Scholarlink

<p>With the extensive application of Automated Guided Vehicles (AGVs) in intelligent manufacturing, path planning for AGVs has emerged as a critical issue. Solving conflict and deadlock problems is particularly challenging, and deadlocks are especially difficult to identify due to the independence of channels. After the conflict and deadlock models are analyzed, the Directional Channel-Extended Time Window (DCETW) model is designed in which the relationship between channels is established through the cross node, and the deadlock is identified and converted into conflicts. The Channel-Borrowed A* (CBA*) algorithm is proposed, in which AGV waits in the adjacent channel of node and then returns to the original path after another AGV smoothly passes through the node when a conflict occurs. Additionally, the same direction travel strategy is integrated to reduce the direction alternations, thereby improving the channel utilization. Finally, orthogonal experiments are conducted to validate that deadlock is effectively avoided in the DCETW model and the CBA* algorithm outperforms the traditional A* algorithm in solution quality.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel constraint model for conflict-free path planning in multi-AGV system

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Zhengfeng Li,
  • Wenxiao Ye,
  • Huichao Shang,
  • Guohui Zhang,
  • Shunqi Mei

摘要

With the extensive application of Automated Guided Vehicles (AGVs) in intelligent manufacturing, path planning for AGVs has emerged as a critical issue. Solving conflict and deadlock problems is particularly challenging, and deadlocks are especially difficult to identify due to the independence of channels. After the conflict and deadlock models are analyzed, the Directional Channel-Extended Time Window (DCETW) model is designed in which the relationship between channels is established through the cross node, and the deadlock is identified and converted into conflicts. The Channel-Borrowed A* (CBA*) algorithm is proposed, in which AGV waits in the adjacent channel of node and then returns to the original path after another AGV smoothly passes through the node when a conflict occurs. Additionally, the same direction travel strategy is integrated to reduce the direction alternations, thereby improving the channel utilization. Finally, orthogonal experiments are conducted to validate that deadlock is effectively avoided in the DCETW model and the CBA* algorithm outperforms the traditional A* algorithm in solution quality.