Mobile Robot Path Planning with a Novel Multiobjective Slime Mold Algorithm

Abstract

In the path planning problem of mobile robot in static complex obstacle environment, the original slime mold algorithm (SMA) has some shortcomings, such as initial solution error and easy to fall into local optimum, which leads to low accuracy of robot path planning. A novel Baldwin learning effect mobile robot Path planning with Multi-Objective Slime Mold Algorithm (BPMOSMA) is proposed to solve the related problems. The Baldwin learning effect is introduced into to guide the development direction of slime mold in the phase of slime mold surrounding food and the k-nearest neighbor measure is added to prune nondominated solutions to generate a well-distributed solution set. For comparison purposes, five state-of-the-art multiobjective optimization algorithms are carried on nine multiobjective test problems with kinds of types of Pareto set. The qualitative and quantitative experimental results show that BPMOSMA is very competitive performance on unconstrained multi-objective optimization problems. BPMOSMA is finally successfully applied in the mobile robot path planning problem.