AGV conflict-free path planning method and system, and storage medium
By improving the conflict search algorithm and the adaptive bounded suboptimal A* algorithm, the problems of insufficient high-level search and high low-level search overhead of the CBS algorithm in the cross-regional scheduling scenario of the spinning workshop are solved, realizing efficient AGV path planning, reducing computation time and alleviating conflicts and deadlocks.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG SCI-TECH UNIV
- Filing Date
- 2026-05-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing CBS algorithms suffer from insufficient directionality in high-level search, low efficiency in conflict splitting, and high overhead in low-level search in cross-regional scheduling scenarios in spinning workshops. This results in low efficiency in AGV path planning and a tendency for node conflicts and deadlocks.
An improved conflict search algorithm is adopted, which selects the candidate node set through high-level optimization factors, expands the node with the smallest heuristic function value, and uses an adaptive bounded suboptimal A* algorithm to plan the path for a single AGV at the low level. A dynamic relaxation factor is introduced to control the greediness of the search and prioritizes the handling of key conflicts and constraints.
It effectively reduces the expansion of invalid nodes, improves the efficiency of path replanning, significantly reduces computation time, and alleviates the problems of vehicle collisions and local deadlocks in narrow channels, thereby improving the operating efficiency of the AGV system.
Smart Images
Figure CN122408816A_ABST