Scholarlink

In this paper, we study a hybrid approach for the combined truck-drone delivery problem. This approach integrates DCA (Difference of Convex functions Algorithms) into the Deep Q-learning (DQN) algorithm with the purpose of modifying the Q-values update. A numerical comparison on the performance of DQN and DQN-DCA is given on synthesis datasets to show the effectiveness of the proposed approach for solving the truck-drone delivery problem.

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

An Online DCA and DQN Based Approach for Truck-Drone Delivery Problems

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hoang Viet Nguyen,
  • Vinh Thanh Ho,
  • Hoai Minh Le

摘要

In this paper, we study a hybrid approach for the combined truck-drone delivery problem. This approach integrates DCA (Difference of Convex functions Algorithms) into the Deep Q-learning (DQN) algorithm with the purpose of modifying the Q-values update. A numerical comparison on the performance of DQN and DQN-DCA is given on synthesis datasets to show the effectiveness of the proposed approach for solving the truck-drone delivery problem.