Research On Path Planning Technology Of Mobile Car In Indoor Dynamic Environment

The autonomous navigation of mobile robots is mainly divided into positioning and path planning,and path planning is the decision-making part of autonomous navigation.Due to the uncertainties of the environment and its own state,the requirements for the online path planning algorithm of the mobile robot are high,making the path planning of the mobile robot under dynamic environment a highly complex problem.However,the existing path planning algorithms have not been able to solve the robot oscillation and deadlock problems effectively,and the local path planning algorithm is still relatively lacked.Therefore,there is practical significance to study the path planning technology of mobile robots in dynamic environment.A mobile robot platform based on Robot Operating System(ROS)is built in this paper,and mainly research on the global path planning and local path planning method.Then the algorithms are transplanted into the mobile robot's embedded platform.Simulation experiments and prototype experiments are used to verify the effectiveness of the algorithm.The specific research work and achievements of this paper are as follows.In this paper,the ROS-based mobile robot platform was built,it can run normally and perform human-computer interaction.In the global path planning,a global optimal path is proposed using the improved artificial potential field method(APFM)and optimized by the particle swarm optimization algorithm(PSO).In order to solve the problem of local optimum problem existing in the path planning of traditional APFM,it is proposed to make the area unreachable by filling the local minimum area with obstacle potential field.Then,the path generated by the improved APFM is optimized by PSO algorithm.In addition to the path length,the evaluation function of the PSO algorithm also includes the evaluation factors of path safety and smoothness.The feasibility and superiority of this method is verified by simulation experiments.In the local path planning,the solution to the problems existing in the traditional dynamic window approach(DWA)method is given.The DWA and the behavior control idea are combined to plan the local path.According to the sensor's feedback information in the rolling window,the environment is predicted to determine whether the mobile robot will collide with an obstacle: If there is a collision risk,the robot takes a behavior to avoid obstacles,otherwise,a tracking path behavior is taken.The feasibility of the algorithm is verified by simulation experiments.Finally,the improved path planning algorithms proposed were verified on the mobile platform in the complex static and dynamic environment separately.The results show that under the static environment,the path length,smoothness and security indexes of the PSO-APFM algorithm proposed in this paper are better than the A* algorithm.In dynamic environment,the improved DWA algorithm proposed in this paper is better than the traditional DWA,in the aspect of dynamic obstacle avoidance time and the stability.