Intelligent Robot Path Planning Based On Improved Particle Swarm Optimization And Ant Colony Algorithm

Intelligent mobile robot is an important content in robot research field. It centralizes the latest research achievements of many disciplines including such sectors as machinery, electron, computer, automatic control and artificial intelligence, etc. And it represents the paramount achievement in mechanotronics currently. Among the technical researches related to mobile robot, path planning technology has become an important research field.The key task of path planning is to search a smooth and collision-free path which can enable the mobile robot to reach the target point from the starting point smoothly in an obstacle space with certainty or uncertainty in line with certain evaluation criteria. In this thesis, after determining the obstacle space, the particle swarm optimization and ant colony optimization respectively improved and fused to calculate the optimal path as the evaluation criterion of the overall path planning. This thesis includes the follows:Firstly, the background and development status of path planning technology for intelligent mobile robot was introduced; meanwhile, the particle swarm optimization and ant colony optimization are also introduced and the respective principles of the two optimizations are analyzed. Given the problem of uneven search of particle swarm optimization, the thesis improves it and new location updating formula was given. As for ant colony optimization, given the fact that it cannot gradually tend to the optimal solution with the constant progressing of the search and the solutions obtained, the thesis properly changes the insufficient influence of pheromone concentration on ant colony, and improves it in three respects of local pheromone updating rule, overall pheromone updating rule as well as path concentration value interval.Secondly, the two improved algorithms are fused. The main process is to obtain a comparatively optimal feasible route in the obstacle environment by means of the improved particle swarm optimization, then gives an initial value to the path where this route is located according to the pheromone dissemination rule at the fusing time point proposed in this thesis, and finally solves the optimal solution by means of the improved ant colony optimization. Although the path planning technology based on particle swarm optimization and ant colony optimization has been widely researched both home and abroad, and some papers have also published on path planning by fusing the two algorithms, pheromone dissemination rule is proposed at the fusing time point firstly in this thesis. The process of disseminating pheromone concentration on the comparatively optimal route, i.e., the process of giving initial value can effectively guide the subsequent ant colony, which can accelerate the convergence rate of the algorithm and increase the solving accuracy.Thirdly, different obstacle environments are generated under MATLAB environment; three groups of simulation comparisons are carried out, including basic particle swarm optimization, improved particle swarm optimization and fusion algorithm, basic ant colony optimization, improved ant colony optimization and fusion algorithm, as well as improved particle swarm optimization, improved ant colony optimization and fusion algorithm. The performances of the algorithms are compared and analyzed by recording the indexes such as the shortest path, all the routes and average path length solved by each algorithm as well as the time used. Test simulation and index calculation can conclude that the improved particle swarm optimization and ant colony optimization can both improve the solving accuracy than before; moreover, fusion algorithm has the highest solving accuracy and the fastest convergence rate among the five algorithms.In addition, the fusion algorithm of particle swarm optimization and ant colony optimization can be applied to physical platform for the first time. So the effectiveness of the algorithm can be verified in physical environment by PIONEER 3DX intelligent mobile robot. The test of the physical platform can conclude that the improved particle swarm optimization and ant colony optimization can be well applied to the path planning problems for intelligent mobile robot, which can increase the robot’s ability of solving the optimal path, thus providing basis for exploring and researching more effective overall or local path planning schemes in future.Concluding remarks are presented at the end of this dissertation. Some open problems are also pointed out, which deserve further study.