Research On Intelligent Algorithm Of Path Planning Based On Sampling

Path planning is an important research content in autonomous navigation of mobile robots.This paper studies path planning algorithms based on sampling ideas.By analyzing its existing problems,a duple constraints RRT algorithm and an interactive intelligent path planning algorithm for mobile robots based on geographic information are proposed,and the proposed algorithm is validated by simulation.The main research contents of this article are as follows:Firstly,duple constraints RRT algorithm is proposed.To solve the problem of poor sampling quality of RRT algorithm,a probabilistic sampling pool is constructed to improve the quality of sampling points by restricting the distance between the points in the sampling pool and the planned end point.Aiming at the problem that the angle amplitude of path planning by RRT algorithm is too large,and considering the kinematic constraints of mobile robots,a new node generation function containing length and angle is proposed to constrain the new nodes to reduce the path turning angle.Then,an interactive intelligent path planning algorithm based on geographic information is constructed.Aiming at the problem that the duple constraints RRT algorithm doesn’t consider the actual size of the robot and isn’t suitable for practical operation,a collision detection method based on the simplified obstacle model is proposed.Aiming at the low utilization of map information by duple constraints RRT algorithm,an efficient expansion method based on geographic information is proposed.Firstly,the current node neighborhood and target neighborhood are determined according to the map information,and sampling points are generated according to obstacle information in the current node domain.An initial path is efficiently formed through dynamic step size combined with node expansion.In order to solve the insufficient smoothness of the path of duple constraints RRT algorithm,the redundant nodes are eliminated based on skip point filtering and maximum rotation constraint,and then a cubic spline curve is used to smooth the path.In order to verify the effectiveness of the proposed algorithm,simulation experiments are conducted on Jupyter Notebook,and then the simulation experiments are built using a robot operating system.The experimental results are analyzed to further verify that the proposed algorithm can effectively shorten the path length,reduce the path finding time,smooth the path planning,and improve the efficiency of the algorithm compared with the traditional RRT algorithm on the premise of ensuring the success rate of planning.