# 用于非合作目标端到端捕获与消旋的相位门控约束强化学习

> 原文链接：https://robotics.hellobike.com/papers/2026-10-11-2610-15491
> 频道：papers | 更新日期：2026-10-11 | 标签：arxiv, cs.RO
> 核心摘要：针对非合作翻滚目标捕获与消旋，提出带平滑接近门的约束强化学习单策略，并在随机试验中验证。

# 用于非合作目标端到端捕获与消旋的相位门控约束强化学习

::: v-pre
> 原文链接：[Phase-Gated Constrained Reinforcement Learning for End-to-End Capture and Detumbling of Non-Cooperative Targets](https://arxiv.org/abs/2610.15491)（arXiv 2610.15491 · 发布日期：2026-10-11）

**Phase-Gated Constrained Reinforcement Learning for End-to-End Capture and Detumbling of Non-Cooperative Targets**

针对非合作翻滚目标捕获与消旋，提出带平滑接近门的约束强化学习单策略，并在随机试验中验证。

## 核心要点

- 传统分阶段控制依赖惯量、抓捕接口或手工切换，接触衔接处脆弱。
- 在约束MDP的奖励、动作空间和约束集中加入平滑接近门。
- 单策略完成接近、无接口摩擦捕获和消旋，适应未知质量惯量与噪声估计。
- 1000余次随机试验成功率99.5%，且保持接触力、执行器等限制。

**验证范围与局限**：验证范围为1000余次随机试验，摘要未说明是否包含真机或在轨实验。

## 英文摘要 (Abstract)

Active debris removal requires a servicer spacecraft to approach a non-cooperative tumbling object, capture it, and arrest its rotation. This sequence is conventionally divided among phase-specific controllers, each depending on knowledge of the target's inertia, a dedicated grappling fixture, or a hand-designed switch at the contact-rich seam where such pipelines are most fragile. We show that a single constrained reinforcement learning policy can execute the entire mission, driving a dual-arm free-flying servicer from free-space closing through friction-based fixture-free capture to full detumbling of a target whose mass and inertia are unknown and whose state is observed only through noisy estimates. The enabling mechanism is a smooth proximity gate applied at three loci of a constrained Markov decision process. In the reward it induces the phase structure rather than imposing it; in the action space it forecloses a degenerate grab-and-pull exploit as a hard guarantee; and in the constraint set it resolves a normalization failure of constraint-based learning, in which a constraint spanning phases that load it unequally is rendered nearly inert in the phase where precision matters most. Both the failure mode and the remedy apply to any multi-phase problem of this kind. Over 1000 randomized trials the policy succeeds in 99.5% while holding its contact-force, actuator, base-reaction and collision limits throughout, transfers zero-shot to unseen geometries and centre-of-mass offsets, and extrapolates across a tenfold range of target mass.

作者：Álvaro Belmonte-Baeza, Celia Redondo-Verdú, José L. Ramón, Miguel Cazorla, Leonard Felicetti, Jorge Pomares
:::

## 原始参考来源 (Sources)

- [arXiv 2610.15491：Phase-Gated Constrained Reinforcement Learning for End-to-End Capture and Detumbling of Non-Cooperative Targets](https://arxiv.org/abs/2610.15491)