Vision-Based Autonomous Car Racing Using Deep Imitative Reinforcement Learning

TL;DR

This paper introduces a deep imitative reinforcement learning approach for autonomous car racing that combines imitation learning and model-based reinforcement learning to improve efficiency and performance using visual inputs.

Contribution

The work presents a novel DIRL method that learns from both human demonstrations and self-improvement via offline world models, enhancing autonomous racing capabilities.

Findings

Outperforms previous IL and RL methods in simulation and real-world tests.

Achieves high sample efficiency and task performance.

Successfully operates on a real RC-car with limited computation.

Abstract

Autonomous car racing is a challenging task in the robotic control area. Traditional modular methods require accurate mapping, localization and planning, which makes them computationally inefficient and sensitive to environmental changes. Recently, deep-learning-based end-to-end systems have shown promising results for autonomous driving/racing. However, they are commonly implemented by supervised imitation learning (IL), which suffers from the distribution mismatch problem, or by reinforcement learning (RL), which requires a huge amount of risky interaction data. In this work, we present a general deep imitative reinforcement learning approach (DIRL), which successfully achieves agile autonomous racing using visual inputs. The driving knowledge is acquired from both IL and model-based RL, where the agent can learn from human teachers as well as perform self-improvement by safely interacting with an offline world model. We validate our algorithm both in a high-fidelity driving simulation and on a real-world 1/20-scale RC-car with limited onboard computation. The evaluation results demonstrate that our method outperforms previous IL and RL methods in terms of sample efficiency and task performance. Demonstration videos are available at https://caipeide.github.io/autorace-dirl/