Scholarlink

<p>In modern aviation systems, the wide application of unmanned aerial vehicles (UAVs) promotes the development of path planning technology. However, UAV path planning still faces challenges such as complex obstacle avoidance and efficient path optimization, which makes it difficult for traditional algorithms to meet the requirements of real-time and dynamic environments. Meta-heuristic (MH) algorithms have become an effective tool for solving such problems because of their high global optimization capability, local optimum avoidance, and adaptability. This study proposes a novel MH algorithm, quaternion-based dung beetle optimizer (QDBO), to improve the path planning performance of UAVs in complex mountainous environments. In QDBO, we design three improvement strategies based on quaternion theory to enhance the algorithm’s ability to explore, exploit, and escape the local optimum. The performance of QDBO is evaluated on the designed 14 simulated flight missions. The experimental results show that the QDBO can generate smooth and direct paths, avoiding unnecessary detours, and the performance in a high-density obstacle environment is better than other algorithms. Specifically, compared to the worst algorithm, the average total cost of QDBO in the 14 flight missions is reduced by 4.70%, 7.67%, 13.44%, 30.68%, 32.60%, 30.09%, 19.03%, 17.50%, 20.82%, 17.89%, 22.61%, 21.69%, 21.12%, and 17.86%, respectively. The numerical results verify the QDBO’s excellent performance in solving the UAV path planning problem in challenging mountainous environments.</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

3D path planning in complex mountainous environments for UAVs using quaternion-based dung beetle optimizer

  • 期刊论文

  • 发表日期：

    2025年9月30日

查看全文
  • Huangzhi Xia,
  • Linqi Sun,
  • Huanghui Xia,
  • Seyedali Mirjalili

摘要

In modern aviation systems, the wide application of unmanned aerial vehicles (UAVs) promotes the development of path planning technology. However, UAV path planning still faces challenges such as complex obstacle avoidance and efficient path optimization, which makes it difficult for traditional algorithms to meet the requirements of real-time and dynamic environments. Meta-heuristic (MH) algorithms have become an effective tool for solving such problems because of their high global optimization capability, local optimum avoidance, and adaptability. This study proposes a novel MH algorithm, quaternion-based dung beetle optimizer (QDBO), to improve the path planning performance of UAVs in complex mountainous environments. In QDBO, we design three improvement strategies based on quaternion theory to enhance the algorithm’s ability to explore, exploit, and escape the local optimum. The performance of QDBO is evaluated on the designed 14 simulated flight missions. The experimental results show that the QDBO can generate smooth and direct paths, avoiding unnecessary detours, and the performance in a high-density obstacle environment is better than other algorithms. Specifically, compared to the worst algorithm, the average total cost of QDBO in the 14 flight missions is reduced by 4.70%, 7.67%, 13.44%, 30.68%, 32.60%, 30.09%, 19.03%, 17.50%, 20.82%, 17.89%, 22.61%, 21.69%, 21.12%, and 17.86%, respectively. The numerical results verify the QDBO’s excellent performance in solving the UAV path planning problem in challenging mountainous environments.