The research article delves into autonomous driving with a particular focus on tractor-trailer vehicles. The researchers developed truck and trailer models using a high-fidelity simulation software Carla, and several roundabout scenarios to establish a baseline dataset for benchmarks. Using a twin-q soft actor-critic algorithm, they trained an autonomous driving model that achieved a 73% success rate on different roundabouts. The paper emphasizes that autonomous vehicles can help to reduce road accidents significantly, lower emissions, increase mobility, and reduce congestion. The application of autonomous navigation to heavy goods vehicles can lead to substantial economic growth by increasing the efficiency of the logistics sector.
Publication date: 12 Jan 2024
Project Page: https://orcid.org/0000-0002-8274-6177
Paper: https://arxiv.org/pdf/2401.04980
