Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Multi-agent Path Finding for Cooperative Autonomous Driving

Zhongxia Yan, Han Zheng, Cathy Wu

TL;DR

The paper addresses coordinating autonomous vehicles at signal-free intersections by separating the problem into a crossing-ordering stage and an order-conditioned trajectory generation stage. It introduces OBS-KATS, a MAPF-inspired, optimal and complete algorithm that uses Kinematic Arrival Time Scheduling (KATS) for subzone reservations and a subsequent trajectory optimization step under a kinematic bicycle model. The main contributions include solving for a crossing order with OBS, proving soundness, completeness, and optimality of the approach, and demonstrating substantial delay reductions across diverse intersection configurations compared to FIFO, PP, and MCTS baselines. The work demonstrates robustness to arrival rates, lane lengths, and crossing speeds and provides public code for reproducibility, with implications for scalable cooperation in traffic and multi-robot settings.

Abstract

Anticipating possible future deployment of connected and automated vehicles (CAVs), cooperative autonomous driving at intersections has been studied by many works in control theory and intelligent transportation across decades. Simultaneously, recent parallel works in robotics have devised efficient algorithms for multi-agent path finding (MAPF), though often in environments with simplified kinematics. In this work, we hybridize insights and algorithms from MAPF with the structure and heuristics of optimizing the crossing order of CAVs at signal-free intersections. We devise an optimal and complete algorithm, Order-based Search with Kinematics Arrival Time Scheduling (OBS-KATS), which significantly outperforms existing algorithms, fixed heuristics, and prioritized planning with KATS. The performance is maintained under different vehicle arrival rates, lane lengths, crossing speeds, and control horizon. Through ablations and dissections, we offer insight on the contributing factors to OBS-KATS's performance. Our work is directly applicable to many similarly scaled traffic and multi-robot scenarios with directed lanes.

Multi-agent Path Finding for Cooperative Autonomous Driving

TL;DR

The paper addresses coordinating autonomous vehicles at signal-free intersections by separating the problem into a crossing-ordering stage and an order-conditioned trajectory generation stage. It introduces OBS-KATS, a MAPF-inspired, optimal and complete algorithm that uses Kinematic Arrival Time Scheduling (KATS) for subzone reservations and a subsequent trajectory optimization step under a kinematic bicycle model. The main contributions include solving for a crossing order with OBS, proving soundness, completeness, and optimality of the approach, and demonstrating substantial delay reductions across diverse intersection configurations compared to FIFO, PP, and MCTS baselines. The work demonstrates robustness to arrival rates, lane lengths, and crossing speeds and provides public code for reproducibility, with implications for scalable cooperation in traffic and multi-robot settings.

Abstract

Anticipating possible future deployment of connected and automated vehicles (CAVs), cooperative autonomous driving at intersections has been studied by many works in control theory and intelligent transportation across decades. Simultaneously, recent parallel works in robotics have devised efficient algorithms for multi-agent path finding (MAPF), though often in environments with simplified kinematics. In this work, we hybridize insights and algorithms from MAPF with the structure and heuristics of optimizing the crossing order of CAVs at signal-free intersections. We devise an optimal and complete algorithm, Order-based Search with Kinematics Arrival Time Scheduling (OBS-KATS), which significantly outperforms existing algorithms, fixed heuristics, and prioritized planning with KATS. The performance is maintained under different vehicle arrival rates, lane lengths, crossing speeds, and control horizon. Through ablations and dissections, we offer insight on the contributing factors to OBS-KATS's performance. Our work is directly applicable to many similarly scaled traffic and multi-robot scenarios with directed lanes.
Paper Structure (19 sections, 3 theorems, 3 equations, 3 figures, 1 table, 1 algorithm)

This paper contains 19 sections, 3 theorems, 3 equations, 3 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Cooperative Driving at Intersections
  4. Multi-agent Path Finding
  5. Continuous MAPF and Multi-robot Motion Planning
  6. Problem Formulation
  7. Kinematic Bicycle Model
  8. Optimal Crossing Order Search with MAPF
  9. Kinematic Arrival Time Scheduling (KATS)
  10. Prioritized Planning (PP) with Traffic Heuristics
  11. Order-based Search (OBS) with Traffic Heuristics
  12. Trajectory Optimization
  13. Why is Crossing Order Useful?
  14. Default Experimental Setup
  15. Experimental Results
  16. ...and 4 more sections

Key Result

Theorem 1

If a crossing order is feasible, calling KATS in this order obtains the optimal constant-speed crossing times for all vehicles consistent with the crossing order.

Figures (3)

  • Figure 1: Geometry of our studied intersection. Our algorithms are applicable to junctions in general, e.g. merging, as the exact geometry is encoded by the start and end positions of subzones along vehicle routes, $x_{z, r}$.
  • Figure 2: Delay vs computation time per crossing order replan.
  • Figure 3: Delay along entire route vs crossing geometry

Theorems & Definitions (6)

  • Theorem 1
  • proof
  • Theorem 2
  • proof
  • Theorem 3
  • proof