As part of the project, Kpler participated in the 28th International Conference on Extending Database Technology (EDBT2025) to showcase two data-intensive applications designed for real-time maritime safety and situational awareness.
🔹 Automated Collision Avoidance System
Based on a distributed actor model using Apache Pekko, this system processes live AIS streams (via MarineTraffic/Kpler) and forecasts vessel-to-vessel interactions in real-time. It integrates COLREG-compliant path planning using a modified Frenét Frame Optimal Trajectory Generation algorithm, producing dynamic, navigable rerouting paths.
🔹 Hazardous Weather Rerouting System
This solution uses fused spatiotemporal data from Copernicus Marine Service and NOAA GFS weather forecasts over an H3 hexagonal grid, combined with historical AIS mobility patterns. It generates multi-objective vessel routing graphs and applies A* search with edge weights informed by weather and traffic patterns. The system supports ETA calculation and scenario comparison via a simplified waypoint interface.
Both systems are deployed on AWS and accessed via a unified middleware API, enabling simulation, visualization, and scenario exploration for both manned and unmanned vessels.