Robot control method unites model-based and learning approaches
End-to-end QP-based policies: A unified perspective on robust control and robot learning
Robotics
Summary
Controlling robots usually requires designing special rules based on how they work or teaching them entirely through trial and error. This paper introduces a new way to create robot controllers that combines both approaches in a single method, making it transparent and easier to adjust. Their method can automatically tune itself even when the robot faces different disturbances or when its exact inner workings are unknown. The authors tested this method both through simulations and on real robots, showing it can handle a wide range of tasks reliably.
What this means in practice
- •For robotics engineers: Create robot controllers that automatically adapt to hardware differences and external disturbances without detailed manual design.
- •For automation system integrators: Deploy control strategies for systems with unknown or changing dynamics by using a unified framework that blends model-based and learning methods.
Authors
Fausto Vega, Priyanka Supraja Balaji, Chase Dunaway, Joe Koszut, Jon Arrizabalaga, Zachary Manchester
Abstract
We present an end-to-end QP-based policy framework that enables systematic policy construction with minimal domain-specific design, while preserving the transparency and interpretability of model-based control. The proposed policy representation supports both domain-randomized model-based auto-tuning, where policy parameters are optimized over distributions of disturbances and model variations, and black-box policy construction, where the problem is formulated in terms of a (possibly) unknown model without requiring explicit notions of states, inputs, or the underlying system dynamics. We establish connections to existing policy representations and control paradigms, including robust control, multilayer perceptrons, and robot learning, and interpret our end-to-end QP policies as a common abstraction of these approaches. We validate the resulting framework in both simulation and hardware, demonstrating a broad range of applications spanning robustness, automatic policy tuning, and control under unknown system dynamics.