Control of Self-Driving Cars using Reinforcement Learning

Engineering notes

Key topics: autonomous driving, self-driving car, self-driving, reinforcement learning, control. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

Over the last decade, significant advances have been made in the Autonomous Driving field. Deep Reinforcement Learning has benefited the complex control and navigation-related challenges. This paper presents the implementation of a Deep Reinforcement Learning based algorithm for the control of a lane following car. A self-driving car is designed to keep a set speed and a safe distance from the lead vehicle while moving along the center of the lane. We use a Deep Reinforcement Learning based algorithm called Deep Deterministic Policy Gradient (DDPG) for the implementation of lateral and longitudinal control of the autonomous car system. The simulation results show that the system is able to drive autonomously in the dynamic environment and maneuver along the curved road.

Links and sources

• Official / arXiv page

Comments