A Demonstration of Continuous Lifelong Conflict-Aware AGV Routing with Kinematic Constraints

Abstract

Automated Guided Vehicles (AGVs) are the backbone of modern e-commerce warehouses, but the existing multi-agent pathfinding (MAPF) solutions fail to simultaneously satisfy the practical requirements of continuous space and time, lifelong and dynamic tasks, kinematic movement constraints, online responsiveness, and large-scale scalability in real-world deployments. This demo presents a fully functional lifelong conflict-aware AGV routing system built on our novel routing algorithm OHSMD optimized for warehouse scenarios. The system supports customizable warehouse map configuration, AGV kinematic parameter tuning, lifelong task scheduling, conflict-free path planning, interactive visualization, and physical validation using Sony's Toio™ Core Cube robots. Through this demo, the audience can intuitively experience the system's performance and verify the practical availability of our routing algorithms in industrial warehouse scenarios.