Fast Positioning Model and Systematic Error Calibration of Chang’E-3 Obstacle Avoidance Lidar for Soft Landing

Abstract

Chang‘E-3 is China’s first soft landing mission on an extraterrestrial celestial body. The laser Three-Dimensional Imaging (TDI) sensor is one of the key payloads of the Chang‘E-3 lander. Its main task is to provide accurate 3D lunar surface information of the target landing area in real time for the selection of safe landing sites. Here, a simplified positioning model was constructed, to meet the accuracy and processing timeline requirements of the TDI sensor of Chang‘E-3. By analyzing the influence of TDI intrinsic parameters, a permanent outdoor calibration field based on flat plates was specially designed and constructed, and a robust solution of the geometric calibration adjustment was realized by introducing virtual observation equations for unknowns. The geometric calibration and its absolute and relative positioning accuracy verification were carried out using multi-measurement and multi-angle imaging data. The results show that the error of TDI intrinsic parameters will produce a false obstacle with a maximum height of about 1.4 m on the plane, which will cause the obstacle avoidance system of Chang‘E-3 to fail to find a suitable landing area or find a false flat area. Furthermore, the intrinsic parameters of the TDI have good stability and the accuracy of the reconstructed three-dimensional surface can reach about 4 cm after error calibration, which provides a reliable terrain guarantee for the autonomous obstacle avoidance of the Chang‘E-3 lander.

Visual Navigation and Obstacle Avoidance Control for Agricultural Robots via LiDAR and Camera

“…The pose coordinate of the LiDAR is (x s , y s , φ), where φ is the angle from axis O v X v to axis O s X s of the robot coordinate system. Suppose the coordinate of LiDAR coordinate point s under robot is (x vs , y vs ), then the rotation translation of the coordinate system is as shown in Equation (3) [28].…”