An Intelligent Collision Avoidance Algorithm for Unmanned Surface Vehicles Based on Deep Reinforcement Learning

Bowen Zu; Xiang Wang; Xuehua Zhou; Zhiguo Zhou

doi:10.1142/S2301385026500135

An Intelligent Collision Avoidance Algorithm for Unmanned Surface Vehicles Based on Deep Reinforcement Learning

Bowen Zu, Xiang Wang, Xuehua Zhou, Zhiguo Zhou^*

^*Corresponding author for this work

Beijing Institute of Technology

Research output: Contribution to journal › Article › peer-review

Abstract

To address the obstacle avoidance challenge for unmanned surface vehicles, this paper presents a novel intelligent algorithm based on deep reinforcement learning. The algorithm incorporates human demonstration experience data for quick convergence and efficient decision-making. It features an end-to-end framework for multi-sensor data processing and immediate action decisions. Both simulation and deployment experiments evidence the superiority of this algorithm.

Original language	English
Journal	Unmanned Systems
DOIs	https://doi.org/10.1142/S2301385026500135
Publication status	Accepted/In press - 2025
Externally published	Yes

Keywords

Unmanned surface vehicles
artificial intelligence
deep reinforcement learning
obstacle avoidance

Access to Document

10.1142/S2301385026500135

Cite this

@article{add46455e9d54ad2a2e2798296ad32f0,

title = "An Intelligent Collision Avoidance Algorithm for Unmanned Surface Vehicles Based on Deep Reinforcement Learning",

abstract = "To address the obstacle avoidance challenge for unmanned surface vehicles, this paper presents a novel intelligent algorithm based on deep reinforcement learning. The algorithm incorporates human demonstration experience data for quick convergence and efficient decision-making. It features an end-to-end framework for multi-sensor data processing and immediate action decisions. Both simulation and deployment experiments evidence the superiority of this algorithm.",

keywords = "Unmanned surface vehicles, artificial intelligence, deep reinforcement learning, obstacle avoidance",

author = "Bowen Zu and Xiang Wang and Xuehua Zhou and Zhiguo Zhou",

note = "Publisher Copyright: {\textcopyright} 2026 World Scientific Publishing Company.",

year = "2025",

doi = "10.1142/S2301385026500135",

language = "English",

journal = "Unmanned Systems",

issn = "2301-3850",

publisher = "World Scientific",

}

TY - JOUR

T1 - An Intelligent Collision Avoidance Algorithm for Unmanned Surface Vehicles Based on Deep Reinforcement Learning

AU - Zu, Bowen

AU - Wang, Xiang

AU - Zhou, Xuehua

AU - Zhou, Zhiguo

PY - 2025

Y1 - 2025

N2 - To address the obstacle avoidance challenge for unmanned surface vehicles, this paper presents a novel intelligent algorithm based on deep reinforcement learning. The algorithm incorporates human demonstration experience data for quick convergence and efficient decision-making. It features an end-to-end framework for multi-sensor data processing and immediate action decisions. Both simulation and deployment experiments evidence the superiority of this algorithm.

AB - To address the obstacle avoidance challenge for unmanned surface vehicles, this paper presents a novel intelligent algorithm based on deep reinforcement learning. The algorithm incorporates human demonstration experience data for quick convergence and efficient decision-making. It features an end-to-end framework for multi-sensor data processing and immediate action decisions. Both simulation and deployment experiments evidence the superiority of this algorithm.

KW - Unmanned surface vehicles

KW - artificial intelligence

KW - deep reinforcement learning

KW - obstacle avoidance

UR - https://www.scopus.com/pages/publications/105009138506

U2 - 10.1142/S2301385026500135

DO - 10.1142/S2301385026500135

M3 - Article

AN - SCOPUS:105009138506

SN - 2301-3850

JO - Unmanned Systems

JF - Unmanned Systems

ER -

An Intelligent Collision Avoidance Algorithm for Unmanned Surface Vehicles Based on Deep Reinforcement Learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this