<![CDATA[The path planning (PP) problem is the most concerning and studied in the field of mobile robots. In this paper, we propose an improved non-dominated sorting genetic algorithm (INSGA-II) to solve the multi-objective optimization problem of path planning. We optimized the three objectives of path length, path safety, and path smoothness. Firstly, the RRT algorithm and elite retention strategy were applied to population initialization to improve the distributivity and evolutionary efficiency of the population. Then, the population evolution process was divided into two stages—the population macro-evolution stage and the population micro-adjustment stage. The former stage employed crossover, mutation, and shortening evolutionary operators, which effectively increased the evolutionary speed of the population. The latter stage used the point update operator, which increased the population's chances of rapidly convergent to the pareto optimal solution and prevented it from falling into local optimum. Finally, to verify the algorithm, we used several typical maps and analyzed the influence of three parameters on the algorithm. This algorithm outperformed the most advanced algorithms, MO-PSO and MO-FA, in terms of solutions and convergence.]]>
Copyrights © 2023
