A Web-Based Virtual Test Environment for Autonomous Navigation Systems in High-Traffic Port Waters

Abstract

Testing Autonomous Navigation Systems (ANS) in port waters with heavy traffic demands more than a good algorithm. It requires infrastructure: a way to organize real-world encounter scenarios, replay vessel behavior, connect to external decision systems, and document what happened. This paper describes the architecture of a web-based Virtual Test Environment (VTE) built for ANS testing in the Singapore Strait. The contribution is not a new collision avoidance algorithm but a unified platform that ties together browser-based visualization, AIS-driven scenario management, a backend vessel maneuverability engine, and reporting into a single test environment. A modular layered design separates operator interaction, application services, data persistence, and external ANS integration, so that risk assessment modules, COLREGs compliance checking, and higher-fidelity vessel models can be added incrementally.