Abstract: Continuum robots are bio-inspired devices that seek to find a middle ground between traditional rigid robots and soft robots. With their mix of compliance, payload capability, inherent safety, dexterity, and manipulability, continuum robotic manipulators are perfectly suited to serve as collaborative robots alongside humans in potential areas ranging from manufacturing to healthcare applications. The high degrees of freedom of continuum manipulators, which results in kinematic redundancy, serve as their greatest challenge when it comes to path and motion planning. More specifically, because of their high compliance, performing reliable and stable path planning for such robots poses a major challenge.
In this talk, we will survey some recent progress toward performing path planning for continuum manipulators that addresses some of these issues, while providing performance guarantees. We also survey some work under progress that investigates the use of machine learning techniques for performing stable motion planning, and the use of anticipatory planning methods for obstacle avoidance in dynamic environments.
Bio: Iyad Kanj is a professor of computer science in the School of Computing at DePaul University. He completed his Ph.D. at Texas A&M University. His research interests include parameterized algorithms, graph algorithms, and computational geometry. His recent work focuses on applications of the aforementioned areas to problems in artificial intelligence and robot motion planning.