Cable interference control in physical interaction for cable-driven parallel mechanisms. (February 2019)

• Record Type: Journal Article
• Authors: Meziane, Ramy; Cardou, Philippe; Otis, Martin J.-D.
• Is Part Of: Mechanism and machine theory. Volume 132(2019)
• Page Start: 30
• Page End: 47
• Publication Date: 2019-02
• Subjects: Cable driven parallel mechanism (CDPM) -- Cable interferences -- Physical human-robot interaction (PHRI) -- Karush-Kuhn-Tucker conditions (KKT) -- Admittance control -- Force control; Machine theory -- Periodicals; Machinery -- Periodicals; Machines -- Périodiques; Génie mécanique -- Périodiques; Machine theory; Machinery; Periodicals; 621.81
• DOI: 10.1016/j.mechmachtheory.2018.10.002
• Languages: English
• ISSNs: 0094-114X

Abstract

Highlights: The paper considers a new control strategy to prevent cables from crossing and folding in human-CDPM physical interaction. Our method proposes: The management of interferences between cables until the limit of the mechanism kinetostatic capabilities. An admittance control for physical interaction The use of a virtual force computation based on a gradient vector and a logistic function. The proposition of a new solution for minimum distance determination between cables using Karush-Kuhn-Tucker conditions. Abstract: Cable interferences and collisions can lead to unpredictable behavior when a human physically interacts with a cable-driven parallel mechanism through its mobile platform. This paper presents an interactive control approach to prevent two cables in interference from folding onto one another, and thus preserve the cable-mechanism geometry. In this approach, the controller generates a repulsive force to prevent the cables from crossing. Therefore, the task is executed within the cable-driven parallel mechanism's geometric limits. The repulsive force applied by the controller is derived from the gradient of the minimum distance between any pair of cables of the parallel mechanism. In turn, this minimum distance between cables is computed from the Karush-Kuhn-Tucker conditions of the associated optimization problem. The approach was tested and validated on a parallel mechanism driven by seven cables.