We are working on understanding how human supervisors perceive and plan high-level functions for a swarm of semi-autonomous agents. We are using ElectroEncephaloGraphic signals as human subjects perceive and control collective behaviors of robotic swarms. The overarching goal is to unveil brain mechanisms that perceive multi-agent systems information, and define methods to extract centralized and/or de-centralized control commands for a multi-agent system.
Gait requires kinematic and dynamic coordination of the limbs and muscles, multi-sensory fusion and robust control mechanisms. The force stimulus generated by the interaction of the foot with the walking surface is a vital part of the human gait. We are using a novel system, called Variable Stiffness Treadmill (VST), in order to understand inter-leg coordination sensorimotor mechanisms, and utilize them for a new approach to rehabilitation of hemiparetic gait.
Proportional myoelectric control has been implemented for user-friendly interaction with prosthetics, orthotics, and new advances in human-machine interfaces. This work supports a shift in myoelectric control applications towards proportional controls learned through development of unique muscle synergies. Analyzing long term trends in human motor learning through interaction with a proportionally controlled myoelectric interface, this study reveals the natural emergence of a new muscle synergy space as a user identifies the novel system dynamics of the interface.
Physical human-robot cooperation currently depends primarily on interaction forces. We investigate human-robot cooperation through a novel perspective where neural signals (from the brain and the muscles) are incorporated in the robot controller to optimize human-robot interaction and cooperation. Leader and follower roles and multiple levels of trust between the human and the robot are extracted in real-time, which allows seamless interaction and cooperation between humans and machines-robots.
Unmanned aerial vehicles have received increased attention in the last decade due to their versatility, as well as the availability of inexpensive sensors for their navigation and control. This project introduces a new concept of control involving more than one quadrotors, according to which two quadrotors can be physically coupled in mid-flight. This concept equips the quadrotors with new capabilities, e.g. increased payload or pursuit and capturing of other quadrotors. The combination of ground and aerial vehicles for exploration of unknown environments is also researched.
As robots are increasingly used in human-cluttered environments, the requirement of human-likeliness in their movements becomes essential. This project focuses on the introduction of bio-inspired control schemes for robot arms that coordinate with human arms in bi-manual manipulation tasks. Using data captured from human subjects performing a variety of every-day life tasks employing their two arms, we propose a bio-inspired controller for a robot arm, that is able to learn human inter-arm coordination during those tasks.