When artificial intelligence (AI) and machine learning (ML) models are applied in healthcare, the ability to understand and explain model decisions is an important aspect. Methods in the field of explainable AI (XAI) have been developed to create explanations for such decisions, which provides transparency and trust to the prediction model. However, the use of XAI-based explanations as added data features for the purpose of improving prediction performance remains a little explored topic. Our proposed Explanation Supported Learning (XSL) framework can improve classification performance for ML models used in medical imaging systems, while also providing a new understanding of how medical images are processed by deep learning (DL) models. The XSL framework consists of novel methods to achieve knowledge transfer from one or several teacher models to a student model. The novelty lies in using explanations from the teacher models, obtained from XAI techniques, as added features when training the student model. This approach enables flexible knowledge transfer between models of different architecture types. We further demonstrate how the XSL framework can be used as a new metric for measuring the quality of the explanations provided by XAI methods. The achievement of increased performance in this framework requires that the chosen XAI technique contains useful information based on the learned understanding of the input data by the teacher models. By testing XSL on the HyperKvasir gastrointestinal image dataset, we achieved significant increases in most of the measured classification metrics, and exceeded most benchmark scores of the HyperKvasir paper. Our code is available on GitHub.

An ageing population and the need of providing adequate care have led to developing robots to relieve healthcare workers and to assist individuals in their own homes. However, the successful integration of robots in such settings relies on more than just ensuring physical safety associated with physical risks (e.g., collisions): it also requires the user’s perceived safety – the users perceiving the robot as not doing any harm. This paper explores the potential influence of a robot gripper’s visual and tactile properties, such as materials and texture, on the users’ perceived safety and comfort of human-robot interaction. An initial survey was distributed to 53 participants, exploring five (n=5) robot gripper designs focusing on the robots’ gripper shape. One design shape was thereafter selected to be constructed as a cover to be placed over the parallel grippers of the TIAGo robot, by using 1) wood filament and 2) plastic. The covers were then tested in an experimental setting with 11 participants. The covers were attached to the TIAGo mobile manipulator robot and participants interacted with both of the designed gripper covers within a controlled laboratory environment. A questionnaire was distributed to all 11 experiment participants, at different stages of the interactions. The findings indicate that the material of the gripper influenced participants’ sense of comfort, familiarity, and perceived capabilities of the robot. The study suggests that perceived safety in human-robot interaction (HRI) is shaped not only by physical factors but also by how materials are personally and contextually interpreted. To better support safe and comfortable interactions, further research is needed to understand how material choices shape users’ perceived safety.

As global efforts to explore the Moon intensify, the need for high-quality 3D lunar maps becomes increasingly critical—particularly for long-distance missions such as NASA’s Endurance mission concept, in which a rover aims to traverse 2,000 km across the South Pole–Aitken basin. Kaguya TC (Terrain Camera) images, though globally available at 10 m/pixel, suffer from altitude inaccuracies caused by stereo matching errors and JPEG-based compression artifacts. This paper presents a method to improve the quality of 3D maps generated from Kaguya TC images, focusing on mitigating the effects of compression-induced noise in disparity maps. We analyze the compression behavior of Kaguya TC imagery, and identify systematic disparity noise patterns, especially in darker regions. In this paper, we propose an approach to enhance 3D map quality by reducing residual noise in disparity images derived from compressed images. Our experimental results show that the proposed approach effectively reduces elevation noise, enhancing the safety and reliability of terrain data for future lunar missions.

Studies foresee a dramatic increase in the elderly population of Western Europe over the next decades, putting pressure on healthcare systems. Healthcare robots are developed to facilitate independent living for elderly people. This article aims to provide a qualitative analysis of recent projects in healthcare robotics (2008–2024) and proposes new research directions for healthcare robots for older adults. We provide an overview of current research and a roadmap for upcoming research. Our study began with a literature search using four databases. Searches were performed for articles from research projects containing the words “elderly care,” “assisted aging,” “health monitoring,” or “elderly health.” Additional exclusion criteria were used to focus on elderly healthcare and utilization of commercial robotic systems. Resulting from this endeavor, 20 recent research projects are described and categorized in this article. Then, these projects were analyzed using thematic analysis. Our findings are summarized in common themes: Most projects have a strong bias towards care robots’ functionalities; robots are often seen as outsiders in care settings; there is an emphasis on robots as commercial products; and there is some limited attention to the design and ethical aspects of care robots, but very little attention to their legal aspects. The article concludes with key points representing a roadmap for future research addressing robotics for the elderly.

Expressive robot motion can help establish acceptance of this technology in everyday life, but understanding what makes movement expressive is a complex and multifaceted task. This paper presents the results of an online study with 46 participants, it aims to explore how people perceive and interpret the expressive qualities of human movement and how they envision the translation of their description into an imagined non-humanoid, quadrupedal robot. Through a qualitative analysis of responses, we conceptualize three themes: their understanding of intent, their interpretations of movement qualities, and finally, their translation from human to robot movement. Respondents’ descriptions of their initial understanding of the performer’s intent fall into two modes, bio-mechanical and narrative. We illustrate their interpretations of movement qualities through four strategies: movement features as kinematic indicators, intent indicators, attributed context, and perceived internal states. Lastly, we observe their translation from human to robot movement, with a particular focus on respondents’ use of kinaesthetic empathy and anthropomorphism. Our findings aim to support a bottom-up approach, using users’ general knowledge for designing expressive robot motion.

Striking a balance between utility and privacy holds significant importance in systems that rely on sensor utilization, such as robots. This balance is even more vital in care robots, given the sensitivity of personal data and the necessity for privacy-preserving monitoring to ensure user comfort. This paper presents a comprehensive investigation into the utility-privacy trade-off concerning different vision-based sensors. Specifically, RGB cameras, color and mono-color thermal cameras, and depth sensors are compared, considering technical aspects and users’ perception of privacy. The technical analysis addresses human pose tracking, human presence detection, human vital sign monitoring, and human facial and emotion recognition. The quantitative examination of sensors in real-life scenarios highlights the mono-color thermal camera’s effectiveness for user monitoring. Particularly, this sensor excels in challenging human presence detection scenarios compared to RGB cameras. Furthermore, interview and survey studies, encompassing two different age groups were carried out to compare how sensors are perceived in terms of user privacy. The quantitative and qualitative assessments of users’ feedback in these studies reveal that apart from depth sensors, thermal mono-color, and thermal color sensors are perceived as better at preserving user privacy compared to RGB cameras. The analysis includes the influence of participant age on privacy perception, indicating non-significant effects. Considering both the technical assessment and user preferences, the mono-color thermal camera emerges as the optimal choice for human monitoring purposes.

The search for refining 3D LiDAR data has attracted growing interest motivated by recent techniques such as supervised learning or generative model-based methods. Existing approaches have shown the possibilities for using diffusion models to generate refined LiDAR data with high fidelity, although the performance and speed of such methods have been limited. These limitations make it difficult to execute in real-time, causing the approaches to struggle in real-world tasks such as autonomous navigation and human-robot interaction. In this work, we introduce a novel approach based on conditional diffusion models for fast and high-quality sparse-to-dense upsampling of 3D scene point clouds through an image representation. Our method employs denoising diffusion probabilistic models trained with conditional inpainting masks, which have been shown to give high performance on image completion tasks. We introduce a series of experiments, including multiple datasets, sampling steps, and conditional masks. This paper illustrates that our method outperforms the baselines in sampling speed and quality on upsampling tasks using the KITTI-360 dataset. Furthermore, we illustrate the generalization ability of our approach by simultaneously training on real-world and synthetic datasets, introducing variance in quality and environments.

