让机器人在几秒内适应突发地形变化，实现实时自主导航。
Zero to Autonomy in Real-Time: Online Adaptation of Dynamics in Unstructured Environments
- 用函数编码器+递归最小二乘法在线更新模型参数，无需反向传播。
- 仅需数秒数据即可完成适配，计算开销恒定，支持实时响应。
- 适用于野外、冰面等复杂环境，显著降低导航碰撞率。
自主机器人必须在无先验知识的情况下，于数秒内实现安全控制，以应对非结构化环境。突发地形变化（如突然进入结冰路面）会导致动力学特性突变，若模型无法实时适应，将导致规划系统失稳。本文提出一种在线自适应方法，结合函数编码器与递归最小二乘法，将函数编码器的系数视为可更新的隐状态，基于流式里程计数据进行迭代更新。该方法实现常数时间系数估计，无需梯度驱动的内层优化，可在仅数秒数据下完成有效适配。我们在范德波尔系统上验证算法行为，在Unity仿真中评估高保真越野导航性能，并在Clearpath Jackal机器人上测试，包括在本地冰场挑战性地形中的表现。实验表明，该方法在多种场景下均提升模型精度与下游规划能力，相比静态模型和元学习基线显著减少碰撞事件。
原文摘要 · Abstract (English)
Autonomous robots must go from zero prior knowledge to safe control within seconds to operate in unstructured environments. Abrupt terrain changes, such as a sudden transition to ice, create dynamics shifts that can destabilize planners unless the model adapts in real-time. We present a method for online adaptation that combines function encoders with recursive least squares, treating the function encoder coefficients as latent states updated from streaming odometry. This yields constant-time coefficient estimation without gradient-based inner-loop updates, enabling adaptation from only a few seconds of data. We evaluate our approach on a Van der Pol system to highlight algorithmic behavior, in a Unity simulator for high-fidelity off-road navigation, and on a Clearpath Jackal robot, including on a challenging terrain at a local ice rink. Across these settings, our method improves model accuracy and downstream planning, reducing collisions compared to static and meta-learning baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。