Research And Application Of Path Planning Based On Swarm Intelligence Optimization Algorithm

Nowadays,the application field of path planning is gradually expanding,and it has become one of the key technologies for the development of some industries.However,the traditional research mainly focuses on the algorithm analysis and improvement of the global path planning problem in static environment,but it is hard to get the hopeful target in the face of complex dynamic environment.This paper aims at the problems of global path planning algorithms,such as poor search ability,easy to fall into local optimal and unable to dynamically avoid obstacles.By combining ant colony optimization and rolling window method,a fusion algorithm for path planning in complex dynamic environment was proposed to solve this problem and improve the actual production efficiency.In this paper,the principles of three common and excellent swarm intelligence algorithms are systematically analyzed.The global path planning results of the three algorithms in static environment are compared by simulation experiments.Considering the main application fields and the characteristics of the algorithm,we finally select the ant colony algorithm with good robustness and strong search ability for improvement and optimization.In this algorithm,a back-off strategy is introduced and parameters are dynamically adjusted to improve the searching ability of the algorithm and avoid being caught in local optimum.In the meantime,quadratic programming is carried out to shorten the path length as much as possible.For the sake of settling the question of local path planning,this paper compares the advantages and disadvantages of the rolling window method and the artificial potential field method,and finally selects the rolling window method with better real-time performance.Based on this,a two-level safety distance determination rule is proposed to assist the implementation of dynamic obstacle avoidance strategy.Based on the selected ant colony optimization and the rolling window method,the algorithm fusion is carried out after reasonable optimization.The performance superiority of the improved ant colony algorithm and the rolling window method are given full play to solve the path planning problem in complex environment.In order to verify the actual function effect of the algorithm,this paper uses Gazebo platform to create a 3d scene simulation experiment.In the experiment,Turtlebot robot is chosen as the object of study,the corresponding static and dynamic obstacles were set in the environment.The experimental results show that The Turtlebot can avoid obstacles and reach the designated area smoothly,which further verifies the effectiveness and practicability of the fusion algorithm.