Robot Path Planning Based on Chaos Concise Differential Evolution and RFNN Control

Abstract

In order to reduce the memory footprint and energy consumption of embedded microcontroller in mobile robot, the concise differential evolution algorithm based on chaotic local search (CDE-CLS) is proposed for online optimization of recurrent fuzzy neural network (RFNN) controller in robot path planning so that the robot can be adaptive real-time obstacle avoidance. The CDE-CLS algorithm reduces the memory footprint of the controller using virtual population and increases the ability to explore help to fast convergence introducing a simple and efficient chaotic local fine search and inhibit premature convergence perturbing the virtual population. Contrast tests on the typical Benchmark functions verify the global convergence and stability of the algorithm comparing with other concise evolutionary algorithm. Finally, the simulation result on the robot path planning controller shows the effectiveness of the proposed method.