首页 >

近距离巡检空间翻滚目标的航天器安全关键控制

张小翔1,耿云海2,吴宝林1,丁学良3   

  1. 1. 哈尔滨工业大学
    2. 哈尔滨工业大学卫星技术研究所
    3. 中国航空工业集团公司成都飞机设计研究所
  • 收稿日期:2026-02-04 修回日期:2026-05-29 出版日期:2026-06-01 发布日期:2026-06-01
  • 通讯作者: 吴宝林
  • 基金资助:
    高端重载机器人全国重点实验室开放基金

Safety-Critical Control for Spacecraft Proximity Inspection of a Tumbling Target

  • Received:2026-02-04 Revised:2026-05-29 Online:2026-06-01 Published:2026-06-01
  • Contact: Baolin WU

摘要: 针对复杂结构外形的空间翻滚目标近距离巡检任务,考虑执行机构饱和与位姿耦合下的碰撞规避,基于控制障碍函数方法提出了一种近似最优安全关键控制策略。首先,在李群SE(3)上构建了航天器位姿一体化跟踪误差动力学模型,并构造含控制输入饱和约束的值函数,在自适应动态规划框架下得到位姿一体化近似最优标称控制策略。进一步,针对复杂障碍边界附近易出现的期望状态不可达的问题,设计排斥引导策略作为候选标称控制策略,从而降低原标称控制的保守性。然后,针对目标翻滚诱发的位姿强耦合使安全约束随姿态演化而时变、以及复杂几何结构放大近场碰撞风险的问题,基于相对位姿与几何特征构建代数安全指标函数并形成避碰约束。最后,基于备份控制障碍函数与二次规划构建的安全关键控制框架,在满足避碰约束与输入饱和约束的同时,对标称控制进行最小修正以实现安全跟踪控制。数值仿真验证了所提方法的有效性。

关键词: 在轨维护, 自适应动态规划, 碰撞规避, 控制障碍函数, 李群SE(3)

Abstract: For a close-proximity inspection mission of a tumbling space target with a complex geometry, considering collision avoidance under actuator saturation and coupled position–attitude dynamics, an approximately optimal safety-critical control strategy is proposed based on the control barrier function (CBF) methodology. First, an integrated pose tracking-error dynamics model is formulated on the Lie group SE(3), and a value function incorporating control-input saturation constraints is constructed; an integrated pose-level approximately optimal nominal control policy is then obtained within an adaptive dynamic programming (ADP) framework. Next, to address the issues that target tumbling induces strong pose coupling, making safety constraints time-varying with attitude evolution, and that complex geometry amplifies near-field collision risk, an algebraic safety index function is designed based on relative pose and geometric features to yield collision-avoidance constraints. Furthermore, a repulsion-guidance strategy is developed as a candidate nominal control policy to mitigate reachability degradation near complex obstacle boundaries, thereby reducing the conservatism of the original nominal controller. Finally, a safety-critical control framework combining a backup control barrier function and quadratic programming is established to modify the nominal control input, achieving safe tracking control under motion constraints and input saturation. Numerical simulations validate the effectiveness of the proposed method.

Key words: On-orbit maintenance, Adaptive dynamic programming, Collision avoidance, Control barrier function, Lie group SE(3)

中图分类号: