Research On UAV Indoor Obstacle Avoidance Navigation Technology Based On Lidar Detection

At present,autonomous flight of drones has become an important content in the field of drone research.In the indoor environment,the obstacle avoidance navigation is a prerequisite for the drone to complete other tasks.Aiming at the indoor dynamic environment,this paper proposes a Quadrotor UAV obstacle avoidance navigation method in a two-dimensional plane.First,an improved ant colony algorithm is used to generate a reference path,and extracts inflection points from the reference path as local target points.Then an improved artificial potential field method is used to guide the drone to fly.While ensuring that the drone can avoid obstacles,this method ensures that the flight path is as close to the optimal path as possible.The main research contents of this article are as follows:First,this paper proposes the overall design of the obstacle avoidance navigation system for the quadrotor drone.And the hardware composition and selection of the system are introduced.This system is divided into flight control module,data transmission module and obstacle avoidance navigation module,the software design and implementation for each module are finished.Secondly,the map form used for navigation and the indoor positioning method of drone are studied.Occupied grid maps are used to build the drone's flight environment model,and a match algorithm based on optimization is used to estimate the pose of the drone.This paper realized the mapping and indoor positioning of the drone.Then,in order to solve the problem that the traditional ant colony algorithm has a slow convergence speed when used for path planning,an improved ant colony algorithm is proposed.By changing the update range of the pheromone concentration,the convergence speed of the algorithm is accelerated while ensuring the diversity of the solution.At the same time,an improved artificial potential field method was proposed to solve problems of the traditional artificial potential field method,such as collision caused by great attractive force at the initial time,GNRON,and trapped in local minima.First,collisions between drones and obstacles are avoided by limiting the attractive force.Secondly,the UAV can reach the target point when the target point is close to the obstacle by introducing target information into the force field model.Finally,the escape when UAV trapped in local minima is achieved by defining the influence area of local minima and using the A~* search method to generate the escape path.Finally,a hybrid path planning algorithm combining ant colony algorithm and artificial potential field method is proposed to solve the problem that ant colony algorithm cannot avoid obstacles in real time and the path planned by artificial potential field method is long.By using the inflection point of the global path,which planned by the ant colony algorithm,as the local target point of the artificial potential field method,and using the artificial potential field method to generate navigation,the real-time obstacle avoidance navigation of the drone is realized.At the same time,an experimental platform for drones is set up,and the effectiveness of the method is proved by physical experiments.