Mobile robot navigation using PSO-optimized fuzzy artificial potential field with fuzzy control

• Published in: Journal of intelligent & fuzzy systems Vol. 32; no. 6; pp. 3893 - 3908
• Main Authors: Abdalla, Turki Y., Abed, Ali A., Ahmed, Alaa A.
• Format: Journal Article
• Language: English
• Published: London, England SAGE Publications 01.01.2017; Sage Publications Ltd
• Subjects: Algorithms; Computer simulation; Control methods; Fuzzy control; Fuzzy logic; Minima; Motion control; Navigation; Particle swarm optimization; Planning; Product design; Proportional integral derivative; Robots; Trajectory planning; Unknown environments; fuzzy artificial potential field; Path planning; artificial potential field; PSO; fuzzy logic
• ISSN: 1064-1246; 1875-8967
• DOI: 10.3233/IFS-162205

The objective of this paper is to develop a path planning algorithm that is able to plan the trajectory of mobile robots from its start point to target point in static and dynamic unknown environments. The classical artificial potential field (APF) method is not sufficient and ineffective for that purpose since it has the problem of local minima. To enhance the performance of the classical APF algorithm and to produce a more efficient and effective path planning for mobile robots, a new method based on combination of a modified APF algorithm with fuzzy logic (i.e. FAPF) is proposed. The proposed algorithm is designed to overcome the problems of the classical APF especially the local minima and enhances the navigation in complex environments. The fuzzy logic controller (FLC) is also used for motion control of the mobile robot. The membership functions of the FLC are optimized with particle swarm optimization (PSO) algorithm for optimality. Simulation models for the proposed path planning and motion control methods are built with MATLAB. Simulation results are obtained and proved that the robot with FAPF navigates with smoother path, react much faster in static and dynamic environments, and avoid obstacles efficiently. The work is compared with other implementations that used conventional PID controllers. All the system is then implemented practically to prove the proposed algorithms and tested in complex and unknown environment.