| In recent years,robots have become more and more popular in human social life and work.Robot employees have appeared in various fields such as industry,agriculture,commerce,and service industries.Subsequently,people have higher requirements for robot efficiency and safety.The research and implementation of intelligent path navigation planning methods are the prerequisites and guarantees for mobile robots to complete other additional tasks.Path planning algorithm is one of the basic capabilities in the development of robot autonomous navigation integrated technology.Many researchers have carried out a lot of theoretical and application research on this.However,each algorithm has its own limitations and cannot perfectly adapt to changes in various environments.Therefore,it is necessary to conduct more in-depth research on path planning algorithms.The traditional ant colony algorithm and artificial potential field algorithm are used to study and explore the path planning problem of a single algorithm.The scheme design of two algorithms applied to path planning are introduced respectively,and simulation experiments are performed on the two algorithms in Matlab simulation software.By analyzing the experimental results,the shortcomings and defects of the two traditional algorithms are summarized: the traditional ant colony algorithm has problems such as local sub-optimal solutions,low efficiency of ant colony optimization,and difficulty in achieving stable convergence;the traditional artificial potential field method has problems such as minimum points and path oscillations.In view of these defects,the optimization scheme of the improved potential field ant colony algorithm is designed to improve the path planning performance of the fusion algorithm.Introduce potential field force heuristic information to improve the calculation method of transition probability.According to the ideas of different algorithm advantages fusion and learn from each other,use artificial potential field force factors to construct potential field force heuristic information,optimize the probability calculation method,and effectively improve the early ant colony search of the algorithm.Inefficiency.At the same time,the influence parameters of potential field force heuristic information are introduced to avoid prematureness to obtain a local optimal solution.Improve the heuristic information constructor,dynamically update the heuristic information according to the actual distance between the path node and the end point in the algorithm operation,increase the reasonable induction ability of the actual distance heuristic information to the ants;improve the pheromone update method to improve the effective information utilization efficiency of the ant colony,increase local Path pheromone reward and punishment mechanism,which rewards each node of the complete path with pheromone,and at the same time reduces the pheromone of each node of the incomplete path caught in a deadlock;proposes an adaptive pseudo-random state transition mode to accelerate the algorithm's convergence,and introduces a pseudo-random probability Select the coefficient and change its value to control the state of the path selected by the algorithm,and adjust the definiteness and randomness of the ants selecting the next path node.Simulation experiments were performed in different map environments,and the improved potential field ant colony algorithm was compared with the traditional ant colony algorithm and other improved algorithms.Finally,the algorithm is transplanted on the mobile platform to complete the design of the upper computer controlled by the platform.The improved algorithm of this paper is transplanted into the mobile platform,and the movement method of the platform is analyzed and designed.The ability of path planning is tested in a real environment,and the complete path planning can be realized under the control of host computer. |
PDF Full Text Request