Ship collision avoidance using timespace parametrization: Robust and risk-aware planning Using SeaCharts implementation

The complexity of maritime environments poses several challenges to the development of autonomous ships. While avoiding grounding is usually the primary concern, ships must also navigate around dynamic obstacles. Finding a suitable trade-off between the original path and the trajectory required to avoid dynamic obstacles is challenging, particularly because most existing algorithms use different metrics to quantify the risk posed by static and dynamic obstacles. This paper introduces a COLREGs-compliant trajectory planning framework that accounts for bounded uncertainties in surrounding vessel speed and course angle, effectively converting dynamic obstacle avoidance into a path planning problem where all obstacles are represented as polygons by parameterizing time in space. This unified representation integrates global and local planners, allowing any path planning algorithm to be converted into a trajectory planner. The method accounts for the own ship’s speed and course rate constraints. The framework is supported by new features of SeaCharts, a Python library for computation based on Electronic Navigational Charts, which we extend to read S-57 files and access weather data. The approach is validated through three case studies, where it shows reduced sensitivity to target ship motion uncertainty compared to velocity obstacle and scenario-based MPC methods.