Design And Implementation Of Indoor Autonomous Navigation System Based On UAV Platform

In recent years,UAV has been widely used in outdoor environment by relying on satellite navigation system.In contrast,UAV indoor applications are limited due to the lack of reliable navigation systems in indoor environment.In order to find a wide variety of practical applications,it is an urgent challenge and difficulty to break the limitation of UAV application scenarios that how to accomplish the autonomous navigation in uncharted indoor environment by the sensors mounted on UAV itself.For this reason,this paper takes the quadrotor UAV as a platform,relies on the twodimensional lidar detection environment,and uses the software framework based on ROS to implement two-dimensional laser SLAM and two-dimensional obstacle avoidance,to solve the problems of positioning and obstacle avoidance in the indoor navigation system.The main contents of this study are as follows:1、According to the hardware platform and theoretical requirements of indoor autonomous navigation system,the kinematics model of quadrotor UAV,twodimensional lidar observation model and occupying grid map model are established respectively.2、Aiming at the limitation of motion compensation processing for lidar data by Cartogrpher,A lidar data reverse correction improvement plan based on sequential container data structure and a data processing optimization scheme based on dynamic time-stamp growth value are presented.Experiments show that the improvement program can effectively reduce problems of Cartographer building map divergence caused by fast spin motion of lidar,and minimize effects of lidar frequency fluctuations on map details.3、In order to overcome the drawbacks of Artifical Potential Field,the problem of local minimum value and the problem of excess gravitation for distant targets are solved by introducing random search strategy and modifying gravitation formula.The effectiveness of these improved method is verified by simulation.4、The software design structure based on ROS is analyzed,the software design of autonomous navigation system is completed,and the software is used to simulate flight in virtual environment to verify the feasibility and stability of the system.The completed system software is deployed to the hardware platform,and flight experiments are carried out to verify the application value of the autonomous navigation system in this paper.