proposed method includes identification and documentation of the elementary TRIZ inventive principles from the TRIZ body of knowledge, extension and enhancement of inventive principles by patents and technologies analysis, avoiding overlapping and redundant principles, classification and adaptation of principles to at least following categories such as working medium, target object, useful action, harmful effect, environment, information, field, substance, time, and space, assignment of the elementary inventive principles to the at least following underlying engineering domains such as universal, design, mechanical, acoustic, thermal, chemical, electromagnetic, intermolecular, biological, and data processing. The method includes classification of abstraction level of the elementary principles, definition of the statistical ranking of principles for different problem types, and specific engineering or non-technical domains, definition of strategies for selection of principles sets with high solution potential for predefined problems, automated semantic transformation of the elementary inventive principles into solution ideas, evaluation of automatically generated ideas and transformation of ideas to innovation or inventive concepts.
Disturbances of the cardiac conduction system causing reentry mechanisms above the atrioventricular (AV) node are induced by at least one accessory pathway with different conducting properties and refractory periods. This work aims to further develop the already existing and continuously expanding Offenburg heart rhythm model to visualise the most common supraventricular reentry tachycardias to provide a better understanding of the cause of the respective reentry mechanism.
Radio frequency identification (RFID) antennas are popular for high frequency (HF) RFID, energy transfer and near field communication (NFC) applications. Particularly for wireless measurement systems the RFID/NFC technology is a good option to implement a wireless communication interface. In this context, the design of corresponding reader and transmitter antennas plays a major role for achieving suitable transmission quality. This work proves the feasibility of the rapid prototyping of a RFID/NFC antenna, which is used for the wireless communication and energy harvesting at the required frequency of 13.56 MHz. A novel and low-cost direct ink writing (DIW) technology utilizing highly viscous silver nanoparticle ink is used for this process. This paper describes the development and analysis of low-cost printed flexible RFID/NFC antennas on cost-effective substrates for a microelectronic vital parameter measurement system. Furthermore, we compare the measured technical parameters with existing copper-based counterparts on a FR4 substrate.
Subspace clustering aims to find all clusters in all subspaces of a high-dimensional data space. We present a massively data-parallel approach that can be run on graphics processing units. It extends a previous density-based method that scales well with the number of dimensions. Its main computational bottleneck consists of (sequentially) generating a large number of minimal cluster candidates in each dimension and using hash collisions in order to find matches of such candidates across multiple dimensions. Our approach parallelizes this process by removing previous interdependencies between consecutive steps in the sequential generation process and by applying a very efficient parallel hashing scheme optimized for GPUs. This massive parallelization gives up to 70x speedup for
the bottleneck computation when it is replaced by our approach and run on current GPU hardware. We note that depending on data size and choice of parameters, the parallelized part of the algorithm can take different percentages of the overall runtime of the clustering process, and thus, the overall clustering speedup may vary significantly between different cases. However, even
in our ”worst-case” test, a small dataset where the computation makes up only a small fraction of the overall clustering time, our parallel approach still yields a speedup of more than 3x for the complete run of the clustering process. Our method could also be combined with parallelization of other parts of the clustering algorithm, with an even higher potential gain in processing speed.
During the day-to-day exploitation of localization systems in mines, the technical staff tends to incorrectly rearrange radio equipment: positions of devices may not be accurately marked on a map or their positions may not correspond to the truth. This situation may lead to positioning inaccuracies and errors in the operation of the localization system.This paper presents two Bayesian algorithms for the automatic corrections of positions of the equipment on the map using trajectories restored by the inertial measurement units mounted to mobile objects, like pedestrians and vehicles. As a basis, a predefined map of the mine represented as undirected weighted graph was used as input. The algorithms were implemented using the Simultaneous Localization and Mapping (SLAM) approach.The results prove that both methods are capable to detect misplacement of access points and to provide corresponding corrections. The discrete Bayesian filter outperforms the unscented Kalman filter, which, however, requires more computational power.
This paper presents an extended version of a previously published Bayesian algorithm for the automatic correction of the positions of the equipment on the map with simultaneous mobile object trajectory localization (SLAM) in underground mine environment represented by undirected graph. The proposed extended SLAM algorithm requires much less preliminary data on possible equipment positions and uses an additional resample move algorithm to significantly improve the overall performance.
