The abstracts of the invited talks can be downloaded.
Deutschen Zentrums für Luft- und Raumfahrt
Design and control of robots with variable stiffness actuation
Adjustable compliance is a key feature for robots designed to interact skilfully and safely with unknown environments and with humans. While controlled compliance based on joint torque sensing or on measurement of the tip wrench is meanwhile well established, newest developments aim at integrating adjustable mechanical compliance into the robot joints. Mechanical compliance has the major advantage that it increases the robustness of the system for unexpected impacts and at the same time can provide higher peak torque and velocity due to intermediate energy storage into the elastic element.
However, the mechatronic design of such systems poses major challenges due to the increased system complexity through the additional stiffness adjusting device. Moreover, the control of such systems is challenging due to the weakly damped, very non-linear springs and the resulting highly nonlinear robot behaviour.
The talk will present design and control ideas of variable compliant robots using as example the DLR hand-arm system. Its control is providing an exciting link to neuroscience by helping to understand human motion principles and to validate biological sensory-motor control hypotheses.
However, the mechatronic design of such systems poses major challenges due to the increased system complexity through the additional stiffness adjusting device. Moreover, the control of such systems is challenging due to the weakly damped, very non-linear springs and the resulting highly nonlinear robot behaviour.
The talk will present design and control ideas of variable compliant robots using as example the DLR hand-arm system. Its control is providing an exciting link to neuroscience by helping to understand human motion principles and to validate biological sensory-motor control hypotheses.
Carnegie Mellon University
A user-centered approach to workspace limitations in tele-operation
A common problem in tele-operation with force feedback is that the operator workspace is small, relative to the target workspace in which sensing and manipulation take place. Various methods have been proposed to deal with this problem, but with limited success, particularly when the goal is to support remote manipulation by transmitting forces. I will present work with collaborators Hanns Tappeiner and Ralph Hollis describing a novel solution to workplace limitations. The method is motivated by cognitive science research on people’s ability to update spatial location, based on sensory-motor cues. I will describe the approach and its grounding in research on human spatial perception and action. Finally, I will present evaluation data showing the benefits of force feedback overall, and our method in particular.
LAAS-CNRS
Understand the computational foundations of anthropomorphic locomotion
Walking towards a goal requires to make choices. Why do we choose one trajectory and not another one? Is it because it is the shortest one? the fastest one? the most comfortable? in what sense? What are the laws underlying the formation of locomotor trajectories? The talk will overview a pluridisciplinary approach to the questions. The approach combines computational neuroscience methodology, numerical optimization and robotics research. We will first establish that human locomotion is stereotyped. Then we will introduce the so-called inverse optimal control problem to provide a generic control model of human locomotion. The model is effective: it has been implemented on the humanoid robot platform HRP2. The presentation gathers work done in collaboration with Prof. A. Berthoz (Collège de France, Paris, France) and Prof. K. Mombaur (Heidelberg University, Germany).
Vrije Universiteit Brussel
Variable stiffness actuators in lower limb rehabilitation robots, prostheses and orthotic devices
Torque-angle characteristics and power-angle characteristics of human walking e.g. running show clearly that actuators for the joints of lower limb rehabilitation devices, prostheses or orthotic aids should be able to produce high forces/torques as well as high power peaks. For wearability reasons these actuators should be lightweight. Energy storage during periods of negative work will increase the efficiency. During walking/running impacts occur at heel strike. Finally, a prosthetic limb or rehabilitation device must always interact with the human in a safe manner.
The talk will present, using examples of rehabilitation robots and prosthetic devices developed by the R&MM research group of the Vrije Universiteit Brussel, how variable stiffness actuators can be used to approximate human performance as close as possible.
The talk will present, using examples of rehabilitation robots and prosthetic devices developed by the R&MM research group of the Vrije Universiteit Brussel, how variable stiffness actuators can be used to approximate human performance as close as possible.
University of Twente
Robotics for minimally invasive surgery
This talk provides two examples that emphasize the importance of pre-operative plans and intra-operative control to improve the accuracy of minimally invasive surgical robotic procedures. Soft tissue displacements during minimally invasive surgical procedures may cause target motion and subsequent misplacement of the surgical tool. First, a technique is presented to predict target displacements using a combination of ultrasound elastography and finite element modeling. This talk provides experimental evidence that organ geometry and boundary conditions surrounding the organ are important factors influencing target motion. This technique could be used for pre-operative planning of minimally invasive surgical interventions.
Manually steering the tip of a flexible endoscope to navigate through an endoluminal path relies on the physician’s dexterity and experience. Second, a robotic flexible endoscope steering system was developed that uses the images to control the endoscope tip orientation towards the direction of the lumen. Using this setup, the endoscope was steered using a haptic device, while haptic feedback was given based on the output from the image-processing algorithm. This intra-operative control method could be used to robotically steer advanced endoscopes.
Manually steering the tip of a flexible endoscope to navigate through an endoluminal path relies on the physician’s dexterity and experience. Second, a robotic flexible endoscope steering system was developed that uses the images to control the endoscope tip orientation towards the direction of the lumen. Using this setup, the endoscope was steered using a haptic device, while haptic feedback was given based on the output from the image-processing algorithm. This intra-operative control method could be used to robotically steer advanced endoscopes.
University of Napoli
Robots moving closer to humans
Robots! Robots on Mars and in oceans, in hospitals and homes, in factories and schools, robots fighting fires, making goods and products, saving time and lives. Robots today are making a considerable impact on many aspects of modern life, from manufacturing to healthcare. Reaching for the human frontier, robotics is also vigorously engaged in the growing challenges of new emerging domains. Interacting, exploring, and working with humans, the new generation of robots will increasingly touch people’s lives. Unlike the industrial robotics domain where the workspace of machines and humans can be segmented, applications of intelligent machines that work in contact with humans are increasing, which involve e.g. haptic interfaces and teleoperators, cooperative material-handling, power extenders and such high-volume markets as rehabilitation, physical training, entertainment. In this context, it is customary to distinguish between Cognitive Human-Robot Interaction (cHRI) and Physical Human-Robot Interaction (pHRI). This talk is aimed at discussing a number of issues in pHRI concerning with safety, dependability and dexterity. The presentation will be accompanied by videos illustrating experimental tests on both conventional and new lightweight robots endowed with force and vision sensors.
University of Twente
Modeling and control of interaction using port-based robotics
Mechanical interaction between a robot and a human can be very effectively and elegantly described using the concept of power-ports. In this presentation a short introduction will be presented about modeling and control using ports. Issues like stability and passivity, representations of interaction will be discussed and illustrated with some examples.
École Nationale Supérieure de Techniques Avancées
Towards human-robot customized interaction
This presentation focuses on Socially Assistive Robotics. The research aims to address a new niche of assistive robotics, one that involves an autonomous robot providing contact-free rehabilitation monitoring, assistance, and encouragement to vulnerable individuals (i.e., stroke patients, people suffering of Alzheimer’s disease, and children with autism), while also being capable of providing detailed reports of individuals progress to physicians, therapists, and family members. Specifically, this research develops and evaluates different methodologies of learning and online robot behavior adaptation to the user personality, preferences, disability, and profile aimed at both personalizing the therapy/interaction process and maximizing its health related outcomes. The different approaches are evaluated on assistive robotic systems with different physical embodiments, including human-like and non human-like physical appearance.
