Dr. Timothy Wiley

Title: Learning Autonomous Robot Behaviours

Dr. Timothy Wiley

Autonomous robots execute complex behaviours to operate and perform tasks in real-world environments. Machine learning has been employed to acquire such behaviours, usually by trial-and error learning, however, this process often requires a large number of iterations. If an online learning process must be performed, that is, learning on board the robot as it operates, simplistic approaches to trial-and-error learning are infeasible. The robot will break down long before any significant progress is made. Instead, multi-stage approaches improve the feasibility and efficiency of learning robot behaviours.

I will discuss aspects of learning robot behaviours in two fields of research. The first field is in social robotics, where robots are interacting with humans on a regular basis. This field covers both the autonomous behaviour of the robots, and their morphology, as both play a key role in developing acceptable and "natural" interactions. The secondly field is in applications on autonomous robots for urban search and rescue. In this domain, the use of robot is important to assist and protect the safety of first responders to disaster sites. However, due to the challenges of communication, search and rescue robots require autonomous behaviours.