"Human-Robot Interactive Control using Brain and Muscle Interfaces", Peter Allen (CS, Columbia) - Hosted by Kostas Bekris
Tuesday, November 27, 2018, 01:00pm - 02:30pm
Busch Campus, Psych 101
Peter K. Allen, Department of Computer Science, Columbia University
Abstract: As robotic systems begin to populate our daily life, Human-Robot interaction modes and control become increasingly important. Besides the traditional methods of interacting with a robot such as speech, gesture, or keyboard/joystick, Brain and Muscle Computer Interfaces (BMCI) can also be used. These interfaces provide an intriguing method for controlling remote agents. Human operators can easily become overloaded as task responsibilities increase. By using a BMCI, possibly in concert with other interfaces, a human operator can begin multi-task control of multiple agents, some controlled by more traditional interfaces, and others using a BMCI. This work also finds high applicability in the Assistive Robotics space, where it can assist individuals with limited speech, mobility and motion control. In this talk, we present three systems we have built that utilize these interfaces. The first is an assistive robotic grasping system that allows impaired individuals to interact with the system in a human-in-the-loop manner, including the use of a novel cranio-facial electromyography input device. The system uses an augmented reality interface that allows users to plan grasps online that match their task-oriented intents. The second is a shared control online grasp planner. It uses an EEG-based interface to recognize the user’s grasping preferences. The EEG signal classifier is fast and simple to train, and the system as a whole requires almost no learning on the part of the subject. The last system is a novel hierarchical system for shared control of a humanoid robot. Our framework uses a Brain Computer Interface (BCI) to interpret EEG signals via Steady- State Visual Evoked Potentials (SSVEP). Our system leverages the ability of the robot to accomplish low level tasks on its own, while the user assists the robot with high level directions when needed.
Biosketch: Peter Allen is Professor of Computer Science at Columbia University, and Director of the Columbia Robotics Lab. He is the recipient of the CBS Foundation Fellowship, Army Research Office fellowship, the Rubinoff Award for innovative uses of computers, and the NSF PYI award. His current research interests include robotic grasping, medical robotics and Brain-Computer Interfaces for Human-Robot interaction
- Grasping with Your Brain: A Brain-Computer Interface for Fast Grasp Selection by Robert Ying, Jonathan Weisz and Peter K. Allen, Robotics Research: Volume 1, eds. A. Bicchi and W. Burgard, pp. 325-340, 2018, Springer
- Task Level Hierarchical System for BCI-enabled Shared Autonomy by Iretiayo Akinola, Boyuan Chen, Jonathan Koss, Aalhad Patankar, Jake Varley and Peter Allen, IEEE/RAS International Conference on Humanoid Robotics, Nov. 2017