发表机构
NOVA School of Science and Technology; University of Oxford（新里斯本大学科学与技术学院; 牛津大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对延迟反馈下的鲁棒控制难题，提出受小脑启发的多路复用预测与内部反馈框架，实现快速在线修正并将学习时间减少一个数量级。
AI 中文摘要
在延迟感觉反馈下进行鲁棒控制，仍然是机器人和神经科学中的关键挑战。经典小脑模型通过前向预测解释延迟补偿，但无法解释生物系统中观察到的快速在线修正和快速适应。我们提出了一种受小脑启发的控制框架，该框架将多路复用预测表征与内部反馈相结合。通过联合编码运动学变量和任务相关误差信号，该模型能够在延迟反馈下实现准确的在线修正。此外，在小脑环路内纳入反馈显著加速了适应过程，将学习时间减少了一个数量级。我们的结果表明，在延迟条件下，单一信号预测是不充分的，而多路复用与反馈共同为在线控制和快速学习提供了统一机制。
英文摘要
Robust control under delayed sensory feedback remains a key challenge in both robotics and neuroscience. Classical cerebellar models explain delay compensation through forward prediction but fail to account for fast online corrections and rapid adaptation observed in biological systems. We propose a cerebellum-inspired control framework that combines multiplexed predictive representations with internal feedback. By jointly encoding kinematic variables and task-relevant error signals, the model enables accurate online correction despite delayed feedback. Furthermore, incorporating feedback within the cerebellar loop significantly accelerates adaptation, reducing learning time by an order of magnitude. Our results show that single-signal predictions are insufficient under delay, while multiplexing and feedback together provide a unified mechanism for online control and rapid learning.
