HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Design and stability analysis of visual servoing on cable-driven parallel robots for accuracy improvement

Abstract : This thesis presents accuracy improvement of Cable-Driven Parallel Robots (CDPRs) by visual servoing (VS) and the use of stability analysis to evaluate the robustness of the robotic system. CDPRs are a kind of parallel robots with cables instead of rigid links. They are characterized by a large workspace, a large payload capacity and reconfigurability, including a changeable moving-platform. However, CDPRs lack accuracy, which prevents them to be widely used. With an onboard camera on the moving-platform (MP) used in VS control of CDPRs, it is possible to have high accuracy with respect to a target object. Indeed, as the object is perceived, the control is only stopped when the desired accuracy is achieved. However, the MP is not observed and its pose must be estimated. The contributions of this thesis are the following. Three moving-platform pose estimation methods were proposed and evaluated on different tasks. It was found that estimation by control integration is the most versatile. Thorough Lyapunov stability analysis was performed on a planar and a spatial CDPRs. A link between the MP pose and system stability was determined and thus a novel workspace named Control Stability Workspace was defined. It was computed for several VS approaches on multiple CDPRs. The impact of different perturbations and modeling errors was evaluated. In experimental validation it was shown that CDPR accuracy always remains the same as long as the system is stable. Perturbations in the system affect only the trajectory to the goal. It was shown that trajectory tracking greatly improves CDPR behavior despite the perturbations. Finally, to deal with cable slackness, a Tension Correction Algorithm for VS was proposed and validated.
Document type :
Complete list of metadata

Contributor : Abes Star :  Contact
Submitted on : Tuesday, June 29, 2021 - 5:15:11 PM
Last modification on : Wednesday, April 27, 2022 - 3:51:25 AM
Long-term archiving on: : Thursday, September 30, 2021 - 7:18:54 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03274078, version 1


Zane Zake. Design and stability analysis of visual servoing on cable-driven parallel robots for accuracy improvement. Automatic. École centrale de Nantes, 2021. English. ⟨NNT : 2021ECDN0010⟩. ⟨tel-03274078⟩



Record views


Files downloads