Abstract:<div class="section abstract"><div class="htmlview paragraph">Roundabouts are one of the most complex traffic environments in urban roads, and a key challenge for intelligent driving decision-making. Deep reinforcement learning, as an emerging solution for intelligent driving decisions, has the advantage of avoiding complex algorithm design and sustainable iteration. For the decision difficulty in roundabout scenarios, this paper proposes an artificial potential field based Soft Actor-Critic (APF-SAC) algorithm. Firstly, based on the Carla simulator and Gym framework, a reinforcement learning simulation system for roundabout driving is built. Secondly, to reduce reinforcement learning exploration difficulty, global path planning and path smoothing algorithms are designed to generate and optimize the path to guide the agent. Then, considering the complex interactions between vehicles in roundabouts, a Markov decision process model is constructed, and a coupled longitudinal and lateral action space, a vectorized state space based on roundabout scenarios, and a reward function based on artificial potential field are designed, and the APF-SAC algorithm is proposed. Finally, simulation experiments under different traffic densities show that compared to rule-based driving decisions, the deep reinforcement learning method can significantly improve decision safety and driving efficiency in roundabout scenarios, with the maximum safety improvement of 10.4% and the maximum driving efficiency improvement of 13.2%, demonstrating the superior performance of the APF-SAC algorithm for roundabout driving decisions. This research provides an effective approach for applying reinforcement learning algorithms to complex urban autonomous driving decisions.</div></div>

Research on robust decision making for intelligent connected vehicle at highway on-ramp

Decision-Making under On-Ramp merge Scenarios by Distributional Soft Actor-Critic Algorithm

Autonomous Highway Merging in Mixed Traffic Using Reinforcement Learning and Motion Predictive Safety Controller

Distributional Soft Actor-Critic for Decision-Making in On-Ramp Merge Scenarios

Towards Robust Decision-Making for Autonomous Highway Driving Based on Safe Reinforcement Learning

Decision-Making for Autonomous Vehicles in Random Task Scenarios at Unsignalized Intersection Using Deep Reinforcement Learning

Behavioral Decision-Making for Autonomous Driving Using Soft Actor-Critic Algorithm

Towards Robust Decision-Making for Autonomous Driving on Highway

Autonomous Ramp Merge Maneuver Based on Reinforcement Learning with Continuous Action Space

Deep Reinforcement Learning Based Decision-Making Strategy of Autonomous Vehicle in Highway Uncertain Driving Environments

A Rear Anti-Collision Decision-Making Methodology Based on Deep Reinforcement Learning for Autonomous Commercial Vehicles

Novel Decision-Making Strategy for Connected and Autonomous Vehicles in Highway On-Ramp Merging

A Comprehensive Driving Decision-Making Methodology Based on Deep Reinforcement Learning for Automated Commercial Vehicles

A Roadside Decision-Making Methodology Based on Deep Reinforcement Learning to Simultaneously Improve the Safety and Efficiency of Merging Zone.

Enhancing Cooperation of Vehicle Merging Control in Heavy Traffic Using Communication-Based Soft Actor-Critic Algorithm

Decision-making Strategy on Highway for Autonomous Vehicles using Deep Reinforcement Learning

RACE-SM: Reinforcement Learning Based Autonomous Control for Social On-Ramp Merging

Highway Decision-Making and Motion Planning for Autonomous Driving via Soft Actor-Critic

Investigating Autonomous Vehicle Driving Strategies in Highway Ramp Merging Zones

Research on Artificial Potential Field based Soft Actor-Critic Algorithm for Roundabout Driving Decision

Towards Safe Decision-Making for Autonomous Vehicles at Unsignalized Intersections