Scholarlink

<p>The development of a collision-free motion path to achieve efficient positioning of end-effector and safety of all links for multi-degree-of-freedom robotic arms operating in narrow spaces represents a considerable challenge. A novel hybrid algorithm named Whale Hybrid Improved Dung Beetle Optimizer (WHIDBO) is proposed by combining the Whale Optimization Algorithm (WOA) and improved Dung Beetle Optimization algorithm (DBO) with multi-strategy integration. The convergence pattern and spiral search strategy of the WOA are utilized to optimize the breeding and stealing behaviors of the DBO to enhance its exploitation ability. The random walk and adaptive <i>t</i>-distribution strategies are employed to strengthen the ball-rolling and foraging behaviors of the DBO to improve its exploration ability. The crisscross mechanism effectively enhances the balance between exploitation and exploration phases during optimization. Evaluated on CEC2022 functions, WHIDBO demonstrates superior convergence speed and stability compared to seven benchmark algorithms. Practical validation using a kinematic-constrained fitness function with cubic B-spline smoothing shows WHIDBO generates collision-free paths 21.15% shorter than DBO, while improving computational efficiency by 23.33%. These results highlight the potential of WHIDBO for industrial automation systems requiring reliable motion planning in geometrically complex 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

A novel hybrid improved Dung beetle optimization algorithm for robotic arm obstacle avoidance planning

  • 期刊论文

  • 发表日期：

    2025年9月22日

查看全文
  • Dong Chen,
  • Wuyin Jin,
  • Zhiyuan Rui,
  • Lan Luo,
  • Jiazhen Li,
  • Wentao Wang

摘要

The development of a collision-free motion path to achieve efficient positioning of end-effector and safety of all links for multi-degree-of-freedom robotic arms operating in narrow spaces represents a considerable challenge. A novel hybrid algorithm named Whale Hybrid Improved Dung Beetle Optimizer (WHIDBO) is proposed by combining the Whale Optimization Algorithm (WOA) and improved Dung Beetle Optimization algorithm (DBO) with multi-strategy integration. The convergence pattern and spiral search strategy of the WOA are utilized to optimize the breeding and stealing behaviors of the DBO to enhance its exploitation ability. The random walk and adaptive t-distribution strategies are employed to strengthen the ball-rolling and foraging behaviors of the DBO to improve its exploration ability. The crisscross mechanism effectively enhances the balance between exploitation and exploration phases during optimization. Evaluated on CEC2022 functions, WHIDBO demonstrates superior convergence speed and stability compared to seven benchmark algorithms. Practical validation using a kinematic-constrained fitness function with cubic B-spline smoothing shows WHIDBO generates collision-free paths 21.15% shorter than DBO, while improving computational efficiency by 23.33%. These results highlight the potential of WHIDBO for industrial automation systems requiring reliable motion planning in geometrically complex environments.