Estimation of motion state for small celestial body detection based on sequence images

Abstract

Recently, with advances in aerospace technology, most countries have gradually shifted their attention to detecting small celestial bodies, which belong to the noncooperative space target task. In contrast to the cooperative space target, the noncooperative space target has typical characteristics, such as no prior knowledge, no cooperative sign, and no communication. However, necessary information on the noncooperative space target must be acquired before capturing. Obtaining the attitude, angular velocity, and inertia parameters of targets without contacting them is becoming a prominent technology for detecting small celestial bodies. This work proposes a new method for estimating the attitude, motion, and inertia properties of an unknown noncooperative space object. Simulations are presented to verify the method for targets with different properties and states. Moreover, the accuracy of the ratios of the principal inertia parameters can reach 90%. In some cases, the most important step is that the computer vision algorithm must assure good and precise information about the feature points on the target. In summary, this paper presents a promising and valuable method for identifying the motion and parameters of noncooperative space targets before the landing of small celestial bodies. The proposed method is highly efficient and suitable for spacecraft applications.