Research On Autonomous Mapping And Navigation Technology In Indoor Environment Based On Lidar And MEMS Inertial Components

In recent years, natural disasters occurred frequently and terrorism form is still grim. It is a big challenge for micro small unmanned aerial vehicle(UAV) whether the tasks of unknown detection of indoor environment and indoor rescue, indoor anti-terrorism reconnaissance could be solved smoothly. Autonomous mapping and navigation in indoor environment is the key technology to those tasks mentioned above. Meanwhile, it is an important support for micro small UAVs to implement indoor flight control and path planning. Main work is done based on the Lidar/IMU for indoor mapping and navigation technology.First of all, according to the problem of SLAM implementation in indoor environment based on Lidar only, the characteristics associated scan matching SLAM algorithm based on complete endpoint is proposed. Complete endpoint feature extraction algorithm is acquired to solve the problem of feature points' change suffered from the relative movements. Based on linear fitting cognitive, the attitude angle weighted calculating method is carried to optimize the convergence of attitude angle. As for the map management and the closed-loop detection problem, the map database of backup feature method is proposed. The experimental results show that the algorithm of the characteristics associated scan matching SLAM algorithm based on complete endpoint can effectively realize the high-precision indoor positioning and geometric mapping. It avoids the problem of computational complexity increasement and the shortcoming of relying on closed-loop detection for probability based SLAM.In view of the Lidar/IMU integrated navigation problems, the micro SINS of random error parameter identification method is improved. Using ?-? filter to get the Lidar observation of speed and position, which greatly inhibit the noise of speed observation. Based on loose combination scheme of Lidar/IMU, state estimation of UAV is presented. Combined with random error parameter identification method of IMU and the pose calculation method of IMU, the navigation result is presented in a short period of time, which satisfies the requirement of indoor navigation.Finally, using micro small UAVs as the application aim and integrated navigation system factors such as cost, performance and load, the hardware platform of navigation system and the software system are set up. Through reasonable design of experiment scheme, both of the characteristics associated scan matching SLAM algorithm based on complete endpoint and Lidar/IMU integrated navigation method are proved effective. Results have certain reference value in practical application.