Ship Collision Risk Evaluation Using Track Deviation and DCPA
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current risk determination methods for ship collisions based on relative distance between ships are inadequate, often leading to erroneous detection of avoidance behaviors, particularly in complex navigation scenarios like harbor entries, where course changes may be misinterpreted.
Innovation Solution
A computer-readable medium and evaluation device that extract sailing logs from ships in close proximity to identify deviations from established sailing routes and calculate the distance of closest approach (DCPA) to accurately detect avoidance behaviors and evaluate collision risks, generating a hazard map for high-risk areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If risk determination is executed based on relative distance between ships, then collision risk assessment is performed, but dangerous places with high collision risk cannot be detected accurately
Solution Approach 1:
The patent segments the risk assessment process into multiple evaluation dimensions: relative distance, ship track deviation from sailing route, and number of course changes. By dividing the assessment into these distinct segments, the system can accurately identify dangerous places while avoiding false positives from normal navigation maneuvers.
Solution Approach 2:
The patent adds a new dimension to the risk assessment by incorporating ship track analysis and deviation from sailing routes. Instead of relying solely on relative distance (one dimension), the system evaluates multiple dimensions including whether ships are deviating from their designated routes and how frequently they are changing courses, thereby improving detection accuracy.
2Productivity
If risk determination is based on relative distance between ships, then collision risk is evaluated, but course changes for harbor entry are erroneously detected as avoidance behavior
Solution Approach 1:
The patent segments the behavior analysis into distinct components: normal course changes (such as harbor entries) versus avoidance maneuvers. By analyzing whether course changes occur at expected locations (harbor entrances) and evaluating the pattern of deviations, the system can distinguish between routine navigation and genuine avoidance behavior, reducing false detections.
Solution Approach 2:
The patent introduces an intermediary analysis layer that examines ship track deviations and compares them against known sailing routes and harbor entry patterns. This intermediary evaluation acts as a filter between raw distance data and final avoidance behavior classification, preventing erroneous detections by contextualizing course changes within known navigation patterns.
3Device complexity
If only relative distance is used for risk determination, then the evaluation process is simple, but dangerous places cannot be accurately identified
Solution Approach 1:
The patent segments the evaluation process into modular components: distance calculation, track deviation analysis, course change counting, and dangerous place identification. This segmentation allows the system to maintain relative simplicity while incorporating multiple evaluation criteria, as each module can be independently processed and combined to achieve accurate detection.
Data Source
AI summary
A non-transitory computer-readable recording medium stores a program that causes a computer to execute a process. The process includes extracting a sailing log of a ship in a sea area with a sailing route existing therein, the sea area being such that a distance between two ships of the ship and another ship is less than a predetermined value and at least one of the two ships has a ship track deviating from the sailing route; and evaluating risks at a plurality of points that are included in the sea area, based on the extracted sailing log.