Collision Risk-Perceptive Landing Control for Small-Body Missions

In lack of information with regard to the diverse topology on the small-body surface, operation safety of the spacecraft can be hardly guaranteed in advance. To improve onboard environment adaptability in future missions, an agile landing control that reacts to real-time environment interactions is investigated. An online collision prediction model is established on the basis of environment sensing and state uncertainty propagation. By searching the minimal residual space margin and the most dangerous direction on the finite horizon, a dynamic environment feedback is formulated. The feedback information is incorporated into model-predictive control, upon which a collision risk-perceptive landing control is obtained. Compared with the existing methods, the proposed method is independent of a priori knowledge of the environment. It is capable of adjusting control objectives according to onboard environment perceptions. Recursive feasibility and stability of the controller are proved. Its effectiveness is demonstrated in a 433-Eros-based landing simulation.