We propose to combine the learnability and stability of dynamic movement primitives (DMPs) with the analyzable sequencing of stable heteroclinic channels (SHCs) in soft robotics applications.

Looks to expand the scale and sophistication of synthetic nervous systems (SNS), a continuous time dynamical model of the insect nervous system, which will be implemented to control our six-legged MantisBot, endowing it with online learning and intelligent autonomy. Also will investigate its distributed intelligence.

Engineering of CPS System Design and Verification will be addressed—How can cyber physical walking systems (CPWS) be designed to be safe, secure, and resilient in a variety of unanticipated disturbances? How can this mixed autonomous system with a human in the loop be verified? CPS Real-time Control and Adaptation will also be targeted—How can real-time dynamic control and behavior adaptation be achieved in a diversity of environments?