Rapidly Deployable Cable-Driven Parallel Robots

可快速部署的电缆驱动并联机器人

• 批准号: RGPIN-2021-03294
• 负责人: Cardou, Philippe
• 金额: $ 1.97万
• 依托单位: Université Laval
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=754419
• 关键词: Rapidly; Deployable; Cable; Driven; Parallel

Cable--driven parallel robots (CDPRs) consist of an end effector suspended by a number of cables. Each cable is wound onto a servo-actuated spool, which is rigidly attached to a fixed frame. The end-effector displacements are controlled by simultaneously winding or unwinding the cables on their respective spools. Our interest in CDPRs stems from four advantages they hold over classical rigid-link robots: large workspaces; low cost; high speeds and accelerations; portability and reconfigurability. In our opinion, this latter advantage, the portability and reconfigurability of CDPRs, has been relatively neglected in previous research. We believe that this could be significantly improved by solving three problems: (1-2) making the CDPR calibration fully autonomous (3) designing new spools that can be mounted on towers rather than buttressed structures and (4-5) quickly tracing and certifying their workspace. The hardware will not have to be rebuilt from scratch for each project, because a dynamically reconfigurable CDPR is available at the Laboratoire de robotique. In projects (1-2), we will devise a mathematical method that allows to identify the CDPR mechanical model without the prior estimate of its parameters required by current calibration methods. Measuring the dimensions of a newly deployed cable robot is time consuming and constitutes an important limitation to its portability. The proposed method will be demonstrated through the dynamically reconfigurable CDPR already available and by designing a new highly portable lightweight CDPR. Project (3) was established from the observation that current CDPRs resemble more gantry cranes than tower cranes. To minimise the CDPR structure, we want to attach its spools on towers free of large buttresses, stabilisers or tethers. We will achieve this objective by cancelling the moment induced at the base of the tower by the cable tension applied at its top. A moving counterweight driven by the winch and cable direction will be used to cancel the moment. We will demonstrate this concept on a two--degree--of-freedom cable--suspended parallel robot in the vertical plane. In projects (4-5), we wish to develop faster methods for tracing and certifying the wrench-feasible workspace of CDPRs. The fastest available methods for tracing the workspace are based on line search, following rays emanating from a point until they intersect the workspace surface. We will propose a method where a path is followed along the workspace surface by using a predictor-corrector scheme. This method should be faster, as every point predicted along the path will be close to the true workspace surface, ensuring a fast correction step. As this type of workspace tracing method is not guaranteed to find the whole workspace, we will also develop a certification method. This method will consist in a branch-and-prune algorithm in which the lower bounds are computed through convex relaxations.

缆索驱动并联机器人由一个末端执行器悬挂在多根缆索上组成。每根电缆都缠绕在一个伺服驱动的线轴上，线轴刚性地连接在一个固定的框架上。末端执行器的位移通过同时在其各自的线轴上缠绕或退绕线缆来控制。我们对CDPR的兴趣来自于它们比传统刚性连杆机器人具有的四个优点：大的刚度;低成本;高速度和加速度;便携性和可重构性。在我们看来，后者的优势，CDPR的可移植性和可重构性，在以前的研究中相对被忽视。我们认为，这可以通过解决三个问题来显著改善：（1-2）使CDPR校准完全自主（3）设计可以安装在塔架上而不是支撑结构上的新线轴（4-5）快速跟踪和验证其工作空间。每个项目都不需要从头开始重建硬件，因为机器人实验室提供了可动态重新配置的CDPR。在项目（1-2）中，我们将设计一种数学方法，该方法允许识别CDPR力学模型，而无需当前校准方法所需的参数先验估计。测量一个新部署的电缆机器人的尺寸是耗时的，并构成了其便携性的重要限制。所提出的方法将证明通过动态可重构CDPR已经可用，并通过设计一个新的高度便携式轻量级CDPR。项目（3）是根据观察结果建立的，即目前的CDPR更像龙门起重机而不是塔式起重机。为了最大限度地减少CDPR结构，我们希望将其线轴连接在没有大型扶壁、稳定器或系绳的塔架上。为了达到这个目的，我们将取消由施加在塔顶的缆索张力在塔底引起的力矩。由绞车和电缆方向驱动的移动配重将用于抵消力矩。我们将证明这个概念上的两个-度-的自由度电缆-悬挂并联机器人在垂直平面。在项目（4-5）中，我们希望开发更快的方法来跟踪和证明CDPR的扳手可行工作空间。跟踪工作空间的最快可用方法是基于线搜索，跟踪从一个点发出的光线，直到它们与工作空间表面相交。我们将提出一种方法，其中路径是遵循沿着工作空间表面通过使用预测校正计划。该方法应该更快，因为沿着路径沿着预测的每个点将接近真实的工作空间表面，从而确保快速校正步骤。由于这种类型的工作空间跟踪方法不能保证找到整个工作空间，我们还将开发一种认证方法。这种方法将包括在一个分支和修剪算法，其中的下限计算通过凸松弛。