Research On Path Planning And Trajectory Tracking Algorithm For Mobile Robots

In recent years,the application scope of mobile robots has been continuously expanded and widely used in various indoor scenes such as production workshop,catering service,logistics management and daily life,therefore,the requirements for robot autonomy are also constantly improving.However,the indoor environment is often much more complex than the outdoor environment,which poses a greater challenge to the autonomous navigation performance of mobile robots.Autonomous navigation includes two core parts: path planning and trajectory tracking,however,there is still room for improvement in distance cost of path planning,accuracy of trajectory tracking.Therefore,this paper makes further research and exploration on the related algorithm of path planning and trajectory tracking involved in the process of robot autonomous navigation.Path planning is one of the core research problems in autonomous navigation technology of robot,among the existing path planning algorithm,the classic heuristic search algorithm-A~* algorithm has become the mainstream solution algorithm because of its high algorithm efficiency and simple implementation.In this paper,in order to further reduce the value of distance generation,A~* algorithm is combined with differential evolution algorithm.According to the known environment map,the corresponding grid model is established,A~* algorithm is used to search path quickly in advance,and then set a certain ratio to expand it,a series of feasible points are generated randomly in the region,and the path through the feasible points generated by A~* algorithm is used as the initial group of differential evolution algorithm,further mutation,crossover and selection operations are carried out to find the optimal path.In view of the proposed algorithm in this paper,under the Matlab simulation environment,build a complicated grid model and give a starting point and a target point,and the path planning simulation experiment is carried out,A~* algorithm is combined with differential evolution algorithm,respectively control population size,relaxation coefficient,extension ratio,copy location ratio,crossover probability and so on with single factor design method to determine the optimal parameters,and then obtain the optimal path.The experimental result show that the fusion algorithm of A~* and differential evolution can always find path shorter than that of A~* alone,the experiment shows that the distance cost of the path is lower and the path is better.After solving the optimal path,another core problem that needs to be solved is the trajectory tracking problem of robot.In this paper,a trajectory tracking control law is designed using back-stepping method based on adaptive grid ant colony algorithm.However,the initial parameters of the control law designed by the back-stepping method are uncertain and need to be adjusted continuously manually,which leads to the randomness of the tracking effect,Aiming at the back-stepping method initial parameters selection blindness which affects the controller design efficiency and robustness problem,the ant colony algorithm with strong robustness is introduced in this paper to optimize the back-stepping method,the parameters can be given automatically on the premise of ensuring the minimum error of the accumulated position and posture to improve the efficiency of the back-stepping method.At the same time,the adaptive grid method is combined with ant colony algorithm to further improve the accuracy of the solution.Aiming at the algorithm proposed in this paper,a wheeled mobile robot model is established.In the Matlab simulation environment,trajectory tracking simulation experiments are carried out on three typical curves,namely,line with infinite curvature,circle with fixed curvature,attenuation sine curve with unstable curvature,and two commonly used curves in practice,namely rectangle and domed rectangle,the pose error curve and the cumulative V value error are also given.The experimental result show that the proposed algorithm can automatically select the initial parameters of the back-stepping method which can minimize the cumulative V value,and achieve a good trajectory tracking effect.Aiming at the problems existing in the existing algorithm of path planning and trajectory tracking related to the autonomous navigation of mobile robots,this paper has carried out targeted improvement research,on the basis of theoretical analysis,has carried out sufficient simulation experiment research,which provides theoretical reference for the research in related fields.