A Bayesian Approach to Reinforcement Learning of Vision-Based Vehicular Control

TL;DR

This paper introduces a Bayesian reinforcement learning method for autonomous vehicle control using vision data, demonstrating improved performance and efficiency in a simulated urban environment.

Contribution

It presents a novel Bayesian temporal difference learning approach for vision-based vehicle control, with comprehensive evaluation in a realistic simulator.

Findings

Training on ground truth segmentation improves performance.

The proposed method reduces training time compared to existing approaches.

System outperforms competitors on the CARLA benchmark.

Abstract

In this paper, we present a state-of-the-art reinforcement learning method for autonomous driving. Our approach employs temporal difference learning in a Bayesian framework to learn vehicle control signals from sensor data. The agent has access to images from a forward facing camera, which are preprocessed to generate semantic segmentation maps. We trained our system using both ground truth and estimated semantic segmentation input. Based on our observations from a large set of experiments, we conclude that training the system on ground truth input data leads to better performance than training the system on estimated input even if estimated input is used for evaluation. The system is trained and evaluated in a realistic simulated urban environment using the CARLA simulator. The simulator also contains a benchmark that allows for comparing to other systems and methods. The required training time of the system is shown to be lower and the performance on the benchmark superior to competing approaches.