Online Multi-Agent Path Planning in Runaway Scenarios

Not yet accepted / published

Authors

Lesur, Paul & Bajcinca, Naim

Abstract

In this work we present a real-time capable algorithm for solving path planning problems for the application of Runaway Scenarios. In such applications, a main agent has to follow a path, and other secondary agents should create space for the main agent, without colliding with one another. Our algorithm uses a low-level path planner, similar to what can be found in the literature for the problem of Multi-Agent Path Finding (MAPF), and combines it with our novel Planner Scheduler, a high-level scheduler that allows us to find a sub-optimal solution quickly.

Results

One can find the detailed results (numbers and animations) here.