Successfully implementing robots to support senior adults requires their acceptance. Leveraging nonverbal communication could enhance the ease and intuitiveness of accepting robot assistance. However, it is essential to see how different age groups understand nonverbal communication cues to understand the dynamics between different user groups and assistive robots. Our research specifically delves into the intuitive understanding of handshaking gestures across multiple interactions, focusing on seniors (between 70 and 97) and young (21 and 26) adults. Through a combination of observations and open-ended surveys, we conducted a video observation and thematic analysis. Interestingly, our findings indicate no significant differences between the two age groups, except for reactions and interaction time variables. Furthermore, we report on possible motivations behind the initial reactions in the two age groups, familiarity, and ways to improve the overall Human-Robot Interaction experience potentially.

This work presents a reinforcement learning-based switching control mechanism to autonomously move a ferromagnetic object (representing a milliscale robot) around obstacles within a constrained environment in the presence of disturbances. This mechanism can be used to navigate objects (e.g., capsule endoscopy, swarms of drug particles) through complex environments when active control is a necessity but where direct manipulation can be hazardous. The proposed control scheme consists of a switching control architecture implemented by two sub-controllers. The first sub-controller is designed to employ the robot’s inverse kinematic solutions to do an environment search for the to-be-carried ferromagnetic particle while being robust to disturbances. The second sub-controller uses a customized rainbow algorithm to control a robotic arm, i.e., the UR5 robot, to carry a ferromagnetic particle to a desired position through a constrained environment. For the customized Rainbow algorithm, Quantile Huber loss from the Implicit Quantile Networks (IQN) algorithm and ResNet are employed. The proposed controller is first trained and tested in a real-time physics simulation engine (PyBullet). Afterward, the trained controller is transferred to a UR5 robot to remotely transport a ferromagnetic particle in a real-world scenario to demonstrate the applicability of the proposed approach. The experimental results on the UR5 robot show an average success rate of 98.86% over 30 episodes for randomly generated trajectories, demonstrating the viability of the proposed approach for real-life applications. In addition, two classical path finding approaches, Attractor Dynamics and the execution extended Rapidly-Exploring Random Trees (ERRT), are also investigated and compared to the RL-based method. The proposed RL-based algorithm is shown to achieve performance comparable to that of the tested classical path planners whilst being more robust to deploy in dynamical environments. Note to Practitioners —Deep reinforcement learning methods have been widely applied in computer games and simulations. However, employing these algorithms for practical, real-world applications such as robotics becomes challenging due to the difficulty of obtaining training samples. This paper predominantly focuses on bridging the gap between simulations and the real-world implementation of a reinforcement learning algorithm for a robotic application in the context of miniaturized drug delivery robots and robotic capsule endoscopes. This paper presents the derivation and experimental validation of a reinforcement learning-based algorithm for controlling a magnetically-actuated small-scale robot within a simplified model of the large intestine in the presence of disturbances. We demonstrate the possibility of training a high-fidelity reinforcement learning algorithm fully within a simulated environment before deploying it as-is in a real-world scenario by carrying out different experiments and simulations. Implementing the presented control framework complements a large body of this work, and the results offer a feasibility study of using reinforcement learning algorithms in practice.

What expectations exist in the minds of dancers when interacting with a generative machine learning model? During two workshop events, experienced dancers explore these expectations through improvisation and role-play, embodying an imagined AI-dancer. The dancers explored how intuited flow, shared images, and the concept of a human replica might work in their imagined AI-human interaction. Our findings challenge existing assumptions about what is desired from generative models of dance, such as expectations of realism, and how such systems should be evaluated. We further advocate that such models should celebrate non-human artefacts, focus on the potential for serendipitous moments of discovery, and that dance practitioners should be included in their development. Our concrete suggestions show how our findings can be adapted into the development of improved generative and interactive machine learning models for dancers’ creative practice.

The field of synthetic medical data has become increasingly important due to the urgent need for large and diverse datasets in the medical sector. Using diffusion models in data generation has created more authentic and varied medical data. In this study, a framework is presented that utilizes diffusion models trained on openly accessible data to generate realistic-looking colon polyps, along with their corresponding ground truth masks. The usefulness of the synthetic polyps is evaluated by using them to train segmentation models designed to segment colon polyps in real-world images. The results demonstrate that the generated synthetic data is highly accurate and suggest that including synthetic polyps in the training dataset improves the predictive performance and generalization of the segmentation models. When the training dataset consists of pre-generated synthetic data from our model, we achieve a mean intersection over union (mIoU) improvement of 4.64% on the validation data and a 4.14% mIoU improvement when testing across different datasets. These results indicate that generating synthetic medical data using diffusion models is valuable for addressing the need for diverse and extensive medical datasets.

Robots have the potential to provide everyday life care and support for senior adults, but acceptance is essential for successful implementation in the domestic environment. Nonverbal social behavior can enhance this acceptance, and behavioral cues should be easy and intuitive to understand. However, which factors contribute to senior adults’ intuitive understanding of social cues, such as handshakes? Our research aims to address this question using video observations and semi-structured interviews. Based on a thematic analysis and video observations, our findings indicate that some participants intuitively understood how to shake hands. Most did not shake hands due to not understanding the robot’s behavior or fear. Other identified themes included: contributing features for intuitive handshakes, design improvements, and experiences with the robot’s end effector. Lastly, we found no significant effect between the initial response of the participants to the handshake and either the reaction time or the handshake duration. By designing the gripper and the robot itself in a more familiar, less fear-eliciting way, senior adults might understand the gesture of shaking hands more intuitively.

We used the story dialog method (SDM) to gather the viewpoints of health professionals about the use of social robots in the home and healthcare services with vulnerable users. SDM consists of participants bringing stories that they discuss together. The aim of the study was to address universal design and accessibility issues with robots in specific use situations. We used three social robots in four stories: TIAGo, Romibo and robot pets. We used the SDM method in two workshops with eight participants. The participants uncovered issues regarding ethics, responsibility, use of data, infrastructure, design, and user concerns based on provided stories and their own experiences. These issues provide important aspects that researchers and roboticists should consider when using robots with vulnerable users and ensuring that a robot is usable by as many people as possible.

Internet-delivered psychological treatments (IDPT) are seen as an effective and scalable pathway to improving the accessibility of mental healthcare. Within this context, treatment adherence is an especially pertinent challenge to address due to the reduced interaction between healthcare professionals and patients. In parallel, the increase in regulations surrounding the use of personal data, such as the General Data Protection Regulation (GDPR), makes data minimization a core consideration for real-world implementation of IDPTs. Consequently, this work proposes a Self-Attention-based deep learning approach to perform automatic adherence forecasting, while only relying on minimally sensitive login/logout-timestamp data. This approach was tested on a dataset containing 342 patients undergoing Guided Internet-delivered Cognitive Behavioral Therapy (G-ICBT) treatment. Of these 342 patients, 101 ( ∼ 30%) were considered non-adherent (dropout) based on the adherence definition used in this work (i.e. at least eight connections to the platform lasting more than a minute over 56 days). The proposed model achieved over 70% average balanced accuracy, after only 20 out of the 56 days ( ∼ 1/3) of the treatment had elapsed. This study demonstrates that automatic adherence forecasting for G-ICBT, is achievable using only minimally sensitive data, thus facilitating the implementation of such tools within real-world IDPT platforms.

