Hierarchical Behavior Planning Framework for Networked MSV Systems in Inland Waterway Traffic Separation Schemes

Abstract

Ensuring safety and preventing accidents in waterway channels are critical challenges for networked marine surface vessel systems (NMSVs). This study introduces a regulation-aware decision-making system designed to minimize traffic conflicts and enhance navigational safety in inland waterway traffic separation schemes. The proposed framework integrates a hierarchical conditional state machine with chance-constrained model predictive control, allowing NMSVs to handle complex traffic situations while complying with safety regulations. The hierarchical conditional state machine effectively identifies vessel maneuver states, implementing safety constraints that proactively avoid collisions. Meanwhile, the chance-constrained model predictive control optimizes vessel trajectories, factoring in uncertainties and potential risks, while simultaneously enhancing operational efficiency. Simulation and experimental results demonstrate that the proposed system significantly reduces the likelihood of accidents and improves overall safety by efficiently managing vessel interactions. Compared to traditional methods, the regulation-aware approach ensures better collision avoidance, greater regulation compliance, and superior safety performance. This study confirms that the proposed decision-making system can be effectively implemented in real time, offering practical benefits for improving waterway safety and mitigating accident risks.