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

Area of Science:

• Lunar exploration
• Robotics and autonomous systems
• Geospatial information science

Background:

• Chang'E-3 mission aimed for China's first extraterrestrial soft landing.
• Accurate 3D terrain data is vital for safe landing site selection.
• The Three-Dimensional Imaging (TDI) sensor is a key payload for lunar surface mapping.

Purpose of the Study:

• To develop and validate a geometric calibration method for the Chang'E-3 TDI sensor.
• To assess the impact of TDI intrinsic parameter errors on 3D reconstruction.
• To ensure reliable terrain data for autonomous obstacle avoidance during lunar landing.

Main Methods:

• Construction of a simplified positioning model for the TDI sensor.
• Design and implementation of a permanent outdoor calibration field with flat plates.
• Application of virtual observation equations for robust geometric calibration adjustment.
• Verification of calibration accuracy using multi-measurement and multi-angle imaging data.

Main Results:

• TDI intrinsic parameter errors can create false obstacles up to 1.4m high, potentially leading to landing failures.
• The TDI sensor exhibits good intrinsic parameter stability.
• Post-calibration, the reconstructed 3D lunar surface accuracy reaches approximately 4 cm.

Conclusions:

• Precise geometric calibration of the TDI sensor is essential for the Chang'E-3 mission's success.
• Accurate 3D terrain mapping provides a reliable foundation for autonomous obstacle avoidance.
• The validated calibration method ensures the TDI sensor's capability to support safe lunar landings.