This work introduces a class of network performance models designed to capture variations in network quality on diverse timescales. By explicitly modeling how quality changes over time, the proposed models enable computation of performance metrics that are beyond the scope of steady-state methods such as Markov chains. We use the quality attenuation (ΔQ) metric to quantify network quality, and ΔQ generative models specify how quality attenuation varies over time. Variation over time is modeled using a finite state machine with timed state transitions. We show how the models can be used to shed light on practical problems by presenting novel results for the problem of buffer sizing. In addition to the buffer sizing results, this work presents the ΔQ generative model structure and the basic algorithms needed to work with the models.

This paper establishes a physics-based simulation framework for steering a magnetically actuated guidewire based on the linear elasticity and dipoles theories. Interaction wrenches resulting from an external magnetic field and embedded magnets in a continuum rod, i.e., guidewire, serves as actuators for steering. In the presented framework, a simplified integration scheme based on the finite-volume method is employed to model guidewire using the linear elasticity theory and forces resulting from the interference of magnetic fields to provide a rapid model reconstruction. Furthermore, orienting the external magnetic field is employed to steer a guidewire into a constrained environment. Finally, simulations illustrate the approach performance on a soft rod where an external magnetic field is orientated to form the desired shape for a continuum rod and steer it within an environment. The results open up possibilities to construct a rapid model for continuum manipulators in practice.

WiFi is one of the most widely deployed networking technologies, and understanding WiFi performance is therefore of great importance. The WiFi MAC layer sometimes introduces significant and variable delays. No existing models of the WiFi protocol describe WiFi performance in terms of complete latency distributions. In this work, we present a novel model of WiFi performance. We explicitly define our model in terms of the latency introduced at each step in the protocol state machine, and the model produces complete latency distributions. We validate the model by comparing its outputs to previous modeling work and real-world measurements. Finally, we use our results to quantify the latency distribution of WiFi as a function of the duration of transmit opportunities and the number of stations competing for the channel. Quantifying this relation represents a significant improvement in our understanding of WiFi performance that would not be possible with existing models.

This paper presents a Nonlinear Model Predictive Control-based Reinforcement Learning (NMPC-based RL) framework for robot manipulators. The controller is developed to address the motion planning problem for robot manipulators in the presence of obstacles. The proposed control scheme includes a parametrized NMPC structure used as an approximator for the RL framework’s value function and action-value function. In the NMPC structure, the cost function, system constraints, and the manipulator’s model are parameterized. The Q-Learning algorithm based on the Temporal Difference method adjusts the parameters of the NMPC to increase the closed-loop performance of the whole control scheme. The controller has been applied to a 6-degrees-of-freedom (DoF) model of a robot manipulator, aimed at moving its end-effector to reach the desired pose when static obstacles are in the robot’s workspace. Numerical simulations demonstrate that the proposed controller can effectively control the end-effector’s pose in such a way as to avoid any collisions between the manipulator and the obstacles. It is shown that the learning capability of the proposed NMPC-based RL framework can enhance the efficiency of the control loop up to 21%.

Ethical considerations are getting increased attention with regards to providing responsible personalization for robots and autonomous systems. This is partly a result of the currently limited deployment of such systems in human support and interaction settings. There are many different ethical considerations, and it is important to identify those relevant to one's own work within user modelling and personalization. The tutorial paper will give an overview of the most commonly expressed ethical challenges and ways being undertaken to reduce their impact using the findings in an earlier undertaken review supplemented with recent work and initiatives. That includes the identified challenges in a “Statement on research ethics in artificial intelligence”.

Through dance, a wide range of emotions can be expressed. As virtual agents and robots continue to become part of our daily lives, the need for them to efficiently convey emotion and intent increases. When trained to dance, to what extent can AI learn to model the tacit mappings between sound and motion? Here, we explore the creative capacity of a generative model trained on 3D motion capture recordings of improvised dance. We perform a perceptual judgment experiment wherein respondents rate movement generated by our model as well as human performances. While the sound-motion mappings remain somewhat elusive, particularly when compared to examples of human dance, our study shows that in certain aspects related to perceived dance-likeness and expressivity, the model successfully mimics human dance movement. By employing a perceptual study to evaluate our generative model, we aim to further our ability to understand the affordances and limitations of creative AI.

WiFi is a crucial part of internet infrastructure. Performance issues are common in WiFi networks, but well-tested solutions exist for some known problems. In this work, we look for evidence of known WiFi issues in chipsets commonly found in consumer WiFi routers. We document serious performance problems in most of the tested WiFi chipsets and point to existing solutions for the detected problems. The prevalence of these problems has implications for network management, network research, and for users and creators of performance-sensitive network applications running over WiFi. To our knowledge, this is the most comprehensive documentation of these issues to date.

Introducing robots in healthcare facilities and homes may reduce the workload of healthcare personnel while providing the users with better and more available services. It may also contribute to interactions that are engaging and safe against transmitting contagious diseases for senior adults. A major challenge in this regard is to design and adapt the robot’s behavior based on the requirements and preferences of the different users. In this paper, we report a conducted use study on how people perceive different kinds of robot encounters. We had two groups of target users: one with senior residents at a care center and another with young students at a university, which would be representative for the visitors and care volunteers in the facility. Several common scenarios have been created to evaluate the perception of the robot’s behavior by the participants. Two sets of questionnaires were used to collect feedback on the behavior and the general perception of the users about the robot´s different styles of behavior. An exploratory analysis of the effect of age shows that the age of the targeted user group should be considered as one of the main criteria when designing the social parameters of a care robot, as seniors preferred slower speed and closer distance to the robot. The results can contribute to improving a future robot’s control to better suit users from different generations.

Ethical considerations are getting increased attention with regards to providing responsible personalization for robots and autonomous systems. This is partly as a result of the currently limited deployment of such systems in human support and interaction settings. The tutorial will give an overview of the most commonly expressed ethical challenges and ways being undertaken to reduce their impact using the findings in an earlier undertaken review supplemented with recent work and initiatives. The tutorial will exemplify the challenges related to privacy, security and safety through several examples from own and others’ work.

Robots operating in the real world will experience a range of different environments and tasks. It is essential for the robot to have the ability to adapt to its surroundings to work efficiently in changing conditions. Evolutionary robotics aims to solve this by optimizing both the control and body (morphology) of a robot, allowing adaptation to internal, as well as external factors. Most work in this field has been done in physics simulators, which are relatively simple and not able to replicate the richness of interactions found in the real world. Solutions that rely on the complex interplay among control, body, and environment are therefore rarely found. In this article, we rely solely on real-world evaluations and apply evolutionary search to yield combinations of morphology and control for our mechanically self-reconfiguring quadruped robot. We evolve solutions on two distinct physical surfaces and analyze the results in terms of both control and morphology. We then transition to two previously unseen surfaces to demonstrate the generality of our method. We find that the evolutionary search finds high-performing and diverse morphology-controller configurations by adapting both control and body to the different properties of the physical environments. We additionally find that morphology and control vary with statistical significance between the environments. Moreover, we observe that our method allows for morphology and control parameters to transfer to previously unseen terrains, demonstrating the generality of our approach.

