Closed-loop feedback optimization for autonomous vehicles using deep reinforcement learning

Sifan Wu

doi:10.1016/j.eswa.2026.131383

Closed-loop feedback optimization for autonomous vehicles using deep reinforcement learning

2026-02-12

0

PRE

AI

DOI：10.1016/j.eswa.2026.131383

Original

Original request for help

Save

Abstract

En 中文

• A novel closed-loop feedback framework optimizes motion planning trajectories and updates decision-making in real time. • A dual-value priority sampling strategy identifies high-value samples from both high-reward and low-reward experiences. • A hybrid action selection mechanism integrates the MOBIL model to guide exploration. • The framework outperforms state-of-the-art models and demonstrates its strong potential for real-world applications.

Keywords:

closed-loop feedback

deep reinforcement learning

motion planning

trajectory optimization

autonomous vehicles

Closed-loop feedback optimization for autonomous vehicles using deep reinforcement learning

Abstract

Journal

Organization