Earth sensor and accelerometer based autonomous heading detection algorithm research of lunar rover

Lunar rover heading detection is one of the most important part of rover localization. In this thesis, a absolute heading detection algorithm, which can be applicable in long-range and long duration navigation of lunar rover, is presented. Nonlinear least squares bundle adjustment is used to process the images from CCD earth sensor, consequently, the vector observation function of CCD earth sensor is achieved on lunar environment. Also based on the altitude information from accelerometer, the heading of lunar rover is calculated with a high-efficiency algorithm. The feasibility of this scenario is verified by digital simulation.