Scholarlink

<p>This work presents a complete framework for the modeling, workspace analysis, and trajectory planning of a tendon-driven continuum manipulator (TDCM). For modeling purposes, a Petrov–Galerkin Cosserat rod finite element formulation is adopted, and material parameters are identified through dedicated experiments. The identified model shows good agreement with the optical motion capture data, with a position error of no greater than 4.5% of the total TDCM length. We use brute-force sampling and the alpha shape method to numerically determine the workspace of the TDCM, which is bounded by two umbrella-shaped surfaces. By formulating a constrained optimization problem, the inverse statics problem is solved sequentially for trajectory planning. To systematically analyse the model accuracy within the workspace, 10 reference trajectories with different radii were generated and implemented on the test bench. The worst-case position error of 11.1&#xa0;mm hints at the influence of unmodeled effects, such as tendon routing friction. All trajectories were successfully evaluated without failure, demonstrating the feasibility of the proposed framework.</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

Modeling, workspace analysis and trajectory planning of a tendon-driven soft continuum manipulator

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月25日

查看全文
  • Tianxiang Dai,
  • Remco I. Leine,
  • Simon R. Eugster

摘要

This work presents a complete framework for the modeling, workspace analysis, and trajectory planning of a tendon-driven continuum manipulator (TDCM). For modeling purposes, a Petrov–Galerkin Cosserat rod finite element formulation is adopted, and material parameters are identified through dedicated experiments. The identified model shows good agreement with the optical motion capture data, with a position error of no greater than 4.5% of the total TDCM length. We use brute-force sampling and the alpha shape method to numerically determine the workspace of the TDCM, which is bounded by two umbrella-shaped surfaces. By formulating a constrained optimization problem, the inverse statics problem is solved sequentially for trajectory planning. To systematically analyse the model accuracy within the workspace, 10 reference trajectories with different radii were generated and implemented on the test bench. The worst-case position error of 11.1 mm hints at the influence of unmodeled effects, such as tendon routing friction. All trajectories were successfully evaluated without failure, demonstrating the feasibility of the proposed framework.