Terrain knowledge around unmanned ground vehicles (UGVs) is vital for autonomous navigation. Having global understanding of the surroundings of UGVs is important, although the field of view from UGVs is very limited. Thus, we utilize an aerial vehicle to provide a large terrain map from sequential aerial images. In this paper, we present multiple techniques to accelerate the process of terrain classification so that it can run onboard on the aerial platform. The main techniques used to accelerate the process is a "knowledge distillation" of a deep neural net to a shallower one, and a super pixel implementation. We evaluated our system on Jetson TX1 with actual images collected from a weather balloon which confirmed the effectiveness of the proposed system.

Ethical considerations are getting increased attention with regards to providing responsible personalization for robots and autonomous systems. This is partly as a result of the currently limited deployment of such systems in human support and interaction settings. The tutorial will give an overview of the most commonly expressed ethical challenges and ways being undertaken to reduce their impact using the findings in an earlier undertaken review supplemented with recent work and initiatives. The tutorial will exemplify the challenges related to privacy, security and safety through several examples from own and others' work.Ethics, Robotics, Autonomous systems, Privacy, Security and Safety

The structure and performance of neural networks are intimately connected, and by use of evolutionary algorithms, neural network structures optimally adapted to a given task can be explored. Guiding such neuroevolution with additional objectives related to network structure has been shown to improve performance in some cases, especially when modular neural networks are beneficial. However, apart from objectives aiming to make networks more modular, such structural objectives have not been widely explored. We propose two new structural objectives and test their ability to guide evolving neural networks on two problems which can benefit from decomposition into subtasks. The first structural objective guides evolution to align neural networks with a user-recommended decomposition pattern. Intuitively, this should be a powerful guiding target for problems where human users can easily identify a structure. The second structural objective guides evolution towards a population with a high diversity in decomposition patterns. This results in exploration of many different ways to decompose a problem, allowing evolution to find good decompositions faster. Tests on our target problems reveal that both methods perform well on a problem with a very clear and decomposable structure. However, on a problem where the optimal decomposition is less obvious, the structural diversity objective is found to outcompete other structural objectives-and this technique can even increase performance on problems without any decomposable structure at all.

Neural networks have been widely used in agent learning architectures; however, learnings for one task might nullify learnings for another. Behavioural plasticity enables humans and animals alike to respond to environmental changes without degrading learned knowledge; this can be achieved by regulating behaviour with neuromodulation—a biological process found in the brain. We demonstrate that by modulating activity-propagating signals, neurally trained agents evolving to solve tasks in dynamic environments that are prone to change can expect a significantly higher fitness than non-modulatory agents and also achieve their goals more often. Further, we show that while behavioural plasticity can help agents to achieve goals in these variable environments, this ability to overcome environmental changes with greater success comes at the cost of highly volatile evolution.

With the emerging interest in the ubiquitous sensing field, it has become possible to build assistive technologies for persons during their daily life activities to provide personalized feedback and services. For instance, it is possible to detect an individual’s behavioral pattern (e.g., physical activity, location, and mood) by using sensors embedded in smart-watches and smartphones. The multi-sensor environments also come with some challenges, such as how to fuse and combine different sources of data. In this article, we explore several methods of fusion for multi-representations of data from sensors. Furthermore, multiple representations of sensor data were generated and then fused using data-level, feature-level , and decision-level fusions . The presented methods were evaluated using three publicly available human activity recognition (HAR) datasets. The presented approaches utilize Deep Convolutional Neural Networks (CNNs). A generic architecture for fusion of different sensors is proposed. The proposed method shows promising performance, with the best results reaching an overall accuracy of 98.4% for the Context-Awareness via Wrist-Worn Motion Sensors (HANDY) dataset and 98.7% for the Wireless Sensor Data Mining (WISDM version 1.1) dataset. Both results outperform previous approaches.

Nowadays, virtual reality exposure therapy (VRET) is suggested as a way to treat social anxiety disorder (SAD), including the more specific fear of public speaking (FoPs). The major advantage of using VRET is that the virtual environment represents something in between in vitro and in vivo exposure, and therefore, lower the threshold for engaging in exposure tasks. In this paper, we present a VR system for addressing public speaking anxiety. Furthermore, wireless wristband was worn to detect the heartbeat of the participant in real-time and head movement data were collected from the VR headset. In initial experiments, we use the movement from the VR headset to predict heart rate with the goal of being able to get heart rate from VR equipment without the need for additional equipment allowing larger follow up studies.

Collaborative robots are becoming more common on factory floors as well as regular environments, however, their safety still is not a fully solved issue. Collision detection does not always perform as expected and collision avoidance is still an active research area. Collision avoidance works well for fixed robot-camera setups, however, if they are shifted around, Eye-to-Hand calibration becomes invalid making it difficult to accurately run many of the existing collision avoidance algorithms. We approach the problem by presenting a stand-alone system capable of detecting the robot and estimating its position, including individual joints, by using a simple 2D colour image as an input, where no Eye-to-Hand calibration is needed. As an extension of previous work, a two-stage transfer learning approach is used to re-train a multi-objective convolutional neural network (CNN) to allow it to be used with heterogeneous robot arms. Our method is capable of detecting the robot in real-time and new robot types can be added by having significantly smaller training datasets compared to the requirements of a fully trained network. We present data collection approach, the structure of the multi-objective CNN, the two-stage transfer learning training and test results by using real robots from Universal Robots, Kuka, and Franka Emika. Eventually, we analyse possible application areas of our method together with the possible improvements.

If robots are to become ubiquitous, they will need to be able to adapt to complex and dynamic environments. Robots that can adapt their bodies while deployed might be flexible and robust enough to meet this challenge. Previous work on dynamic robot morphology has focused on simulation, combining simple modules, or switching between locomotion modes. Here, we present an alternative approach: a self-reconfigurable morphology that allows a single four-legged robot to actively adapt the length of its legs to different environments. We report the design of our robot, as well as the results of a study that verifies the performance impact of self-reconfiguration. This study compares three different control and morphology pairs under different levels of servo supply voltage in the lab. We also performed preliminary tests in different uncontrolled outdoor environments to see if changes to the external environment supports our findings in the lab. Our results show better performance with an adaptable body, lending evidence to the value of self-reconfiguration for quadruped robots.

We present results and comparative analysis on the prediction of sensor events in a smart home environment with a limited number of binary sensors. We apply two probabilistic methods, namely Sequence Prediction via Enhanced Episode Discovery - SPEED, and Active LeZi - ALZ, as well as Recurrent Neural Network (RNN) with Long Short-Term Memory (LSTM) in order to predict the next sensor event in a sequence. Our dataset has been collected from a real home with one resident over a period of 30 weeks. The binary sensor events are converted to two different text sequences as dictated by SPEED and ALZ, which are also used as inputs for the LSTM networks. We compare the performance of the algorithms regarding the number of preceding sensor events required to predict the next one, the required amount of data for the model to reach peak accuracy and stability, and the execution time. In addition, we analyze these for two different sets of sensors. Our best implementation achieved a peak accuracy of 83% for a set with fifteen sensors including motion, magnetic and power sensors, and 87% for seven motion sensors.

