# 特权Critic训练使端到端RL中的无传感器推进器故障适应成为可能

> 原文链接：https://robotics.hellobike.com/papers/2026-08-24-2608-22976
> 频道：papers | 更新日期：2026-08-24 | 标签：arxiv, cs.RO
> 核心摘要：针对推进器故障难观测且缺少专用传感器的问题，提出RAFT用特权非对称Critic和循环记忆训练，并在浮动平台机器人上验证。

# 特权Critic训练使端到端RL中的无传感器推进器故障适应成为可能

::: v-pre
> 原文链接：[Privileged Critic Training Enables Sensor-Free Thruster Fault Adaptation in End-to-End RL](https://arxiv.org/abs/2608.22976)（arXiv 2608.22976 · 发布日期：2026-08-24）

**Privileged Critic Training Enables Sensor-Free Thruster Fault Adaptation in End-to-End RL**

针对推进器故障难观测且缺少专用传感器的问题，提出RAFT用特权非对称Critic和循环记忆训练，并在浮动平台机器人上验证。

## 核心要点

- 推进器可能连续退化、失效或卡在开启，传统检测需部署时不可用的专用传感器。
- RAFT在训练中让PPO价值函数访问真实退化状态，Actor仅接收标准任务观测。
- 在最多四个同时推进器故障下，RAFT四故障成功率70.2%，接近部署时可见完整退化状态的oracle策略。
- 论文称所有代码、检查点和数据均已开源。

**验证范围与局限**：验证范围为一台浮动平台机器人（8个推进器、1个反作用轮）上最多四个同时推进器故障。

## 英文摘要 (Abstract)

Fault-tolerant navigation for thruster-actuated robots requires online adaptation to failures that are neither binary nor fully observable: thrusters may degrade continuously, fail dead, or jam stuck-open. Classical fault detection pipelines require dedicated sensors unavailable at deployment; oracle controllers that observe the true failure state are equally impractical. We show that privileged critic training is sufficient for sensor-free fault adaptation: giving the PPO value function access to the true degradation state dgt during training, while the actor receives only standard task observations, shapes a policy that compensates for failures at deployment without any dedicated fault sensing. We propose RAFT (Recurrent Asymmetric Fault Tolerant), a policy with recurrent memory trained with a privileged asymmetric critic. Evaluated on a floating-platform robot (8 thrusters, 1 reaction wheel) under up to four simultaneous thruster failures, RAFT achieves 70.2% success at four concurrent failures, closing 84% of the gap from a failure-naive baseline (4.8%) to an oracle policy that sees the full degradation state at deployment (82.4%). All code, checkpoints, and data are open-source.

作者：Ricard Marsal I Castan, Miguel A. Olivares-Méndez
:::

## 原始参考来源 (Sources)

- [arXiv 2608.22976：Privileged Critic Training Enables Sensor-Free Thruster Fault Adaptation in End-to-End RL](https://arxiv.org/abs/2608.22976)