# 宇树Go2热感知策略：无温感输入实现续航从1.1km增至10km

> 原文链接：https://robotics.hellobike.com/papers/2026-11-18-2611-22231-c16c1d79
> 频道：papers | 更新日期：2026-11-18 | 标签：auto, arxiv
> 核心摘要：研究提出无需温度反馈的强化学习热感知运动策略，在宇树Go2上验证。5kg负载1m/s工况下，持续运动距离从基线1.1km提升至10.0km，运行时间由19分钟延至174分钟，成功将足式机器人瓶颈从热关机转为电池耗尽。

# 宇树Go2热感知策略：无温感输入实现续航从1.1km增至10km

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[From Temperature-Limited to Battery-Limited Legged Locomotion](https://arxiv.org/abs/2611.22231)（arXiv 2611.22231）

## 核心内容

针对足式机器人持续运动常因关节电机过热而非电量耗尽停机的痛点，研究者提出一种热感知强化学习运动策略。该方法在宇树Go2四足机器人上实测验证，在5kg负载、1m/s速度工况下，将持续运动距离从基线的1.1km（19分钟热关机）大幅提升至10.0km（174分钟电池耗尽），最高电机温度低于关机阈值14°C，彻底改变了系统限制因素。

技术核心在于构建13节点集总RC热网络模型，仅凭电功率校准即可预测12个电机温度，跨实例RMSE低于1°C。针对热时间常数远超RL信用分配周期的难题，设计基于步态平均平方扭矩与关节速度的渐近温度奖励函数。该奖励使策略能在单步态周期内响应热趋势，无需实时温度输入即可学会降低并分散热负荷，在含楼梯斜坡的工厂场景中亦将续航从6.2km提升至11.2km。

## 关键看点

- **热网络建模**：构建13节点集总RC热网络，仅用电功率校准即可以低于1°C RMSE预测12个电机温度，适配不同负载与步态
- **奖励函数设计**：利用步态平均平方扭矩与关节速度计算渐近温度作为奖励，解决热时间常数远超RL信用分配周期的训练难题
- **重载续航突破**：5kg负载1m/s工况下，持续运动时间从19分钟延至174分钟，距离由1.1km增至10.0km，消除热停机瓶颈
- **复杂场景验证**：在无负载工厂楼梯斜坡回路测试中，单次电池续航圈数从25圈增至47圈，里程达11.2km且保留22°C热裕度

## 工程与产业观察

该工作证明了在不增加硬件传感器前提下，通过机理模型与RL奖励 shaping 结合可显著解锁本体性能上限。其基于电学量推断热状态的方法为具身智能体在算力与传感受限条件下的长时可靠运行提供了低成本工程范式，对工业巡检与物流机器人的连续作业能力具有直接参考价值。

## 英文摘要 (Abstract)

On legged robots, sustained locomotion often ends in thermal shutdown rather than battery depletion: the heat of walking concentrates in a few joint motors, which overheat long before the battery is empty. We train thermally aware locomotion policies by reinforcement learning without ever observing temperature, and demonstrate this on a Unitree Go2 quadruped. We first identify a coupled 13-node lumped RC thermal network, fitted on two runs of a single robot and calibrated against the measured electrical power, that predicts all twelve motor temperatures with a mean RMSE below 1 $^\circ$C across different robot instances, gaits, payloads, and speeds. Under a sustained gait, the temperatures the motors will settle at follow directly from the gait-averaged squared torques and joint speeds. Our core contribution is a reward that drives down the hottest predicted asymptotes: thermal time constants exceed the credit-assignment horizon of reinforcement learning by orders of magnitude, but predicted asymptotes respond to torque and joint speed within a single gait cycle, so the policy learns to lower and spread heat with no temperature input. On the robot, the policy turns sustained locomotion from temperature-limited to battery-limited: with a 5 kg payload at 1 m/s, a learned baseline reaches thermal shutdown after 19 minutes (1.1 km), whereas our policy runs the battery down after 174 minutes (10.0 km) with its hottest motor 14 $^\circ$C below the shutdown threshold in a 26 $^\circ$C room; on an unloaded factory circuit with stairs and ramps, the baseline overheats after 25 laps (6.2 km), while our policy completes 47 (11.2 km) on one battery with 22 $^\circ$C of margin.

作者：Rapha\"el S\'egas, Bal\'azs Opra, L\'eo Ghafari, Thomas Stewart
:::

## 原始参考来源 (Sources)

- [arXiv 2611.22231：From Temperature-Limited to Battery-Limited Legged Locomotion](https://arxiv.org/abs/2611.22231)