Lidar Odometry and Mapping Optimized by the Theory of Functional Systems in the Parking Lot

We propose a lidar odometry and mapping (LOAM) method optimized by the theory of functional system to map the parking lot and estimate the pose of the ground vehicle in real-time, providing sufficient and reliable prior information for subsequent intelligent parking. An important intelligent component in the theory of functional system is action result acceptor. The constructed action result acceptor is able to correct the motion distortion in the LOAM algorithm by combining the information of IMU or GPS. The action result acceptor can evaluate the credibility of the data from various sensors, build a prediction model from these data, compare the predicted results with the results of the actual behavior of the vehicle, and then estimate the state of the vehicle through the extended Kalman filter, so as to correct results of LOAM algorithm.