Psychology Departmental Seminar Talk Dr. Ehsan Hashemi

Abstract: Advances in applications of cooperative control and sensing in networked autonomous systems, such as intelligent transport and networked robotics, facilitate robust control of robots/vehicles, motion planning, and reliable navigation. In this regard, computationally efficient architecture for cooperation among different control systems in the network (and individual robots), or even human-autonomy plays a key role to perform tasks or reach a certain trajectory reliably. However, such architectures for trajectory planning and control between a human and a fully (or partially) autonomous system are facing challenges due to not having access to the human objectives, and more specifically human cost functions. Human cognitive processes and behavior might be useful for identification of such objectives. In the first part of presentation, a game-theory-based cooperative control framework for automated driving systems, in which guidance, path tracking, and corner traction control strategies are formulated in terms of players in a differential game, will be presented. Then, opportunities for extending such coordination schemes for human-autonomy will be discussed, with applications in: i) human-in-the-loop compliance control for walker-assisted gait technologies; and ii) interactions between human and automated driving systems during take over in self-driving cars. Inference of human cognitive processes during such cooperative interactions helps developing more adaptive and personalized control methods resilient to disturbances.

Lab URL: https://sites.google.com/view/ehsan-hashemi-uwaterloo