Evolutionary Optimization of Sliding Contact Positions in Powered Floor Systems for Mobile Robots




Eric Medvet, Stefano Seriani, Alberto Bartoli, Paolo Gallina


at - Automatisierungstechnik
(rank Q3 in Computer Science Applications)



Links and material:

Abstract #

Mobile robotics is a rapidly expanding technology due to its potential for increased safety and lower costs. In many applications, power is supplied to the robot through sliding contacts and a powered floor. Deciding the positions of the contacts on the robot is a difficult task: for any position/orientation of the robot, at least one contact has to touch a positive strip and at least one a negative strip. In this work, we tackle the problem using Differential Evolution (DE). We formally define problem-specific constraints and objectives and then describe how to use DE for evolving contact positions that satisfy those constraints and maximize those objectives. We validate experimentally our proposal by applying it to three real robots and by studying the impact of the main problem parameters on the effectiveness of the evolved designs for the sliding contacts.