Reinforced Iterative Learning control improves robot arm tracking accuracy
Reinforced Iterative Learning Control for elastic joint arms: Guaranteed cross-trajectory generalization under model uncertainties
Robotics
Summary
Robot arms that bend and flex are hard to control because they don't always move exactly as expected. The authors combined two learning methods to make the arms follow planned movements more accurately and adapt faster. One method learns from repeated tries, while the other generalizes to new movements. Their combined approach lets the robot perform well from the first try and improves over time, even when the robot's model isn’t perfect. They tested this on a robot with flexible joints and showed it works for different tasks.
What this means in practice
- •For industrial robot integrators: Improve precise motion control of flexible joint robots for diverse tasks without separate tuning for each movement.
- •For robotics simulation engineers: Test advanced control algorithms that adapt quickly and generalize across tasks on simulated elastic joint robot models.
Authors
Yuri De Santis, Michele Pierallini, Franco Angelini, Manolo Garabini
Abstract
Trajectory tracking for compliant robots is challenging due to nonlinear dynamics and modeling uncertainties, often requiring learning-based methods to avoid explicit parameter identification. Iterative Learning Control (ILC) ensures theoretically high precision and preserves the passive compliance of the robot. However, traditional ILC requires a separate learning phase for each new trajectory. Conversely, Reinforcement Learning (RL) generalizes to new trajectories, but practically suffers from high sample complexity and limited tracking precision. This work proposes Reinforced Iterative Learning Control (RILC) that combines the strengths of both approaches to achieve fast adaptation and guaranteed tracking convergence across diverse trajectories. By establishing a concurrent coupling between the two controllers, RILC overcomes the high initial errors and slow adaptation of traditional ILC, enabling high-performance tracking from the very first trials. In this architecture, ILC provides high-quality samples, while RL contributes as a generalizing best-effort policy across iterations. Additionally, a theoretical proof is presented to show that the RL component preserves ILC stability and, under specific conditions, can also reduce the error bound asymptotically. The efficacy of this framework is demonstrated through simulation and experiments on an underactuated 4-degree-of-freedom serial elastic robot using different trajectories and modeling mismatches.