Finding the ideal number of layers and size for each layer is a key challenge in deep neural network design. Two approaches for such networks exist: filter learning and architecture learning. While the first one starts with a given architecture and optimizes model weights, the second one aims to find the best architecture. Recently, several visual analytics (VA) techniques have been proposed to understand the behavior of a network, but few VA techniques support designers in architectural decisions. We propose a hybrid methodology based on VA to improve the architecture of a pre-trained network by reducing/increasing the size and number of layers. We introduce Activation Occurrence Maps that show how likely each image position of a convolutional kernel’s output activates for a given class, and Class Selectivity Maps, that show the selectiveness of different positions in a kernel’s output for a given label. Both maps help in the decision to drop kernels that do not significantly add to the network’s performance, increase the size of a layer having too few kernels, and add extra layers to the model. The user interacts from the first to the last layer, and the network is retrained after each layer modification. We validate our approach with experiments in models trained with two widely-known image classification datasets and show how our method helps to make design decisions to improve or to simplify the architectures of such models.

With the emerging ubiquitous sensing field, it has become possible to build assistive technologies for persons during their daily life activities to provide personalized feedback and services. For instance, it is possible to detect an individual's behavioral information (e.g. physical activity, location, and mood) by using sensors embedded in smartwatches and smartphones. To detect human's daily life activities, accelerometers have been widely used in wearable devices. In the current research, usually a single data representation is used, i.e., either image or feature vector representations. In this paper, a novel method is proposed to address two key aspects for the future development of robust deep learning methods for Human Activity Recognition (HAR): (1) multiple representations of a single sensor's data and (2) fusion of these multiple representations. The presented method utilizes Deep Convolutional Neural Networks (CNNs) and was evaluated using a publicly available HAR dataset. The proposed method showed promising performance, with the best result reaching an overall accuracy of 0.97, which outperforms current conventional approaches.

Recognizing human activities from sensor data is one of the key areas of image processing, computer vision, and pattern recognition researches today. The target of human activity recognition (HAR) is usually to detect and analyze distinguished activities from the data acquired via different sensors (e.g. thermal cameras). This work proposes a HAR approach from videos recorded via a thermal camera. Skeletons of human bodies are extracted from thermal frames using an opensource deep convolutional neural network (CNN)-based approach named OpenPose. It is generally applied on videos of typical color cameras. However, this work adopts OpenPose on thermal images to extract useful features so that the HAR system can be deployed in environments with low lights as well. Once skeletons of human silhouettes are obtained from the thermal images, robust spatiotemporal features are extracted followed by discriminant analysis. Finally, the discriminant features are fed into a deep recurrent neural network (RNN) for activity training and recognition. The proposed HAR method can be applied to monitor the users such as elderly in both bright and dark environments to prolong their independent life, unlike other typical color cameras which are generally applied in bright environments.

We investigated how a robot moving with different velocity profiles affects a person's perception of it when working together on a task. The two profiles are the common linear profile and a profile based on the animation principles of slow in, slow out. The investigation was accomplished by running an experiment in a home context where people and the robot cooperated on a clean-up task. We used the Godspeed series of questionnaires to gather people's perception of the robot. Average scores for each series appear not to be different enough to reject the null hypotheses, but looking at the component items provides paths to future areas of research. We also discuss the scenario for the experiment and how it may be used for future research into using animation techniques for moving robots and improving the legibility of a robot's locomotion.

There are many different ways a robot can move in Human-Robot Interaction. One way is to use techniques from film animation to instruct the robot to move. This article is a systematic literature review of human-robot trials, pilots, and evaluations that have applied techniques from animation to move a robot. Through 27 articles, we find that animation techniques improves an individual’s interaction with robots, improving the individual’s perception of qualities of a robot, understanding what a robot intends to do, and showing the robot’s state or possible emotion. Animation techniques also help people relate to robots that do not resemble a human or robot. The studies in the articles show further areas for research, such as applying animation principles in other types of robots and situations, combining animation techniques with other modalities, and testing robots moving with animation techniques over the long term.

Wearable sensors measuring different parts of people's activity are a common technology nowadays. Data created using these devices holds a lot of potential besides measuring the quantity of daily steps or calories burned, since continuous recordings of heart rate and activity levels usually are collected. Furthermore, there is an increasing awareness in the field of psychiatry on how these activity data relates to various mental health issues such as changes in mood, personality, inability to cope with daily problems or stress and withdrawal from friends and activities. In this paper we present the analysis of a unique dataset containing sensor data collected from patients suffering from depression. The dataset contains motor activity recordings of 23 unipolar and bipolar depressed patients and 32 healthy controls. We apply machine learning to classify patients into depressed and nondepressed. For evaluation of the algorithms, leave one patient out validation is performed. The best results achieved are an F1 score of 0.73 and a MCC of 0.44. The overall findings show that sensor data contains information that can be used to determine the depression status of a person.

For robots to handle the numerous factors that can affect them in the real world, they must adapt to changes and unexpected events. Evolutionary robotics tries to solve some of these issues by automatically optimizing a robot for a specific environment. Most of the research in this field, however, uses simplified representations of the robotic system in software simulations. The large gap between performance in simulation and the real world makes it challenging to transfer the resulting robots to the real world. In this paper, we apply real world multi-objective evolutionary optimization to optimize both control and morphology of a four-legged mammal-inspired robot. We change the supply voltage of the system, reducing the available torque and speed of all joints, and study how this affects both the fitness, as well as the morphology and control of the solutions. In addition to demonstrating that this real-world evolutionary scheme for morphology and control is indeed feasible with relatively few evaluations, we show that evolution under the different hardware limitations results in comparable performance for low and moderate speeds, and that the search achieves this by adapting both the control and the morphology of the robot.

Recent advances in image processing and computer vision have driven to numerous initiatives to recognize human activities from video data. This work proposes a human activity recognition approach using robust translation and scale-invariant body silhouette features recurrent neural network. First, Human body silhouette is extracted from a depth image after background subtraction. Then, body parts are segmented using random forests to get corresponding body skeleton in the image. Furthermore, scale-invariant skeleton features are extracted by representing the body joints in the spherical coordinate system. Then, the skeleton features are augmented with the motion features of the skeleton in consecutive frames. To combine with the skeleton features, Radon transformation is applied on the depth silhouettes to extract translation and scale-invariant silhouette features. The robust features extracted from the depth image sequences are then applied to a deep recurrent neural network for activity training and recognition. The proposed approach shows the superiority over other approaches by achieving greater than 98 % mean recognition rate on private and public datasets where others can yield around 95%.

Wearable sensors measuring different parts of people's activity are a common technology nowadays. In research, data collected using these devices also draws attention. Nevertheless, datasets containing sensor data in the field of medicine are rare. Often, data is non-public and only results are published. This makes it hard for other researchers to reproduce and compare results or even collaborate. In this paper we present a unique dataset containing sensor data collected from patients suffering from depression. The dataset contains motor activity recordings of 23 unipolar and bipolar depressed patients and 32 healthy controls. For each patient we provide sensor data over several days of continuous measuring and also some demographic data. The severity of the patients' depressive state was labeled using ratings done by medical experts on the Montgomery-Asberg Depression Rating Scale (MADRS). In this respect, the here presented dataset can be useful to explore and understand the association between depression and motor activity better. By making this dataset available, we invite and enable interested researchers the possibility to tackle this challenging and important societal problem.

