Robust Regression for Safe Exploration in Control
This work addresses the challenge of safe exploration for control systems, particularly in scenarios with uncertain environments, offering a method that is more robust than prior approaches but is incremental in its advancement.
The paper tackles the problem of safe learning and exploration in sequential control by introducing a deep robust regression model to predict uncertainty bounds, enabling provably-safe actions for data collection and improved controller safety and optimality. Empirically, it outperforms Gaussian process-based methods in settings where specifying a good prior is difficult.
We study the problem of safe learning and exploration in sequential control problems. The goal is to safely collect data samples from operating in an environment, in order to learn to achieve a challenging control goal (e.g., an agile maneuver close to a boundary). A central challenge in this setting is how to quantify uncertainty in order to choose provably-safe actions that allow us to collect informative data and reduce uncertainty, thereby achieving both improved controller safety and optimality. To address this challenge, we present a deep robust regression model that is trained to directly predict the uncertainty bounds for safe exploration. We derive generalization bounds for learning, and connect them with safety and stability bounds in control. We demonstrate empirically that our robust regression approach can outperform the conventional Gaussian process (GP) based safe exploration in settings where it is difficult to specify a good GP prior.