Research On Motion Planning Application Of Quadrotor Uav In Unknown Environment Based On Search

Due to its hovering,vertical take-off and landing,and the movement characteristics of flying in any direction within the design speed range,as well as the characteristics of simple structure,easy structure,and low cost,the four-rotor UAV is widely used in inspections,disaster relief,and agriculture.Plant protection and other fields.For quad-rotor drones,motion planning is to find a safe trajectory that the drone can execute from the current unknown to the given unknown in a given obstacle environment,and move in an unknown environment Planning is the key capability of drones to perform complex tasks.This article uses this as the application research background.From the software system design level,the research of this article mainly includes three key aspects: path planning,trajectory planning,and the control of the entire motion planning system in an unknown environment.This article first analyzes the motion planning application in the unknown environment based on the hardware platform and flight system of the quadrotor UAV,and divides the entire application into four key components: map representation,motion planning,safety detection,and state management.It is established that the problem of motion planning is the main research object,combined with the evaluation criteria of motion planning,and the use of path planning and trajectory planning to solve the problem of motion planning.For the path planning problem,this paper uses deterministic algorithms and uses the idea of path search to plan a shortest path in a real environment.Using the state grid algorithm,based on the kinematics of the quad-rotor UAV,the search graph is constructed through the discrete control space method and the graph search algorithm is used to find the optimal path.The state grid algorithm is improved,and the optimal path is solved by The boundary value problem is based on kinematics.The heuristic function of the algorithm is redesigned to increase the speed of the image search process,and the results are verified by simulation experiments.For the problem of trajectory planning,this paper adopts the idea of nonlinear geometric optimization.Based on the optimal path searched by the path planning problem,an executable trajectory of the UAV is planned.Through the analysis of the differential flatness problem,the quad-rotor The kinematic nature of the man-machine transforms the planning of the high-dimensional state of the UAV in the input space into the planning of the low-dimensional state of the flat space,which simplifies the complexity of the planning problem.In addition,the quadrotor UAV trajectory was parameterized,and the trajectory optimization was determined with the goal of minimizing Snap.Through comparative experiments,it was established that the trajectory planning problem was transformed into a nonlinear optimization problem using a soft-constrained trajectory optimization algorithm.The trajectory is parameterized with the help of B-spline curve,which speeds up the trajectory optimization,and the results are verified by simulation experiments.For the control problem of the entire motion planning system in an unknown environment,through path planning and trajectory planning,an executable trajectory is obtained in the environment.In order to realize motion planning in an unknown obstacle environment,this paper designs a safety detection algorithm for real-time obstacle detection on the currently executing trajectory,and designs a state machine-based control management algorithm to solve the ability of UAVs to re-plan in an unknown environment,And verified the results through simulation experiments.