# iLQR-MPC全身控制实现宇树G1人形机器人在线鲁棒行走

> 原文链接：https://robotics.hellobike.com/papers/2026-11-04-2611-04204-e0a4dc77
> 频道：papers | 更新日期：2026-11-04 | 标签：auto, arxiv
> 核心摘要：提出基于单刚体模型的iLQR-MPC算法，联合优化接触力与落脚点。Unitree G1在MuJoCo中以100Hz重规划、8.9ms求解耗时，成功抵御245N前向及175N侧向冲击，无需特定策略训练。

# iLQR-MPC全身控制实现宇树G1人形机器人在线鲁棒行走

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[Online iLQR-MPC with Whole-Body Control for Robust Humanoid Locomotion](https://arxiv.org/abs/2611.04204)（arXiv 2611.04204）

## 核心内容

针对人形机器人步态规划中扰动响应滞后问题，研究提出一种在线iLQR-MPC与全身控制（WBC）结合框架。该方法摒弃独立步态规划器，直接在单次滚动时域优化中联合求解接触力与落脚位置。在MuJoCo仿真验证中，Unitree G1实现了稳态行走，并能从245N前向和175N侧向脉冲推力中恢复稳定，展现出强鲁棒性。

技术核心在于采用保留姿态动力学与全接触 wrench 的单刚体模型（SRBM）。通过增广拉格朗日与松弛障碍惩罚，将接触及运动约束嵌入iLQR反向传播过程。底层WBC以500Hz频率将优化轨迹映射为关节力矩并保证接触可行性。MPC层重规划频率达100Hz，平均求解耗时仅8.9ms。该纯模型驱动方法具备可解释性，无需针对特定平台进行强化学习策略训练。

## 关键看点

- **联合优化架构**：iLQR-MPC在单一滚动时域内同时优化接触力分布与落脚点位置，取消独立步态规划器模块
- **实时求解性能**：MPC重规划频率达100Hz，平均计算耗时8.9ms；全身控制器以500Hz运行确保接触约束可行
- **抗扰实测数据**：Unitree G1在仿真中承受245N前向及175N侧向脉冲推力后仍能维持稳态行走与平衡恢复
- **无训练泛化**：基于显式成本与约束的模型驱动公式，无需平台特定策略训练即可实现可解释的扰动响应

## 工程与产业观察

该工作证明了轻量级模型预测控制在人形机器人高动态场景下的实时可行性。8.9ms的求解耗时为端侧部署留出充裕算力余量，且免RL训练特性显著降低了Sim-to-Real迁移门槛。对于追求高可靠性与可解释性的工业级人形机器人落地，这种经典控制与现代优化结合的范式提供了区别于端到端大模型的重要技术参照。

## 英文摘要 (Abstract)

Humanoid locomotion requires deciding where to step and how to distribute contact forces. However, a disturbance can invalidate the nominal plan, requiring coordinated revision under constraints that switch with every footfall. To solve this online, the proposed approach employs the single-rigid-body model, which retains attitude dynamics and the full contact wrench. An iterative linear-quadratic-regulator MPC (iLQR-MPC) then jointly optimizes contact forces and footstep locations in a single receding-horizon problem, without a separate step planner. Specifically, contact and motion constraints are incorporated into the backward pass through augmented-Lagrangian and relaxed-barrier penalties. A whole-body controller subsequently maps the plan to joint torques at 500 Hz while preserving contact feasibility. The model-based formulation makes disturbance responses interpretable through explicit costs and constraints, and requires no platform-specific policy training. In MuJoCo simulations, the iLQR-MPC replans at 100 Hz (mean solve time 8.9 ms); the Unitree G1 sustains steady-state walking and recovers from impulsive 245 N forward and 175 N lateral pushes.

作者：Xiang Li, Linlin Ou, Ruihan Wu, Xinyi Yu
:::

## 原始参考来源 (Sources)

- [arXiv 2611.04204：Online iLQR-MPC with Whole-Body Control for Robust Humanoid Locomotion](https://arxiv.org/abs/2611.04204)