Research Of Path Planning For Mobile Robots Based On Particle Swarm Optimization Algorithm

Path planning for mobile robots is a complex problem that not only guarantees a collision-free path with minimum traveling distance but also requires smoothness and security. This dissertation presents an improved particle swarm optimization algorithm for solving the path planning problem in static mobile robot environments.Firstly, by researching and analyzing the disadvantages of standard particle swarm algorithm which traps into the local optimum value easily, the paper develops a progressive particle swarm optimization algorithm, according to adding the third progressive factors to guide the current particles'velocity updating rule and random re-initialization at particles which go beyond the default position space in each interiors, that can keep the diversity of the particles and reduce the probability of trapping into the local optimization in the later iteration, so keep part from getting into local optima. Five benchmark functions are tested, and the results indicate the feasibility of the improve algorithm as well; Secondly, according to the characteristics of path planning problem, adopt the grid theory to establish the working space of the mobile robot, every component of the improved algorithm is researched carefully, to optimize the path which doesn't meet the constraints during the process, and to reflect the quality of the path by combining path length, smoothness with security; Finally, the computer simulate experiment are carried out in different static environments, by introducing two separate indicators, i.e., path length and dynamic convergence behavior, to evaluate the efficiency of optimization. In addition, the paper emphasizes and analyzes that different parameters of the fitness function coefficients have a great effect on path planning; progressive particle swarm optimization algorithm is compared and analyzed with other algorithms as well.It's proved that the improved algorithm can plan an approximate optimal path successfully in different static environment, and validate the effectiveness of the proposed approach.