In this paper, we describe a first publicly available fine-grained product recognition dataset based on leaflet images. Using advertisement leaflets, collected over several years from different European retailers, we provide a total of 41.6k manually annotated product images in 832 classes. Further, we investigate three different approaches for this fine-grained product classification task, Classification by Image, by Text, as well as by Image and Text. The approach "Classification by Text" uses the text extracted directly from the leaflet product images. We show, that the combination of image and text as input improves the classification of visual difficult to distinguish products. The final model leads to an accuracy of 96.4% with a Top-3 score of 99.2%. We release our code at https://github.com/ladwigd/Leaflet-Product-Classification.
Analysis of Amplitude and Phase Errors in Digital-Beamforming Radars for Automotive Applications
(2020)
Fundamentally, automotive radar sensors with Digital-Beamforming (DBF) use several transmitter and receiver antennas to measure the direction of the target. However, hardware imperfections, tolerances in the feeding lines of the antennas, coupling effects as well as temperature changes and ageing will cause amplitude and phase errors. These errors can lead to misinterpretation of the data and result in hazardous actions of the autonomous system. First, the impact of amplitude and phase errors on angular estimation is discussed and analyzed by simulations. The results are compared with the measured errors of a real radar sensor. Further, a calibration method is implemented and evaluated by measurements.
In many application areas, Deep Reinforcement Learning (DRL) has led to breakthroughs. In Curriculum Learning, the Machine Learning algorithm is not randomly presented with examples, but in a meaningful order of increasing difficulty. This has been used in many application areas to further improve the results of learning systems or to reduce their learning time. Such approaches range from learning plans created manually by domain experts to those created automatically. The automated creation of learning plans is one of the biggest challenges.In this work, we investigate an approach in which a trainer learns in parallel and analogously to the student to automatically create a learning plan for the student during this Double Deep Reinforcement Learning (DDRL). Three Reward functions, Friendly, Adversarial, and Dynamic based on the learner’s reward are compared. The domain for evaluation is kicking with variable distance, direction and relative ball position in the SimSpark simulated soccer environment.As a result, Statistic Curriculum Learning (SCL) performs better than a random curriculum with respect to training time and result quality. DDRL reaches a comparable quality as the baseline and outperforms it significantly in shorter trainings in the distance-direction subdomain reducing the number of required training cycles by almost 50%.
Active participation of industrial enterprises in electricity markets - a generic modeling approach
(2021)
Industrial enterprises represent a significant portion of electricity consumers with the potential of providing demand-side energy flexibility from their production processes and on-site energy assets. Methods are needed for the active and profitable participation of such enterprises in the electricity markets especially with variable prices, where the energy flexibility available in their manufacturing, utility and energy systems can be assessed and quantified. This paper presents a generic model library equipped with optimal control for energy flexibility purposes. The components in the model library represent the different technical units of an industrial enterprise on material, media, and energy flow levels with their process constraints. The paper also presents a case study simulation of a steel-powder manufacturing plant using the model library. Its energy flexibility was assessed when the plant procured its electrical energy at fixed and variable electricity prices. In the simulated case study, flexibility use at dynamic prices resulted in a 6% cost reduction compared to a fixed-price scenario, with battery storage and the manufacturing system making the largest contributions to flexibility.
Diffracted waves carry high resolution information that can help interpreting fine structural details at a scale smaller than the seismic wavelength. Because of the low signal-to-noise ratio of diffracted waves, it is challenging to preserve them during processing and to identify them in the final data. It is, therefore, a traditional approach to pick manually the diffractions. However, such task is tedious and often prohibitive, thus, current attention is given to domain adaptation. Those methods aim to transfer knowledge from a labeled domain to train the model, and then infer on the real unlabeled data. In this regard, it is common practice to create a synthetic labeled training dataset, followed by testing on unlabeled real data. Unfortunately, such procedure may fail due to the existing gap between the synthetic and the real distribution since quite often synthetic data oversimplifies the problem, and consequently the transfer learning becomes a hard and non-trivial procedure. Furthermore, deep neural networks are characterized by their high sensitivity towards cross-domain distribution shift. In this work, we present deep learning model that builds a bridge between both distributions creating a semi-synthetic datatset that fills in the gap between synthetic and real domains. More specifically, our proposal is a feed-forward, fully convolutional neural network for imageto-image translation that allows to insert synthetic diffractions while preserving the original reflection signal. A series of experiments validate that our approach produces convincing seismic data containing the desired synthetic diffractions.
