UAV Motion Planning For Road Bridge Inspection Task

As the service life of bridges grows,bridge diseases gradually increase,and regular bridge inspection and maintenance are required.However,traditional inspection methods have the disadvantages of low inspection efficiency,high cost,poor flexibility,and the need for traffic control.In recent years,the technology of rotor drones has developed rapidly.The use of drones for bridge maintenance tasks has the advantages of low cost,high flexibility,and low impact on traffic.Therefore,it has a very broad application prospect.Bridge inspection has problems such as complex environment,many obstacles,and occlusion of the field of view,which brings challenges to autonomous motion planning based on UAVs.Therefore,the research on motion planning methods for bridge inspection tasks needs to be deepened.This thesis mainly studies the UAV motion planning method for bridge inspection tasks,and the main work is as follows:Firstly,the coordinate system definition and conversion relationship required for the inspection task are given.The dynamic model of the drone and the detection model of the camera are established.The inspection process is divided into three stages of approach,coverage,and return.It is divided into target-based motion planning and coverage planning.Mathematical descriptions of these two types of motion planning problems are established.Secondly,for the problem of target-based motion planning,a grid map model is established,a path planning method based on improved A* algorithm is proposed,and a trajectory generation method based on Bezier curve optimization is given.Thirdly,for bridge deck coverage,a coverage path planning method based on Morse decomposition and Euler loop is proposed,a mathematical description of the UAV coverage space is given,and a coverage trajectory generation method based on Bezier curve is proposed.For the bridge piers,a coverage trajectory generation method based on circular trajectories and straight trajectories is proposed,and a quantitative estimation of the coverage time of the piers is given.Then,to solve the problem that the original trajectory is unavailable due to sudden obstacles and inaccurate information,a path re-planning algorithm based on Elastic Band is proposed,and a B-spline based trajectory generation algorithm that guarantees trajectory stability and meets kinematic constraints is proposed.Finally,a simulation experiment platform is built,the algorithm flow is given and the proposed algorithms are implemented,and the feasibility and effectiveness of the algorithm are demonstrated through simulation experiments.