Research On Positioning And Path Planning Technology Of Inspection Robots

With the development of robotics,mobile robots are gradually used in more and more fields because of their large range of activities and excellent movement ability.The traditional manual inspection method has the disadvantages of high cost,low efficiency,and untimely information update,so the use of mobile robots instead of manual inspection has gradually become a new development trend.Due to the shortage of sensors and algorithms,the current inspection robots have problems in positioning and path planning,such as low positioning accuracy and instability,low planning path security,and unsmooth paths,respectively.In view of the above problems,this paper designed the overall framework of the system of inspection robot,selected Li DAR as the main observation sensor,and based on this,proposed the framework of inspection robot positioning algorithm combining relative positioning and absolute positioning.The path planning algorithm framework of the inspection robot consists of global path planning and local path planning.The main research contents of this paper are as follows.(1)Aiming at the problems such as particle degradation affecting localization accuracy in particle filtering localization algorithm,a global localization algorithm combining improved particle filtering and scan matching was proposed in this paper.Firstly,the systematic and non-systematic errors of odometry are corrected to improve the relative positioning accuracy of the robot.Secondly,the proposed distribution of the particle filtering algorithm was improved using the motion state model incorporating sensor information in relative localization to solve the problem of degradation of localization accuracy due to particle degradation.Adaptive resampling was also introduced to reduce the computational cost.Finally,the Iterative Closest Point(ICP)idea was combined,and the rough global pose obtained by the improved particle filtering was used as the initial value for iterative scan matching of lidar data and map to further improve the global positioning accuracy.Simulation and comparison experiments are conducted between the original algorithm and the proposed algorithm,and the experimental results showed that the proposed localization algorithm has obvious advantages in terms of localization accuracy,response time and stability.(2)To address the problems of the global path planning algorithm A*,such as tortuous planning paths and routes close to obstacles,this paper improved the estimation function,node search method and path generation method of the A*algorithm,respectively.Firstly,the heuristic cost of the parent node was considered in the valuation function of the current node to make the node search more directional.Secondly,nodes that were tangent to the obstacle were ignored when searching for nodes,so that the generated paths were kept at a distance from the obstacle.Finally,the nodes whose connections do not pass through the obstacles in the path were removed to reduce the redundant nodes and achieved the purpose of smoothing the path.The experimental results showed that the improved A* algorithm reduced the running time by 9.15% and the number of nodes searched by 11.27%.The generated paths are smoother and avoid obstacle edges.In order to verify the effectiveness of the improved algorithm in practical applications,this paper built an experimental platform for the inspection robot according to the designed hardware and software system framework,and conducted experiments on the positioning and path planning of the inspection robot.The experimental results showed that the positioning accuracy,stability and path planning effect of the inspection robot can meet the practical application requirements and achieve the expected goal.