Scholarlink

<p>Trajectory tracking for Discretely Actuated Hyper-Redundant Manipulators (DAHRMs)—composed of serially connected modular units—is challenging due to their discrete configuration space, requiring specialized optimization approaches. This paper introduces a novel full-path cost function that evaluates deviation across entire path segments, not just at waypoints. This key innovation enables significantly more accurate tracking without increasing motion time or sacrificing smoothness, providing a more effective alternative to the traditional approach of simply adding more path segments. The method is integrated into a discrete optimization framework and adapted for five established solvers (Two-by-Two Search, Narrowing-Down Search, Genetic Algorithm, Particle Swarm Optimization, and Grey Wolf Optimizer). Validation via 2D and 3D case studies with linear and curved paths confirms the method’s substantial improvement over endpoint-only optimization. The results show solver performance is highly geometry-dependent, with no single algorithm optimal for all scenarios. Furthermore, the framework’s predictable computational scaling confirms its feasibility for real-time applications.</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

Trajectory tracking of discretely actuated hyper-redundant manipulators via full-path cost optimization

  • 期刊论文

  • 发表日期：

    2026年2月12日

查看全文
  • Alireza Motahari,
  • Gholamreza Khalaj,
  • Sadegh Ghorbanhosseini

摘要

Trajectory tracking for Discretely Actuated Hyper-Redundant Manipulators (DAHRMs)—composed of serially connected modular units—is challenging due to their discrete configuration space, requiring specialized optimization approaches. This paper introduces a novel full-path cost function that evaluates deviation across entire path segments, not just at waypoints. This key innovation enables significantly more accurate tracking without increasing motion time or sacrificing smoothness, providing a more effective alternative to the traditional approach of simply adding more path segments. The method is integrated into a discrete optimization framework and adapted for five established solvers (Two-by-Two Search, Narrowing-Down Search, Genetic Algorithm, Particle Swarm Optimization, and Grey Wolf Optimizer). Validation via 2D and 3D case studies with linear and curved paths confirms the method’s substantial improvement over endpoint-only optimization. The results show solver performance is highly geometry-dependent, with no single algorithm optimal for all scenarios. Furthermore, the framework’s predictable computational scaling confirms its feasibility for real-time applications.