Learning When to Drive in Intersections by Combining Reinforcement\n Learning and Model Predictive Control

2019/07/31 by Tommy Tram, Ivo Batkovic, Tram, Tommy +5 · 1 citation
Engineering · #Traffic control and management #Autonomous Vehicle Technology and Safety #Vehicle emissions and performance

paper · pdf · doi:10.48550/arxiv.1908.00177

Abstract

In this paper, we propose a decision making algorithm intended for automated\nvehicles that negotiate with other possibly non-automated vehicles in\nintersections. The decision algorithm is separated into two parts: a high-level\ndecision module based on reinforcement learning, and a low-level planning\nmodule based on model predictive control. Traffic is simulated with numerous\npredefined driver behaviors and intentions, and the performance of the proposed\ndecision algorithm was evaluated against another controller. The results show\nthat the proposed decision algorithm yields shorter training episodes and an\nincreased performance in success rate compared to the other controller.\n

Cited by

• Bridging Reinforcement Learning and Optimal Control via Feasible Action Mapping

Related

• Prediction Based Decision Making for Autonomous Highway Driving
• Autonomous Driving at Unsignalized Intersections: A Review of Decision-Making Challenges and Reinforcement Learning-Based Solutions
• Adaptive Decision Making at the Intersection for Autonomous Vehicles Based on Skill Discovery
• Behavior Planning at Urban Intersections through Hierarchical Reinforcement Learning
• Courteous Behavior of Automated Vehicles at Unsignalized Intersections via Reinforcement Learning