Research On Navigation Technology Of Substation Inspection Robot Based On 2D Laser SLAM

Introduction Substation is an important component of the power system,serving as an important task of power transformation and distribution.In order to ensure the safe and stable operation of the substation,staff need to regularly inspect the equipment in the substation.However,manual inspection has shortcomings such as easy misjudgment,large workload,and low efficiency.In addition to the huge and complex structure of modern power systems and a wide variety of equipment,manual inspection has long been unable to meet the requirements of modern power systems for safe and stable operation.Therefore,there is an urgent need for an intelligent device that can replace manual inspection to meet the requirements of power system inspection.In recent years,with the development of science and technology,artificial intelligence technology and robotics technology have gradually matured.Introducing patrol robots into substations to assist or replace manual patrol is an important way to change the working mode of line patrol and achieve automation,intelligence,and efficiency in substation patrol work.In particular,the development and maturity of robot navigation technologies such as real-time positioning and map building(SLAM)and path planning have been highlighted,This provides a technical guarantee for the patrol robot to patrol the power system under complex conditions.Aiming at the requirements of the development of new power systems,this paper conducts research and improvement on mapping,positioning,and path planning technologies related to ROS based substation inspection robot navigation.The main research contents and innovations of this article are as follows:Firstly,this paper analyzes the working environment of the substation inspection robot,and determine whether to use the tracked vehicle as the moving chassis of the robot,and the laser radar is used as the sensor for the robot to sense the external environment.Then,the composition of the robot hardware system,software system,and communication system is introduced.Based on ROS,the URDF model of the robot is built and optimized.The robot simulation environment is built in Gazebo and the laser radar ranging function is simulated.Secondly,two 2D laser slam algorithms,Gmapping and Hector,were studied and simulated in Gazebo.The advantages and disadvantages of the two algorithms were obtained by comparing their mapping effects.Then,the principle of AMCL localization algorithm is analyzed,and the localization simulation of AMCL algorithm is completed based on Gazebo,which verifies the effectiveness of AMCL algorithm.Then,Ant Colony Algorithm and Artificial Potential Field Method are selected as the global path planning algorithm and local path planning algorithm for robots.The basic principles of Ant Colony Algorithm and Artificial Potential Field Method are described respectively,and simulations are conducted based on MATLAB software to verify the effectiveness of the two algorithms.Then,aiming at the problem that traditional ant colony algorithms are prone to fall into local optimization,a strategy of randomly updating pheromones using adaptive ant colon.An improved adaptive ant colony algorithm is proposed to solve the problem of single search direction in traditional ant colony algorithm and adaptive ant colony algorithm,which improves the efficiency of path search.Then,based on MATLAB software,the improved adaptive ant colony algorithm and the artificial potential field method are combined,and the potential field function is added to the heuristic function of the ant colony algorithm,making the algorithm not only accelerate convergence speed,but also avoid falling into local optimization,while reducing unnecessary path search.Then,a path planning and dynamic obstacle avoidance simulation experiment was conducted on the fusion algorithm based on ROS.The global optimal path was planned using an improved adaptive ant colony algorithm,and the dynamic obstacles encountered while driving on the optimal path were avoided using the artificial potential field method to complete local path planning.The simulation results show that the fusion algorithm shortens the path length,improves the path planning efficiency,and improves the autonomous navigation ability of the robot,verifying the feasibility of the fusion algorithm used in the navigation system.Finally,Hector SLAM mapping algorithm,AMCL localization algorithm,improved adaptive ant colony algorithm and artificial potential field algorithm are tested.Mapping experiments,localization experiments,autonomous navigation experiments,and dynamic obstacle avoidance experiments are conducted using the robot experimental platform.The experimental results show that the fusion algorithm can enable the robot to quickly search for the global optimal path in complex environments,and can avoid dynamic obstacles during driving,verifying the feasibility and effectiveness of the fusion algorithm on real robots.