Biogeography-Based Optimization for Robot Controller Tuning

Abstract

This research involves the development of an engineering test for a newly-developed evolutionary algorithm called biogeography-based optimization (BBO), and also involves the development of a distributed implementation of BBO. The BBO algorithm is based on mathematical models of biogeography, which describe the migration of species between habitats. BBO is the adaptation of the theory of biogeography for the purpose of solving general optimization problems. In this research, BBO is used to tune a proportional-derivative control system for real-world mobile robots. The authors show that BBO can successfully tune the control algorithm of the robots, reducing their tracking error cost function by 65% from nominal values. This chapter focuses on describing the hardware, software, and the results that have been obtained by various implementations of BBO.