Uncertainty-Aware BIT*: Collision-Free Path Planning for Maritime Autonomous Surface Ships under Target Ship Position Uncertainty
Sojin Kim, Jeonghong Park
AI summary
Problem
Remote operation of Maritime Autonomous Surface Ships relies on sensor data (AIS/radar) that inherently contains measurement errors and communication delays, making deterministic collision avoidance unsafe and non-compliant with navigation regulations.
Approach
The authors extend the BIT* sampling-based path planner with a custom edge cost function that quantifies collision risk using a 2D Gaussian uncertainty model, penalizes COLREGs violations, and enforces realistic turning radius constraints.
Key results
- Higher target ship uncertainty directly increases the Distance to Closest Point of Approach (DCPA) during avoidance maneuvers
- The planner generates more conservative starboard avoidance paths under elevated uncertainty levels
- Successfully navigates complex static port obstacles while dynamically avoiding target ships in head-on and crossing scenarios
- Validated in real-time ROS2/OMPL simulations on a high-fidelity ENC grid map of Ulsan Port
Why it matters
Provides a practical, regulation-aware planning framework that enhances the safety and operational readiness of remotely operated autonomous ships in uncertain maritime environments.
Abstract
This paper proposes a BIT*-based collision-free path planning method for remotely operated Maritime Au- tonomous Surface Ships (MASS) under target ship position uncertainty. In a remote operating environment, the Remote Operating Center (ROC) receives target ship position data from long-range sensors such as AIS and radar. This data inherently contains uncertainty due to measurement errors and communication delays, making it essential to account for such uncertainty during path planning. As MASS must comply with the International Regulations for Preventing Collisions at Sea (COLREGs) during navigation, path planning must also reflect these requirements. The proposed method models target ship position uncertainty as a two-dimensional Gaussian distribu- tion and incorporates it into edge evaluation as a collision risk cost, while applying penalty costs to edges that enter COLREGs non-compliant regions. A turning radius constraint is also incorporated into the edge selection process to ensure navigational feasibility. The method is validated through head- on and crossing encounter simulations on an Electronic Nautical Chart (ENC)-based grid map of Ulsan Port, South Korea. The results show that higher levels of position uncertainty lead to more conservative avoidance paths, resulting in greater Distance to Closest Point of Approach (DCPA).