Cerebellar model improves motor control and learning with delayed feedback
Error- and Prediction-Driven Motor Learning in the Cortico-Cerebellar Loop
Machine Learning
Summary
Controlling movements is hard when sensory feedback is delayed, like how your brain sometimes reacts late to what your eyes see. The authors propose a brain-inspired method that predicts movement and errors together, helping to make quick corrections even with delayed information. Their model also uses feedback within the brain loop to learn much faster than before. This approach better mimics how actual brains handle movement and adaptation.
What this means in practice
- •For robotics engineers: Enable robots to correct movements accurately and quickly despite delayed sensor data using a cerebellum-inspired control system.
- •For prosthetics developers: Improve artificial limb control by integrating predictive and error-driven feedback mechanisms for real-time adjustment under signal delays.
Authors
Ana Carolina Filipe, Rui Ponte Costa, Cláudia Soares
Abstract
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.