Safe Reinforcement Learning on Autonomous Vehicles

TL;DR

This paper explores how prediction-based methods can enable safe reinforcement learning for autonomous vehicles, specifically focusing on intersection handling, addressing real-world stochasticity and high-dimensional challenges.

Contribution

It introduces a prediction framework to constrain exploration in safe reinforcement learning, applicable to complex autonomous vehicle tasks.

Findings

Prediction constrains exploration effectively.

Safe intersection handling demonstrated on autonomous vehicle.

Addresses stochasticity and high-dimensionality in real-world systems.

Abstract

There have been numerous advances in reinforcement learning, but the typically unconstrained exploration of the learning process prevents the adoption of these methods in many safety critical applications. Recent work in safe reinforcement learning uses idealized models to achieve their guarantees, but these models do not easily accommodate the stochasticity or high-dimensionality of real world systems. We investigate how prediction provides a general and intuitive framework to constraint exploration, and show how it can be used to safely learn intersection handling behaviors on an autonomous vehicle.