Sensors relative calibration method for landing navigation based on feature matching

In the vision-aided inertial navigation system, optimal information fusion depends on accurate calibration of the six degrees-of-freedom transformation between a camera and an inertial measurement unit. Considering the measurement information optimal fusion problem of autonomous navigation during soft landing on Mars, a sensor-to-sensor relative pose calibration algorithm was proposed based on the extended Kalman filter. The proposed algorithm can accurately calibrate the relative pose of the camera and inertial measurement unit, and simultaneously estimate the position, velocity and attitude of the spacecraft. Moreover, obtaining this calibration information requires no additional measurement equipment except the landmark features on the surface of the Mars. Furthermore, high fidelity sensor models for wide field-of-view camera and inertial measurement unit were developed taking into account effects of the probe maneuver and the Mars rotation. Finally, the validity of the sensors calibration algorithm presented in this paper was demonstrated by mathematical simulation.