# 竞速以恢复：通过学习自动赛车为自动驾驶设计先进安全系统

> 原文链接：https://robotics.hellobike.com/papers/2026-10-14-2610-20873
> 频道：papers | 更新日期：2026-10-14 | 标签：arxiv, cs.RO
> 核心摘要：针对失控车辆恢复问题，先用强化学习训练漂移竞速策略再迁移恢复，在Gym-Khana、294个初始状态及1:10 F1TENTH实车上验证。

# 竞速以恢复：通过学习自动赛车为自动驾驶设计先进安全系统

::: v-pre
> 原文链接：[Race to Recovery: Designing Advanced Safety Systems for Autonomous Driving via Learning Autonomous Racing](https://arxiv.org/abs/2610.20873)（arXiv 2610.20873 · 发布日期：2026-10-14）

**Race to Recovery: Designing Advanced Safety Systems for Autonomous Driving via Learning Autonomous Racing**

针对失控车辆恢复问题，先用强化学习训练漂移竞速策略再迁移恢复，在Gym-Khana、294个初始状态及1:10 F1TENTH实车上验证。

## 核心要点

- 结冰、滑水或紧急避让会使车辆越过操稳极限，现有量产安全系统偏重预防而非恢复。
- 方法先以强化学习学习时间最优漂移竞速策略，再用最短时间奖励继续训练以适配恢复。
- 恢复状态相对于基于相平面的稳定平衡定义，并提出Gym-Khana用于控制器训练与评估。
- 在294个初始失控状态中，迁移策略较Stanley提升186.8%，并优于非线性模型预测控制器和从零训练策略。

**验证范围与局限**：验证范围包括Gym-Khana中294个初始失控状态，以及一辆1:10尺度F1TENTH实车上的已测试失控状态。

## 英文摘要 (Abstract)

Vehicles that lose control at the handling limit through icing, hydroplaning, or an abrupt evasive maneuver fall outside the operating envelope of production safety systems such as electronic stability control, which focus on preventing instability rather than recovering from it. Expert drivers regain authority in these conditions by drifting. Existing autonomous recovery approaches are validated on idealized scenarios, define "recovered" states empirically and inconsistently, or rely on offline-planned trajectories rather than acting online. We address these gaps by first training a racing policy with reinforcement learning that learns drifting as a time-optimal maneuver, then adapting it to recovery through continued training with a minimum-time reward. We define recovery relative to a phase-plane-grounded stable equilibrium and introduce Gym-Khana for controller training and evaluation. Across 294 initial loss-of-control states, the transferred policy improves recovery by 186.8% over a Stanley path tracker and outperforms both a nonlinear model predictive controller and a recovery-only policy trained from scratch, while also recovering faster on states the baseline can already handle. An ablation shows that racing pretraining, rather than broader initial-state exposure alone, drives this gain. Deployed on a physical 1:10-scale F1TENTH vehicle using system identification, actuator-lag modelling, and domain randomization, the transferred policy outperforms Stanley on every tested loss-of-control state, including recovering 50% of the time from a hard state the baseline never recovers from.

作者：Teodor Ilie, Kléber Cabral, Sidney Nascimento Givigi
:::

## 原始参考来源 (Sources)

- [arXiv 2610.20873：Race to Recovery: Designing Advanced Safety Systems for Autonomous Driving via Learning Autonomous Racing](https://arxiv.org/abs/2610.20873)