Personal and ubiquitous sensing technologies such as smartphones have allowed the continuous collection of data in an unobtrusive manner. Machine learning methods have been applied to continuous sensor data to predict user contextual information such as location, mood, physical activity, etc. Recently, there has been a growing interest in leveraging ubiquitous sensing technologies for mental health care applications, thus, allowing the automatic continuous monitoring of different mental conditions such as depression, anxiety, stress, and so on. This paper surveys recent research works in mental health monitoring systems (MHMS) using sensor data and machine learning. We focused on research works about mental disorders/conditions such as: depression, anxiety, bipolar disorder, stress, etc. We propose a classification taxonomy to guide the review of related works and present the overall phases of MHMS. Moreover, research challenges in the field and future opportunities are also discussed.

We study heterogeneity among nodes in self-organizing smart camera networks, which use strategies based on social and economic knowledge to target communication activity efficiently. We compare homogeneous configurations, when cameras use the same strategy, with heterogeneous configurations, when cameras use different strategies. Our first contribution is to establish that static heterogeneity leads to new outcomes that are more efficient than those possible with homogeneity. Next, two forms of dynamic heterogeneity are investigated: nonadaptive mixed strategies and adaptive strategies, which learn online. Our second contribution is to show that mixed strategies offer Pareto efficiency consistently comparable with the most efficient static heterogeneous configurations. Since the particular configuration required for high Pareto efficiency in a scenario will not be known in advance, our third contribution is to show how decentralized online learning can lead to more efficient outcomes than the homogeneous case. In some cases, outcomes from online learning were more efficient than all other evaluated configuration types. Our fourth contribution is to show that online learning typically leads to outcomes more evenly spread over the objective space. Our results provide insight into the relationship between static, dynamic, and adaptive heterogeneity, suggesting that all have a key role in achieving efficient self-organization.

This paper presents a novel methodology for generating and compressing configuration bitstreams for modules that can be executed at different positions of an FPGA. The presented methodology for bitstream generation and compression does not need deep knowledge of the bitstream format and it is independent of the target (Xilinx) FPGA family. The approach consists of a design phase where partial bitstreams are decomposed into sequences of module dependent and module independent pieces of configuration data. At run-time, this data can then be recomposed for the individual placement positions by a special DMA configuration controller as one atomic operation without any further software interaction. Our experiments demonstrate that module relocation and fast partial reconfiguration can be implemented at low logic cost.

To fully exploit the capabilities of runtime reconfigurable FPGAs in self-aware systems, design tools are required that exceed the capabilities of present vendor design tools. Such tools must allow the implementation of scalable reconfigurable systems with various different partial modules that might be loaded to different positions of the device at runtime. This comprises several complex tasks, including floorplanning, communication architecture synthesis, physical constraints generation, physical implementation, and timing verification all the way down to the final bitstream generation. In this article, we present how our GoAhead framework helps in implementing self-aware systems on FPGAs with a minimum of user interaction.

Market mechanisms are a means by which resources in contention can be allocated between contending parties, both in human economies and those populated by software agents. Designing such mechanisms has traditionally been carried out by hand, and more recently by automation. Assessing these mechanisms typically involves them being evaluated with respect to multiple conflicting objectives, which can often be nonlinear, noisy, and expensive to compute. For typical performance objectives, it is known that designed mechanisms often fall short on being optimal across all objectives simultaneously. However, in all previous automated approaches, either only a single objective is considered, or else the multiple performance objectives are combined into a single objective. In this paper we do not aggregate objectives, instead considering a direct, novel application of multi-objective evolutionary algorithms (MOEAs) to the problem of automated mechanism design. This allows the automatic discovery of trade-offs that such objectives impose on mechanisms. We pose the problem of mechanism design, specifically for the class of linear redistribution mechanisms, as a naturally existing multi-objective optimisation problem. We apply a modified version of NSGA-II in order to design mechanisms within this class, given economically relevant objectives such as welfare and fairness. This application of NSGA-II exposes tradeoffs between objectives, revealing relationships between them that were otherwise unknown for this mechanism class. The understanding of the trade-off gained from the application of MOEAs can thus help practitioners with an insightful application of discovered mechanisms in their respective real/artificial markets.

In this paper, we introduce a novel gesture recognition algorithm named the ant learning algorithm (ALA), which aims at eliminating some of the limitations with the current leading algorithms, especially Hidden Markov Models. It requires minimal training instances and greatly reduces the computational overhead required by both training and classification. ALA takes advantage of the pheromone mechanism from ant colony optimization. It uses pheromone tables to represent gestures, which scales well with gesture complexity. Our experimental results show that ALA can achieve a high recognition accuracy of 91.3% with only one training instance, and exhibits good generalization.

GOAHEAD is a tool for easily building complex run-time reconfigurable systems. The tool provides sophisticated features like module relocation, hierarchical reconfiguration, or reusing modules among different systems. This demonstration shows 1) how reconfigurable systems can be built using GOAHEAD with only a few mouse clicks. In addition, 2) we will show how a partial module can be compiled all the way to the final bitstream running on an Atlys Spartan-6 FPGA board in one single batch job. The demonstrated system can simultaneously host up to 75 video overlay modules or 10 partially reconfigurable MIPS CPU systems. In the latter case, the CPUs feature reconfigurable custom instruction set extensions, hence demonstrating a hierarchically reconfigurable multi-core system.

The design and development of digital electronic systems is mainly performed by use of a hardware description language. To prepare students in electrical engineering for a career in hardware design many universities provide courses on VHDL. The traditional approach in teaching VHDL is mainly by means of textbook examples and simulation provided by software applications. These exercises are perceived as monotonous by the students and do not or only very slightly correspond with actual real-life applications based on FPGAs. Moreover, most real-life applications are too expensive to be equipped in student laboratories. To bridge the gap between a simulation-only environment and affordable real-life applications students should be provided access to remote real-life setups with a 24/7 availability and preferably shared between multiple institutes. The eDiViDe platform (European Digital Virtual Design Lab, http://www.edivide.eu), see Fig. 1, provides students with this unlimited and exciting access to FPGA based setups. Instead of theory-only courses and a quick basic lab, they can work their way through digital design courses testing their skills on real-life setups to trigger their interest. The platform hosts multiple FPGA setups at different European institutes. These setups are accessible through a web-based interface with video feedback. VHDL development is performed offline, given an entity and specific setup information. All further steps of the FPGA toolchain are performed on the platform. A reservation system takes care of the FPGA programming and student interaction with the setups. Similar initiatives provide stable solutions with educational support [1,2,3]. The eDiViDe platform differentiates with a distributed platform across several institutes and with the support for advanced setups. It is the result of a joint effort and easily expandable with additional setups at any location. At this moment following setups are available: greenhouse, stepper motor control, sea noise emulator, state machine workshop, Geffe generator, pong / game of life, traffic light control, MIPS CPU. This set will be extended with more advanced setups that include e.g. a partial reconfiguration workshop for audio/video filters, a side-channel analysis setup and a mars rover playfield. Besides promoting digital design education, the eDiViDe platform creates a channel to make the research activities in the contributing universities more visible. Industry could also benefit from this platform to promote their brand and products to soon to be engineers.

