Robotics research at the LSR ranges from autonomous navigation and 3D environment understanding to legged locomotion and fast dexterous manipulation.
Several hardware platforms such as the Autonomous City Explorer (ACE) robot and the Stäubli basketball robot serve as demonstrators and provide a link between theoretical research and practical implementation.
Central robotics research goals within the ACE project are, among others, robust simultaneous localization and mapping in outdoor environments, dynamic path planning and obstacle avoidance, object recognition in 3D laserrange data and people detection in vision and range data.
The main aims of the Stäubli Basketball robot project are the investigation of fast manipulation with non-negligible dynamics as well as visual and haptic sensor fusion.
More theoretical work is conducted in the area of biped walking, where legged locomotion is investigated from control-theoretic point of view. Demonstrators for this field of research are planned for the future.
Telepresence systems allow a human operator to explore and manipulate a remote environment. Thereby, he commands a robotic teleoperator (TO) over a communication channel (COM) through a human sytem interface (HSI). The goal is to extend humans cognitive capabilities to a remote environment that is impenetrable by distance, scaling, or living conditions. Examples are space teleoperation, micromanipulation or working in contamined areas.
Improving the operator's experience and enabling him to perform a given task as in reality are the abstract goals of telepresence reserach. Therefore, multimodal information has to be fed back consistently to the human operator. Research topics for telepresence cover a broad horizon from theoretical control issues over psychophysical analyses to haptic interaction with human beings.
Fundamental research is mainly carried out for hybrid systems, sliding mode control and networked control systems. In the field of hybrid systems, there are works on optimization, reinforcement learning, hierarchical path planning, model predictive control and verification. The research on networked control systems focuses on the development of novel control methodologies which take into account the non-ideal transmission of control signals though a communication network. Besides the research on the mentioned fields itself, it is tried to combine the research results to intelligent control systems for cognitive systems.
As robots are gradually leaving factory environments and moving into human populated environments, they need to achieve higher levels of interaction with humans.
Human-robot interaction (HRI) is a research field with a wide range of applications, future scenarios, and potentially a high economic impact. This interdisciplinary research area comprises classical robotics, cognitive sciences, and psychology. HRI research at LSR focuses on aspects of cooperative decision making, planning and joint action/manipulation including topics of verbal and haptic information exchange as well as emotional robots.
How does thinking work? How do we interpret what we see, hear, smell, and touch? – and how do we decide what we do and how we do it in the world around us? This is one of today's greatest mysteries in science.
Looking at small animals with tiny brains, we get the impression that they act effortlessly in the world, foraging for food and returning home safely. In contrast, today's carefully hand-designed computers and robots with all available sensors and processing power are hardly able to successfully perform such simple behaviors. The world is too complex and too ambiguous to get interpreted reliably with contemporary algorithms. So in which fundamental principles does information processing in brains differ from information processing performed by current computing algorithms?