Robust motion planning for mobile robots under attacks against obstacle localization

Thanks to its real-time computation efficiency, deep reinforcement learning (DRL) has been widely applied in motion planning for mobile robots. In DRL-based methods, a DRL model computes an action for a robot based on the states of its surrounding obstacles, including other robots that may communica...

Full description

Saved in:
Bibliographic Details
Published in:Robotica 2024-08, Vol.42 (8), p.2781-2800
Main Authors: Wu, Fenghua, Tang, Wenbing, Zhou, Yuan, Lin, Shang-Wei, Ding, Zuohua, Liu, Yang
Format: Article
Language:English
Subjects:
Algorithms
Approximation
Barriers
Collision avoidance
Efficiency
Generative adversarial networks
Global positioning systems
GPS
Localization
Methods
Motion planning
Performance evaluation
Planning
Real time
Robot control
Robot dynamics
Robots
Robust control
Sensors
Velocity
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Thanks to its real-time computation efficiency, deep reinforcement learning (DRL) has been widely applied in motion planning for mobile robots. In DRL-based methods, a DRL model computes an action for a robot based on the states of its surrounding obstacles, including other robots that may communicate with it. These methods always assume that the environment is attack-free and the obtained obstacles’ states are reliable. However, in the real world, a robot may suffer from obstacle localization attacks (OLAs), such as sensor attacks, communication attacks, and remote-control attacks, which cause the robot to retrieve inaccurate positions of the surrounding obstacles. In this paper, we propose a robust motion planning method ObsGAN-DRL, integrating a generative adversarial network (GAN) into DRL models to mitigate OLAs in the environment. First, ObsGAN-DRL learns a generator based on the GAN model to compute the approximation of obstacles’ accurate positions in benign and attack scenarios. Therefore, no detectors are required for ObsGAN-DRL. Second, by using the approximation positions of the surrounding obstacles, ObsGAN-DRL can leverage the state-of-the-art DRL methods to compute collision-free motion commands (e.g., velocity) efficiently. Comprehensive experiments show that ObsGAN-DRL can mitigate OLAs effectively and guarantee safety. We also demonstrate the generalization of ObsGAN-DRL.
ISSN:0263-5747
1469-8668
DOI:10.1017/S0263574724001115