ARCADE: 基于在线变化点感知的自适应机器人控制
ARCADE: Adaptive Robot Control with Online Changepoint-Aware Bayesian Dynamics Learning
AI总结：
ARCADE通过在线变化点感知机制实现机器人动态的自适应控制，提升预测精度和恢复能力。
摘要：
现实中的机器人必须在操作条件变化、外部干扰和未建模效应导致的动态演变下运行。这些可能表现为渐进漂移、瞬时波动或突变转移，要求实时适应，既要鲁棒于短期变化，又要有对长期变化的响应能力。我们提出了一种用于建模机器人系统非线性动态的框架，能够从流式数据中实时更新。该方法将表示学习与在线适应解耦，利用离线学习的潜在表示来支持在线的闭式贝叶斯更新。为处理演变条件，我们引入了变化点感知机制，通过从数据似然中推断出的潜在变量来指示连续性或转移。当连续性可能性较高时，证据积累以细化预测；当检测到转移时，过去信息会被削弱以实现快速重新学习。这保持了校准的不确定性，并支持对瞬时、渐进或结构性变化的概率推理。我们证明该框架的自适应遗憾仅随时间增长对数级，并与转移次数线性增长，与知晓转移时间的Oracle相当。我们在cartpole模拟和具有摆动载荷和飞行中掉落的真实四旋翼飞行中进行了验证，显示预测精度、恢复速度和闭环跟踪准确性优于相关基线。
英文摘要：
Real-world robots must operate under evolving dynamics caused by changing operating conditions, external disturbances, and unmodeled effects. These may appear as gradual drifts, transient fluctuations, or abrupt shifts, demanding real-time adaptation that is robust to short-term variation yet responsive to lasting change. We propose a framework for modeling the nonlinear dynamics of robotic systems that can be updated in real time from streaming data. The method decouples representation learning from online adaptation, using latent representations learned offline to support online closed-form Bayesian updates. To handle evolving conditions, we introduce a changepoint-aware mechanism with a latent variable inferred from data likelihoods that indicates continuity or shift. When continuity is likely, evidence accumulates to refine predictions; when a shift is detected, past information is tempered to enable rapid re-learning. This maintains calibrated uncertainty and supports probabilistic reasoning about transient, gradual, or structural change. We prove that the adaptive regret of the framework grows only logarithmically in time and linearly with the number of shifts, competitive with an oracle that knows timings of shift. We validate on cartpole simulations and real quadrotor flights with swinging payloads and mid-flight drops, showing improved predictive accuracy, faster recovery, and more accurate closed-loop tracking than relevant baselines.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