A heterogeneous system with soft CPU tailored to the individual threads of the application, while still software based, offers the potential for improved performance and resource utilization over a homogeneous system. In this paper we present a method to automatically create a heterogeneous multi-core system from a multithreaded software application. The resulting system consists of processing elements based on customized MIPS soft CPUs coupled with their respective programs. Using instruction set architecture (ISA) subsetting, we adapt the individual soft CPUs to the specific computations they have to perform. We have carried out a case study with a constraint solver application for which we find a performance increase of 1.54 accompanied with an area reduction of 22.5% compared to a homogeneous multi-core system. We also present an automated toolchain that generates synthesizable IP-cores from software threads with little additional development overhead.

In this paper, we present an open source partial reconfiguration (PR) system which is designed for portability and usability serving as a reference for engineers and students interested in using the advanced reconfiguration capabilities available in Xilinx FPGAs. This includes design aspects such as floorplanning and interfacing PR modules as well as fast reconfiguration and online management. The system features relocatable modules which can even contain reconfigurable modules themselves, hence, implementing hierarchical PR.

This paper proposes an approach to representing robot morphology and control, using a two-level description linked to two different physical axes of development. The bioinspired encoding produces robots with animal-like bilateral limbed morphology with co-evolved control parameters using a central pattern generator-based modular artificial neural network. Experiments are performed on optimizing a simple simulated locomotion problem, using multi-objective evolution with two secondary objectives. The results show that the representation is capable of producing a variety of viable designs even with a relatively restricted set of parameters and a very simple control system. Furthermore, the utility of a cumulative encoding over a non-cumulative approach is demonstrated. We also show that the representation is viable for real-life reproduction by automatically generating CAD files, 3D printing the limbs, and attaching off-the-shelf servomotors.

Links between music and body motion can be studied through experiments called sound-tracing . One of the main challenges in such research is to develop robust analysis techniques that are able to deal with the multidimensional data that musical sound and body motion present. The article evaluates four different analysis methods applied to an experiment in which participants moved their hands following perceptual features of short sound objects. Motion capture data has been analyzed and correlated with a set of quantitative sound features using four different methods: (a) a pattern recognition classifier, (b) t -tests, (c) Spearman's ρ correlation, and (d) canonical correlation. This article shows how the analysis methods complement each other, and that applying several analysis techniques to the same data set can broaden the knowledge gained from the experiment.

Evolvable hardware (EHW) has shown itself to be a promising approach for prosthetic hand controllers. Besides competitive classification performance, EHW classifiers offer self-adaptation, fast training, and a compact implementation. However, EHW classifiers have not yet been sufficiently compared to state-of-the-art conventional classifiers. In this paper, we compare two EHW approaches to four conventional classification techniques:k-nearest-neighbor, decision trees, artificial neural networks, and support vector machines. We provide all classifiers with features extracted from electromyographic signals taken from forearm muscle contractions, and let the algorithms recognize eight to eleven different kinds of hand movements. We investigate classification accuracy on a fixed data set and stability of classification error rates when new data is introduced. For this purpose, we have recorded a short-term data set from three individuals over three consecutive days and a long-term data set from a single individual over three weeks. Experimental results demonstrate that EHW approaches are indeed able to compete with state-of-the-art classifiers in terms of classification performance.

Reconfigurable hardware allows application specific customization of soft microprocessors. Techniques such as removing unused instructions, software emulation of instructions, custom instruction set extensions, and run-time reconfigurable instructions have been suggested. However, the techniques have largely been studied separately from each other. The contribution of this paper is a classification method enabling integration of these techniques. This allows for generating an application specific microprocessor based system from a given program. The generated microprocessor is optimized with respect to performance per area. The improvement of our methodology is demonstrated for the CoreBench benchmark. The benefit of combining the removal of unused instructions (ISA subsetting) with software emulation of rarely used instructions is shown to increase performance while at the same time reducing resource requirements. Improvement in both area and performance is accomplished thorough simplifying the design allowing an increase in clock frequency for the synthesized soft CPU. Optimizing only by using custom instructions allowed a 12% increase in performance, but also increased resource usage by 6%. Software emulation combined with ISA subsetting allowed area savings of 7%, but only improved performance by 3%. By combining custom instructions, software emulation and ISA subsetting, we achieved an performance improvement of 15% while at the same time reducing resource requirements.

Exploiting the benefits of partial run-time reconfiguration requires efficient tools. In this paper, we introduce the tool Go Ahead that is able to implement run-time reconfigurable systems for all recent Xilinx FPGAs. This includes in particular support for low cost and low power Spartan-6 FPGAs. Go Ahead assists during floor planning and automates the constraint generation. It interacts with the Xilinx vendor tools and triggers the physical implementation phases all the way down to the final configuration bit streams. Go Ahead enables the building of flexible systems for integrating many reconfigurable modules very efficiently into a system. The tool targets (re)usability, portability to future devices, and migration paths among reconfigurable systems featuring different FPGAs or even FPGA families. Moreover, it provides a scripting interface and all features can be accessed remotely.

This paper analyses different hardware sorting architectures in order to implement a highly scaleable sorter for solving huge problems at high performance up to the GB range in linear time complexity. It will be proven that a combination of a FIFO-based merge sorter and a tree-based merge sorter results in the best performance at low cost. Moreover, we will demonstrate how partial run-time reconfiguration can be used for saving almost half the FPGA resources or alternatively for improving the speed. Experiments show a sustainable sorting throughput of 2GB/s for problems fitting into the on-chip FPGA memory and 1 GB/s when using external memory. These values surpass the best published results on large problem sorting implementations on FPGAs, GPUs, and the Cell processor.

While it is feasible with today's commercial tools to swap from one module to another, many applications demand more advanced configuration schemes, for example, to map different dataflow graphs onto an FPGA. To support this, we will improve an existing on-FPGA communication architecture in order to carry out arbitrary routing among multiple freely placed modules. This results in a circuit switching network that will be efficiently implemented directly within the FPGA routing fabric.

A lot of optimal and heuristic algorithms for solving facility layout problem (FLP) have been developed in the past few decades. The majority of these approaches adopt a problem formulation known as the quadratic assignment problem (QAP) that is particularly suitable for equal area facilities. Unequal area FLP comprises a class of extremely difficult and widely applicable optimization problems arising in many diverse areas to meet the requirements for real-world applications. Unfortunately, most of these approaches are based on a single objective. While, the real-world FLPs are multi-objective by nature. Only very recently have meta-heuristics been designed and used in multi-objective FLP. They most often use the weighted sum method to combine the different objectives and thus, inherit the well-known problems of this method. As of now, there is no formal approach published for the unequal area multi-objective FLP to consider several objectives simultaneously. This paper presents an evolutionary approach for solving multi-objective unequal area FLP using multi-objective genetic algorithm that presents the layout as a set of Pareto-optimal solutions optimizing multiple objectives simultaneously. The experimental results show that the proposed approach performs well in dealing with multi-objective unequal area FLPs which better reflects the real-world scenario.

A Coevolutionary, Hyper Heuristic approach to the optimization of Three-dimensional Process Plant Layouts (3DPPLs) is explored. By taking advantage of the natural problem decomposition, one population of layout heuristics, and another population of scheduling heuristics are coevolved. Generalized heuristics are evolved by training on multiple small problem instances, so that training time is reduced. The best generalized heuristic builds arbitrary sized 3DPPLs which reduce the cost by 18% when compared to a handmade heuristic. Specialized heuristics are evolved by optimizing each problem instance and outperforms the generalized heuristics after a fixed number of generations. Compared to a direct-encoded Genetic Algorithm, the benefit of specialized heuristics increases with the size of the problem, and costs are reduced by 30% when compared to the handmade heuristic.