The recent successes and wide spread application of compute intensive machine learning and data analytics methods have been boosting the usage of the Python programming language on HPC systems. While Python provides many advantages for the users, it has not been designed with a focus on multiuser environments or parallel programming - making it quite challenging to maintain stable and secure Python workflows on a HPC system. In this paper, we analyze the key problems induced by the usage of Python on HPC clusters and sketch appropriate workarounds for efficiently maintaining multi-user Python software environments, securing and restricting resources of Python jobs and containing Python processes, while focusing on Deep Learning applications running on GPU clusters.
Due to the rapidly increasing storage consumption worldwide, as well as the expectation of continuous availability of information, the complexity of administration in today’s data centers is growing permanently. Integrated techniques for monitoring hard disks can increase the reliability of storage systems. However, these techniques often lack intelligent data analysis to perform predictive maintenance. To solve this problem, machine learning algorithms can be used to detect potential failures in advance and prevent them. In this paper, an unsupervised model for predicting hard disk failures based on Isolation Forest is proposed. Consequently, a method is presented that can deal with the highly imbalanced datasets, as the experiment on the Backblaze benchmark dataset demonstrates.
The monitoring of industrial environments ensures that highly automated processes run without interruption. However, even if the industrial machines themselves are monitored, the communication lines are currently not continuously monitored in todays installations. They are checked usually only during maintenance intervals or in case of error. In addition, the cables or connected machines usually have to be removed from the system for the duration of the test. To overcome these drawbacks, we have developed and implemented a cost-efficient and continuous signal monitoring of Ethernet-based industrial bus systems. Several methods have been developed to assess the quality of the cable. These methods can be classified to either passive or active. Active methods are not suitable if interruption of the communication is undesired. Passive methods, on the other hand, require oversampling, which calls for expensive hardware. In this paper, a novel passive method combined with undersampling targeting cost-efficient hardware is proposed.
The fast and cost-effective manufacturing of tools for thermoforming is an essential requirement to shorten the development time of products. Thus, additive processes are used increasingly in tooling for thermoforming of plastic sheets. However, a disadvantage of many additive methods is that they are highly cost-intensive, since complex systems based on laser technology and expensive metal powders are needed. Therefore, this paper examines how to work with favorable additive methods, e.g. Binder Jetting, to manufacture tools, which provide sufficient strength for thermoforming. The use of comparatively low-priced inkjet technology for the layer construction and a polymer plaster as material can be expected to result in significant cost reductions. Based on a case study using a cowling (engine bonnet) for an Unmanned Aerial Vehicle (UAV), the development of a complex tool for thermoforming is demonstrated. The object in this study is to produce a tool for a complex-shaped component in small numbers and high quality in a short time and at reasonable costs. Within the tooling process, integrated vacuum channels are implemented in additive tooling without the need for additional post-processing (for example, drilling). In addition, special technical challenges, such as the demolding of undercuts or the parting of the tool are explained. All process steps from tool design to the use of the additively manufactured tool are analyzed. Based on the manufacturing of a small series of cowlings for a UAV made of plastic sheets (ABS), it is shown, that the Binder Jetting offers sufficient mechanical and thermal strength for additive tooling. In addition, an economic evaluation of the tool manufacturing and a detailed consideration of the required manufacturing times for the different process steps are carried out. Finally, a comparison is made with conventional and alternative additive methods of tooling.
This paper describes the concept and some results of the project "Menschen Lernen Maschinelles Lernen" (Humans Learn Machine Learning, ML2) of the University of Applied Sciences Offenburg. It brings together students of different courses of study and practitioners from companies on the subject of Machine Learning. A mixture of blended learning and practical projects ensures a tight coupling of machine learning theory and application. The paper details the phases of ML2 and mentions two successful example projects.
Current Harmonics Control Algorithm for inverter-fed Nonlinear Synchronous Electrical Machines
(2023)
Current harmonics are a well known challenge of electrical machines. They can be undesirable as they can cause instabilities in the control, generate additional losses and lead to torque ripples with noise. However, they can also be specifically generated in new methods in order to improve the machine behavior. In this paper, an algorithm for controlling current harmonics is proposed. It can be described as a combination of different PI controllers for defined angles of the machine with repetitive control characteristics for whole revolutions. The controller design is explained and important points where linearization is necessary are shown. Furthermore, the limits are analyzed and, for validation, measurement results with a permanently excited synchronous machine on the test bench are considered.
