Optimization of Edge Directions and Weights for Mixed Guidance Graphs in Lifelong Multi-Agent Path Finding

Researchers propose Mixed Guidance Graph Optimization (MGGO) to improve multi-agent pathfinding systems by optimizing both edge directions and weights in guidance graphs. The paper introduces two MGGO methods, including one using Quality Diversity algorithms with neural networks, to provide stricter guidance for agent movement in lifelong scenarios.

• →Traditional Guidance Graph Optimization only provides soft guidance through edge weights, while MGGO adds strict guidance through edge direction optimization.
• →Two new MGGO methods are introduced: a two-phase approach separating direction and weight optimization, and a neural network-based Quality Diversity approach.
• →The research addresses limitations in lifelong multi-agent pathfinding where agents continuously receive new goals after completing current ones.
• →Traffic pattern incorporation into guidance graphs enables more sophisticated edge-direction-aware pathfinding solutions.
• →The work represents an advancement in AI pathfinding algorithms with potential applications in robotics and automated systems.