Research On Autonomous Mobile Robot Path Planning Algorithm In Complex Environment

As one of the keys to the autonomous and intelligent mobile robot,path planning becomes the key research sub-domain.In this paper,the algorithms used in mobile robot path planning are studied,mainly from the global and local levels.With the Dr Robot X80Pro system as the research background,the corresponding validation experiments are made while optimizing the relevant algorithms.And apply the improved new algorithm to the platform.Aiming at the complex environment,a hybrid double-layer planning scheme is proposed.The scheme firstly uses the improved A*algorithm in the environment to pre-process and obtains the globally optimized trajectory,extracts the turning point of the path as the sub-target of the local path planning,realizes the real-time planning and processing by using the improved potential field method,The hybrid double-layer algorithm not only ensures the better global trajectory,but also real-time ability to cope with path planning tasks in a dynamic environment.The main research work done in this paper is as follows:(1)Drawing on the idea of raster method to complete the construction and representation of the environment,a better path can be get on the global map by completing the weighting and smoothing of the traditional A*algorithm and reducing the turning points and redundant points of the path.In addition,taking into account the actual robot has its own width problem,the strategy of generating point priority section is introduced to ensure the trajectory for the actual planning is safety.The path simulation experiment based on the indoor environment verifies that the improved A*can obtain a safe and optimal global trajectory.(2)The common the potential field method and the common problems of it are discussed.The common problem of minimizing point is eliminated by adding "distance factor" to the function of potential field and improving the direction of repulsion component.Aiming at obstacle avoidance of dynamic obstacle,velocity and acceleration information is introduced into the function of potential field to increase the ability of dynamic obstacle avoidance.By introducing the angle between the relative velocity and the relative position,the complex potential field function model is simplified.Through the simulation examples,it is verified that the new potential field algorithm can avoid the local minimum problem and can also deal with the dynamic programming well.(3)Aiming at the planning problem in complex environment,a double-layer scheme based on the new A*algorithm and the new potential field method is designed.The idea and implementation steps of the hybrid scheme are constructed,and the comparison experiments under static and dynamic environments are described.The superiority of the hybrid solution is verifies,based on the Dr Robot X80Pro system,the related tests is completed and further confirms that the dual-layer solution can effectively cope with planning tasks in complex environments.