Collision-Free Trajectory Planning of Mobile Robots by Integrating Deep Reinforcement Learning and Model Predictive Control
Paper in proceeding, 2023

In this paper, we present an efficient approach to real-time collision-free navigation for mobile robots. By integrating deep reinforcement learning with model predictive control, our aim is to achieve both collision avoidance and computational efficiency. The methodology begins with training a preliminary agent using deep Q-learning, enabling it to generate actions for next time steps. Instead of executing these actions, a reference trajectory is generated based on them, which avoids obstacles present on the original reference path. Subsequently, this local trajectory is employed within an MPC trajectory-tracking framework to provide collision-free guidance for the mobile robot. Experimental results demonstrate that the proposed DQN-MPC hybrid approach outperforms pure MPC in terms of time efficiency and solution quality.

Author

Ze Zhang

Chalmers, Electrical Engineering, Systems and control

Yao Cai

Chalmers, Electrical Engineering, Systems and control

Kristian Ceder

Chalmers, Electrical Engineering, Systems and control

Arvid Enliden

Student at Chalmers

Ossian Eriksson

Student at Chalmers

Soleil Kylander

Student at Chalmers

Rajath Sridhara

Student at Chalmers

Knut Åkesson

Chalmers, Electrical Engineering, Systems and control

IEEE International Conference on Automation Science and Engineering

21618070 (ISSN) 21618089 (eISSN)

Vol. 2023-August
9798350320695 (ISBN)

19th IEEE International Conference on Automation Science and Engineering, CASE 2023
Auckland, New Zealand,

AIHURO-Intelligent human-robot collaboration

VINNOVA (2022-03012), 2023-02-01 -- 2026-01-31.

Subject Categories

Robotics

Control Engineering

Computer Science

DOI

10.1109/CASE56687.2023.10260515

More information

Latest update

11/20/2023