Introducing Particle Swarm Optimization into a Genetic Algorithm to Evolve Robot Controllers’

Malte Langosz, Kai von Szadkowski

In: Proceedings of the RIC Project Day Workgroups Locomotion&Simulation. RIC Project Day September 17-17 Bremen 14_06 1406 Selbstverlag 9/2015.


This paper presents Swarm-Assisted Behavior Graph Evolution (Sabre), a genetic algorithm which combines elements from genetic programming and neuroevolution to develop Behavior Graphs (Bgs). Sabre evolves graph structure and parameters in parallel using particle swarm optimization (Pso) for the latter. The algorithm’s performance was evaluated on a set of black-box function approximation problems, one of which represents part of a robot controller. We found that Sabre performed significantly better in approximating the mathematically complex test functions than the reference algorithms genetic programming (Gp) and Neat.

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence