Dual-mode deep reinforcement learning for safety-oriented MASS collision avoidance
- 1Navigation College, Dalian Maritime University
Abstract
Safe collision avoidance for Maritime Autonomous Surface Ships (MASS) remains challenging because autonomous controllers must maintain predictable rule-guided behaviour while adapting to dense and uncertain traffic situations. This research introduces a dual-mode, safety-oriented deep reinforcement learning (DRL) framework that integrates model-based predictability with data-driven adaptability for MASS collision avoidance. Encounter scenarios are quantified in accordance with the International Regulations for Preventing Collisions at Sea (COLREGs), and a multi-objective reward function combines pairwise rule-guided reasoning with dynamic risk awareness. In the routine-navigation mode, Proximal Policy Optimisation (PPO) is employed to generate stable and COLREGs-guided trajectories, whilst the heightened-safety mode applies a tree-based safety filter that prunes unsafe actions and supports autonomous switching under elevated close-quarters risk. Two pruning optimisations–Reachable Envelope Pruning (REP) and the Terminal Safety Criterion (TSC)–jointly reduce node expansions by over 90%, thereby markedly improving computational efficiency. Simulation results demonstrate that the safety layer achieved no observed collisions in two-ship encounters and reduced collision rates by approximately 80–90% in congested multi-ship scenarios compared with the routine-navigation PPO baseline without the heightened-safety mode. Even in six-ship traffic, the agent maintains an average minimum passing distance exceeding 0.5 nautical miles. Latency analysis indicates reduced computational overhead under the tested simulation settings, suggesting the decision-level computational feasibility of the proposed framework.
Related
- The impact of temporal resampling on the safety of clinical reinforcement learning: A simulated trial
- Balancing Iliac Crest Avoidance and Neural Safety in L5/S1 TELD: Quantitative Identification of a Candidate Balanced Trajectory Window Using AI‐Assisted 3D MRI Reconstruction
- Forecasting accuracy is not enough: safety-oriented evaluation of neural digital twins prior to in-silico insulin and carbohydrate intervention in type 1 diabetes
- Reinforcement learning–coupled neural ode modelling reveals ferroptotic resolution failure as a shared dynamical axis linking periodontitis and coronary atherogenesis
- DEEP REINFORCEMENT LEARNING IN ONCOLOGY: ADVANCES IN CANCER IMAGING, RADIOTHERAPY, AND PERSONALIZED TREATMENT
BibTeX
@article{du2026dualmode,
title = {{Dual-mode deep reinforcement learning for safety-oriented MASS collision avoidance}},
author = {Yifan Du and Feixiang Zhu and Moxuan Wei and Lai Wei},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62506-2},
url = {https://doi.org/10.1038/s41598-026-62506-2},
note = {CC BY},
}
RIS
TY - JOUR
TI - Dual-mode deep reinforcement learning for safety-oriented MASS collision avoidance
AU - Yifan Du
AU - Feixiang Zhu
AU - Moxuan Wei
AU - Lai Wei
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62506-2
UR - https://doi.org/10.1038/s41598-026-62506-2
AB - Safe collision avoidance for Maritime Autonomous Surface Ships (MASS) remains challenging because autonomous controllers must maintain predictable rule-guided behaviour while adapting to dense and uncertain traffic situations. This research introduces a dual-mode, safety-oriented deep reinforcement learning (DRL) framework that integrates model-based predictability with data-driven adaptability for MASS collision avoidance. Encounter scenarios are quantified in accordance with the International Regulations for Preventing Collisions at Sea (COLREGs), and a multi-objective reward function combines pairwise rule-guided reasoning with dynamic risk awareness. In the routine-navigation mode, Proximal Policy Optimisation (PPO) is employed to generate stable and COLREGs-guided trajectories, whilst the heightened-safety mode applies a tree-based safety filter that prunes unsafe actions and supports autonomous switching under elevated close-quarters risk. Two pruning optimisations–Reachable Envelope Pruning (REP) and the Terminal Safety Criterion (TSC)–jointly reduce node expansions by over 90%, thereby markedly improving computational efficiency. Simulation results demonstrate that the safety layer achieved no observed collisions in two-ship encounters and reduced collision rates by approximately 80–90% in congested multi-ship scenarios compared with the routine-navigation PPO baseline without the heightened-safety mode. Even in six-ship traffic, the agent maintains an average minimum passing distance exceeding 0.5 nautical miles. Latency analysis indicates reduced computational overhead under the tested simulation settings, suggesting the decision-level computational feasibility of the proposed framework.
SN - 2045-2322
ER -