The architecture of Xilinx FPGAs, has changed remarkable with respect to their ability to implement runtime reconfigurable systems throughout the last generations. This paper will discuss these changes and reveal an on-FPGA communication architecture that is especially tailored to Xilinx Virtex-5 FPGAs. With this architecture, modules can be integrated in a two-dimensional grid with more than a hundred of individual tiles while allowing a throughput of several GB/s to reconfigurable modules.

In this paper, we show how short-circuits on FPGAs can be caused by partial runtime reconfiguration. Short-circuit can even occur on FPGAs that do not offer any tristate resources just by using off the shelf vendor tools without any bitstream manipulation. The duration of the here presented short-circuits ranges from short spikes up to persistent short-circuits that remain active during runtime. Short-circuits will result in increased current consumption and can thus harm the system and must therefore be prevented. An algorithm is derived that detects whether configuration data will cause short-circuits. We implemented this algorithm in a bitstream scanner that can also be used in systems at runtime.

By neatly reserving routing resources of an FPGA at design-time, a circuit switching network can be implemented for integrating reconfigurable modules in a two-dimensional manner at run-time. In this network, paths can be set directly by manipulating fractions of the switch matrix configuration. By utilizing disjoint resources for implementing the network and the modules of the system, the network is capable to route paths to partial modules independent of the present module placement layout. This paper proposes concepts, implementation issues, and a design flow for building reconfigurable systems providing such a network. Furthermore, a timing model will be presented for validating the system at run-time. The applicability of the network will be demonstrated in a prototype system by routing I/O pins to partial modules.

In this paper, we demonstrate systems based on Spartan-6 series FPGAs that provide full support for active partial run-time reconfiguration. We will summarize design factors for successfully applying run-time reconfiguration, reveal details on partial reconfiguration on Spartan-6 FPGAs, and introduce our easy to use design flow. In this flow, a module can multiple times be instantiated or even migrated to different systems without the need to physically reimplement such a module. The demo systems can host manifold different partial modules that each are capable to manipulate a video stream.

In component-based system design, systems are composed by integrating fully physically implemented hard-IP cores. This design style has the potential to close the design productivity gap that will arise when FPGAs will head over the one million look-up table boundary by minimizing, or even fully removing, costly verification and place&route steps during the system integration phase. This integration can even be carried out at run-time, and consequently, making component-based design the base for implementing partially reconfigurable systems. In this paper, we will discuss requirements on the routing of the components such that modules can be composed to systems without interfering among each other while still being able of providing the top-level component to component communication. We will propose to relax the strict bounding box constraints that have been traditionally applied to implement reconfigurable components. Our results will demonstrate an area and reconfiguration time improvement of up to 33% as compared to the traditional strict bounding box method in a case study using Xilinx Spartan-6 FPGAs.

Type-2 fuzzy logic systems are proposed as an alternative solution in the literature when a system has a large amount of uncertainties and type-1 fuzzy systems come to the limits of their performances. In this study, an adaptive type-2 fuzzy-neuro system is designed for the position control of a servo system with an intelligent sensor. The sensor gives different resistance values with respect to the stretch of it, and it is supposed to be used in an robotic arm position measurement system. These kinds of sensors can be used in human-assistance robots that have soft surfaces in order not to damage the humans. However, these sensors have time-varying gains and uncertainties that are not very easy to handle. Moreover, they generally have a hysteresis on their input-output relations. The simulation results show that the control algorithm developed gives better performances when compared to conventional type-1 fuzzy controllers on such a highly nonlinear, uncertain system.

Efficient heuristics are required for on-line optimization problems where search-based methods are unfeasible due to frequent dynamics in the environment. This is especially apparent when operating on combinatorial NP-complete problems involving a large number of items. However, designing new heuristics for these problems may be a difficult and time-consuming task even for domain experts. Therefore, automating this design process may benefit the industry when facing new and difficult optimization problems. The Distributor's Pallet Packing Problem (DPPP) is the problem of loading a pallet of non-homogenous items coming off a production line and is an instance of a range of resource-constrained, NP-complete, scheduling problems that are highly relevant for practical tasks in the industry. Common heuristics for the DPPP typically decompose the problem into two sub-problems; one of pre-scheduling all items on the production line and one of packing the items on the pallet. In this paper we concentrate on a two dimensional version of the DPPP and the more realistic scenario of having knowledge about only a limited set of the items on the production line. This paper aims at demonstrating that such an unknown heuristic may be evolved by Gene Expression Programming and Cooperative Coevolution. By taking advantage of the natural problem decomposition, two species evolve heuristics for pre-scheduling and packing respectively. We also argue that the evolved heuristics form part of a developmental stage in the construction of the finished phenotype, that is, the loaded pallet.

Several practical problems in industry are difficult to optimize, both in terms of scalability and representation. Heuristics designed by domain experts are frequently applied to such problems. However, designing optimized heuristics can be a non-trivial task. One such difficult problem is the Facility Layout Problem (FLP) which is concerned with the allocation of activities to space. This paper is concerned with the block layout problem, where the activities require a fixed size and shape (modules). This problem is commonly divided into two sub problems; one of creating an initial feasible layout and one of improving the layout by interchanging the location of activities. We investigate how to extract novel heuristics for the FLP by applying an approach called Cooperative Coevolutionary Gene Expression Programming (CCGEP). By taking advantage of the natural problem decomposition, one species evolves heuristics for pre-scheduling, and another for allocating the activities onto the plant. An experimental, comparative approach investigates various features of the CCGEP approach. The results show that the evolved heuristics converge to suboptimal solutions as the problem size grows. However, coevolution has a positive effect on optimization of single problem instances. Expensive fitness evaluations may be limited by evolving generalized heuristics applicable to unseen fitness cases of arbitrary sizes.

A pattern recognition system that can process a large amount of image data at high speed is required in many fields. In this paper, we propose an on-chip pattern recognition system that utilizes the reconfigurability of the FPGA. The features of the system are not only very high recognition speed but also an adaptive function. For example, when objects to be detected change appearance, recognition parameters must be changed to retain the recognition accuracy. The system can automatically adjust by executing on-chip partial reconfiguration. The system runs at 25MHz and can return a recognition result in one clock cycle, 40ns. To update the system, all processes needed for searching for the best recognition parameters, generating configuration data and reconfiguring the system are carried out within 30s.

One type of network security strategy is using an intrusion detection system (IDS). We are implementing an IDS in FPGA-based (Field Programmable Gate Array) reconfigurable hardware. This is to achieve higher speed and more efficient performance of network security, as networks develop very fast with consequently more demanding constraints. This study provides novel hardware architectures for an IDS system which should be able to monitor networks with a speed up to 2.68 Gbps.

One of the problems concerning the backpropagation training of feed-forward neural networks is the effect of the weight update frequency. This aspect influences the efficiency of parallel implementations of the training algorithm where the training vectors are distributed among processors. In this paper the convergence of two applications for various weight update intervals is reported. Further, several models are proposed for describing convergence and learning rate aspects in the context of a set of weight update intervals. The results show that the convergence by updating the weights after each training vector leads to about 10 times less number of training iterations compared to updating the weights only ones for the whole training set.