可视性约束下多航天器跟踪空间目标的变焦图像视觉伺服控制  

Zooming image-based visual servoing for multiple spacecraft tracking control of space objects under visibility constraints

作　　者：林子杰 吴宝林[1,2] 王郸维 LIN ZiJie;WU BaoLin;WANG DanWei(Research Center of Satellite Technology,Harbin Institute of Technology,Harbin 150001,China;State Key Laboratory of Micro-Spacecraft Rapid Design and Intelligent Cluster,Harbin 150001,China;School of Electrical and Electronic Engineering,Nanyang Technological University,Singapore 639798,Singapore)

机构地区：[1]哈尔滨工业大学卫星技术研究所,哈尔滨150001 [2]微小型航天器快速设计与智能集群全国重点实验室,哈尔滨150001 [3]School of Electrical and Electronic Engineering,Nanyang Technological University,Singapore 639798,Singapore

出　　处：《中国科学:技术科学》2025年第2期322-338,共17页Scientia Sinica(Technologica)

基　　金：国家自然科学基金(批准号:62188101);黑龙江“头雁行动”项目资助

摘　　要：利用多个微纳卫星对空间目标进行在轨服务,具备成本低、覆盖面广、灵活性强、冗余度高和隐蔽性强的优势.然而,目标的非合作性以及航天器成员的增加为动力学控制带来挑战.本文针对多航天器跟踪非合作翻滚目标的相对位姿控制问题,提出一种新型的变焦图像视觉伺服控制方法.所设计的方法优势在于:(1)无需使用非合作目标的位姿信息,直接使用图像信息实现姿轨一体化控制,减少对目标状态估计的依赖.(2)变焦功能动态适应目标远近带来的尺寸变化,扩大了视觉伺服的工作范围.(3)能够同时满足两种特征可视性约束,一方面,特征始终以合适的尺寸维持在成像范围内;另一方面,防止特征被其他航天器遮挡.前者涉及单个航天器运动行为,后者涉及多个航天器运动耦合.首先,本文基于不变空间理论推导了新的航天器视觉伺服模型.然后,基于超螺旋干扰观测器设计固定时间鲁棒控制器.随后,设计了变焦策略满足视场约束和主动遮挡规避机动策略.基于Lyapunov方法证明了系统的稳定性与特征可视性,并通过Matlab和CoppeliaSim仿真验证了控制器的有效性.本文所设计的视觉控制方法能为多航天器复杂在轨服务任务提供技术支持.Multiple microsatellites in on-orbit servicing for space targets provide considerable advantages,including low cost,wide coverage,high flexibility,robust redundancy,and enhanced stealth capabilities.However,the noncooperative nature of the target and the increasing number of spacecraft members pose considerable challenges for dynamic control.This study proposes a novel image-based visual servo control scheme using a zooming camera designed to enable the tracking of a noncooperative tumbling target with multiple spacecraft.The proposed method offers the following advantages.(1)The proposed method eliminates the need for the pose estimation of the noncooperative target,directly leveraging image information for integrated attitude and orbit control,thereby reducing dependency on the estimated state of the target.(2)The zooming capability dynamically adapts to size variations caused by changes in distance,extending the operational range of the visual servoing system.(3)Two feature visibility constraints are simultaneously addressed:on the one hand,the features are kept within the imaging range with appropriate sizing;on the other hand,the occlusion of features by other spacecraft members is effectively prevented.The former involves the motion behavior of a single spacecraft,whereas the latter involves the coupled motion of multiple spacecraft.First,a novel spacecraft visual servoing model is developed based on invariant space theory.Afterward,a fixed-time robust controller is designed using a super-twisting disturbance observer.In addition,an innovative zooming strategy and an active occlusion avoidance maneuvering strategy are proposed.Through the Lyapunov analysis method,the stability of the closed-loop control system and the maintenance of feature visibility are rigorously proven.The effectiveness of the controller is demonstrated through simulations conducted in MATLAB and CoppeliaSim.The visual control method proposed in this study provides critical technical support for complex on-orbit servicing missions invol

关 键 词：航天器控制 非合作目标 可视性约束 图像视觉伺服 不变空间 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]