Adaptive Control For Approaching And Tracking The Tumbling Target

Since 1957,a large number of spacecraft have been launched into space orbit each year.Many of them have become disabled due to over-service or equipment failure.These dead-spacecrafts are unable to maintain a stable attitude and orbit situation due to loss of control,and are often free to tumble and drift under the influence of space regulating moments and various disturbances.The dead-spacecrafts not only occupy the little orbital space left,but also pose a serious threat to the safety of other spacecrafts.Therefore,the technology of on-orbit servicing has emerged and gradually become a popular research field for major spacefaring nations.In this paper,we take a runaway satellite tumbling in space as the target and a tracking satellite as the control object,and focus on the final approaching section of the capture mission in the context of the space capture mission-approaching and tracking to 1m from a relative distance of 5m between the surfaces of the two satellites.Considering that the relative distance is within 2m,the relative visual navigation device installed on the tracking satellite can only take partial pictures of the target satellite and thus cannot obtain accurate attitude and orbit information,it is necessary to change the control method.Therefore,the approaching and tracking task within 5m of the surface between the two satellites is divided into two sub-stages: the first stage is the close approaching and tracking(5m-2m),in which the relative visual navigation device can obtain the target attitude orbit information normally;the second stage is the ultra-close approaching and tracking(2m-1m),in which the camera field of view cannot cover the main body of the target satellite and needs to be replaced by the visual servoing algorithm.First,the orbital and attitude motion models of the tracking satellite and the target satellite are established,followed by the relative attitude motion model between the two satellites.Since the tracking satellite needs to keep synchronous tracking with the captured plane of the target satellite(set by the mission),a relative attitude orbit coupling motion model is needed.Finally,a deviation model is built based on this coupled motion model for subsequent control.Subsequently,based on the above deviation model,a close(5m-2m)approaching and tracking control law is designed.Considering the external disturbances and modeling uncertainties widely existing in space,an adaptive sliding mode control law with a nonlinear disturbance observer is selected,which is able to estimate and compensate for the disturbance terms.It is verified by simulation that the method is able to achieve the close-approaching and tracking tasks under the preset time and accuracy requirements.Finally,the two-satellite imaging motion model is established using the small-aperture imaging principle in the ultra-close range(2m-1m)approaching and tracking phase.The image based visual servoing method was used to establish the position and velocity relationship between the visual feature points and the camera core,and binocular stereo vision was used to obtain the relative distance between the two satellites.The orbit control at ultra-close range(2m-1m)is achieved by tracking the central feature point,while attitude control is achieved using the relative geometric relationship between several visual feature points.Based on the above relative relationships,a fuzzy parameter self-tuning PD control law is designed.The simulation verifies that it can achieve stable approaching and tracking of the tumbling target and meet the required accuracy requirements of the task.