PSO Based Path Planning and Dynamic Obstacle Avoidance in CG Space of a 10 DOF Rover

Abstract

Dynamic path planning is a core research content for intelligent robots. This NP-hard problem is complex enough for an algorithm to plan a realistic path for a velocity-bound point robot in a dynamic environment. This paper presents a new CG-Space based real-time dynamic path planning method for finding the obstacle-free path for 10 DOF rocker-bogie type wheeled mobile robot (Rover) traversing over a 3D uneven terrain with dynamic and static obstacles. CG-Space is the locus of the center of gravity location of Rover while moving on a 3D terrain. A swarm intelligence-based Particle Swarm Optimization (PSO) method has been modified to plan an optimum collision-free path over CG-Space of Rover with dynamic obstacles. Dynamic replanning using a modified penalty in the PSO objective function can handle the randomly moving obstacles of varying sizes and shapes in real-time. Simulations demonstrate that the Rover can obtain the target location in 3D uneven dynamic environments with fixed and randomly moving obstacles. This 3D dynamic replanning study will benefit the cooperative sensing in multiple Rover systems and target tracking in cluttered environments.