The nonlinear behavior of inverters is mainly influenced by the interlocking and switching times of the semiconductors. In the following work, a method is presented that enables the possibility of an online identification of the switching times of the semiconductors. This information allows a compensation of the non-linear behavior, a reduction of the locking time and can be used for diagnostic purposes. First, a theoretical derivation of the method is made by considering different cases when switching of the inverter and deriving identification possibilities. The method is then extended so that the entire module is taken into account. Furthermore, a possible theoretical implementation is shown. After the methodology has been investigated with possible limitations, boundary conditions and with respect to real hardware, an implementation in the FPGA is performed. Finally, the results are presented, discussed
and further improvements are presented in an outlook.
The nonlinear behavior of inverters is largely impacted by the interlocking and switching times. A method for online identifying the switching times of semiconductors in inverters is presented in the following work. By being able to identify these times, it is possible to compensate for the nonlinear behavior, reduce interlocking time, and use the information for diagnostic purposes. The method is first theoretically derived by examining different inverter switching cases and determining potential identification possibilities. It is then modified to consider the entire module for more robust identification. The methodology, including limitations and boundary conditions, is investigated and a comparison of two methods of measurement acquisition is provided. Subsequently the developed hardware is described and the implementation in an FPGA is carried out. Finally, the results are presented, discussed, and potential challenges are encountered.
Engineering, construction and operation of complex machines involves a wide range of complicated, simultaneous tasks, which potentially could be automated. In this work, we focus on perception tasks in such systems, investigating deep learning approaches for multi-task transfer learning with limited training data. We show an approach that takes advantage of a technical systems’ focus on selected objects and their properties. We create focused representations and simultaneously solve joint objectives in a system through multi-task learning with convolutional autoencoders. The focused representations are used as a starting point for the data-saving solution of the additional tasks. The efficiency of this approach is demonstrated using images and tasks of an autonomous circular crane with a grapple.
In this work, we evaluate two different image clustering objectives, k-means clustering and correlation clustering, in the context of Triplet Loss induced feature space embeddings. Specifically, we train a convolutional neural network to learn discriminative features by optimizing two popular versions of the Triplet Loss in order to study their clustering properties under the assumption of noisy labels. Additionally, we propose a new, simple Triplet Loss formulation, which shows desirable properties with respect to formal clustering objectives and outperforms the existing methods. We evaluate all three Triplet loss formulations for K-means and correlation clustering on the CIFAR-10 image classification dataset.
Estimating the Robustness of Classification Models by the Structure of the Learned Feature-Space
(2022)
Over the last decade, the development of deep image classification networks has mostly been driven by the search for the best performance in terms of classification accuracy on standardized benchmarks like ImageNet. More recently, this focus has been expanded by the notion of model robustness, \ie the generalization abilities of models towards previously unseen changes in the data distribution. While new benchmarks, like ImageNet-C, have been introduced to measure robustness properties, we argue that fixed testsets are only able to capture a small portion of possible data variations and are thus limited and prone to generate new overfitted solutions. To overcome these drawbacks, we suggest to estimate the robustness of a model directly from the structure of its learned feature-space. We introduce robustness indicators which are obtained via unsupervised clustering of latent representations from a trained classifier and show very high correlations to the model performance on corrupted test data.
In this work, we evaluate two different image clustering objectives, k-means clustering and correlation clustering, in the context of Triplet Loss induced feature space embeddings. Specifically, we train a convolutional neural network to learn discriminative features by optimizing two popular versions of the Triplet Loss in order to study their clustering properties under the assumption of noisy labels. Additionally, we propose a new, simple Triplet Loss formulation, which shows desirable properties with respect to formal clustering objectives and outperforms the existing methods. We evaluate all three Triplet loss formulations for K-means and correlation clustering on the CIFAR-10 image classification dataset.
Multiple Object Tracking (MOT) is a long-standing task in computer vision. Current approaches based on the tracking by detection paradigm either require some sort of domain knowledge or supervision to associate data correctly into tracks. In this work, we present a self-supervised multiple object tracking approach based on visual features and minimum cost lifted multicuts. Our method is based on straight-forward spatio-temporal cues that can be extracted from neighboring frames in an image sequences without supervision. Clustering based on these cues enables us to learn the required appearance invariances for the tracking task at hand and train an AutoEncoder to generate suitable latent representations. Thus, the resulting latent representations can serve as robust appearance cues for tracking even over large temporal distances where no reliable spatio-temporal features can be extracted. We show that, despite being trained without using the provided annotations, our model provides competitive results on the challenging MOT Benchmark for pedestrian tracking.