Research On Path Planning Of Substation Inspection Robot Based On ROS

As an important part of the power grid,the substation maintains the security and stability of the power system.In order to ensure the normal operation of the substation,the staff need to conduct regular inspections of the power equipment.However,manual inspections have large workload,low efficiency,and easy misjudgment.Therefore,the inspection robot is used to replace the staff to complete the inspection task.The difficulty of the current inspection robot is that according to the requirements of the inspection task,it passes through the power equipment stops of the substation,identifies obstacles in the environment and plans a safe and collision-free inspection route.Therefore,this paper adopts the inspection robot based on the ROS system to carry out the path planning research.The main work of this paper is as follows:First,this paper analyzes the characteristics of the substation environment,builds a ROS development environment,simulates the substation environment with the indoor environment,and uses the Gmapping algorithm and the Cartographer algorithm to construct an environment map.Secondly,the improved A~* algorithm is used for global path planning research.By quantifying the obstacle rate,this paper integrates the obstacle rate information into the search method of the A~* algorithm,improves the evaluation function,increases the algorithm search speed,and shortens the path planning time;optimizes the selection method of sub-nodes to avoid the risk of obstacle collision;adopts the path bidirectional smoothing optimization The algorithm removes redundant nodes in the path and improves the smoothness of the path.The original A~* algorithm and the improved A~* algorithm are compared and simulated on the MATLAB platform.Then the improved DWA algorithm is used for local path planning.By introducing the narrow-band passing algorithm,the situation that the inspection robot may not be able to pass in the narrow area is improved;filter processing is added to the cost map,and the guide line is improved according to the planned trajectory,thereby improving the auxiliary navigation point and increasing the time when the inspection robot passes through the narrow road.success rate.The improved DWA algorithm is simulated on the Gazebo platform.Finally,this paper introduces the global trajectory node into the target orientation evaluation function of the dynamic DWA algorithm,which is used as the local target point of the dynamic programming to integrate the improved A~* algorithm and the improved DWA algorithm.And in the indoor environment,the hybrid algorithm proposed in this paper and the hybrid algorithm fused by the original A~* algorithm and the original DWA algorithm are used to conduct a comparative test using an inspection robot.The research results show that the improved A~* algorithm reduces the path planning time by 9.5% on average,the path planning length by 4.53% on average,and the number of route turns by 44.3% on average.The success rate of the improved DWA algorithm through the narrowband region has been increased from 60% to 80%.The hybrid algorithm proposed in this paper reduces the path search time of the inspection robot by an average of 31.7% and the path planning length by an average of 26.7%.And when local obstacle avoidance is performed in a narrow indoor environment,it can successfully pass through the narrow road area.