Improved Computational Speed of System Dynamics for Cable-Driven Robots through Generalised Model Compilation
Refereed conference paper presented and published in conference proceedings

Full Text

Other information
AbstractThe study of cable-driven parallel robots (CDPRs) has motivated numerous applications in recent years. Using a numerical CDPR modelling approach, the Cable-robot Analysis and Simulation Platform for Research (CASPR) allows simulations and analyses to be performed on arbitrary types of CDPRs. However, the inadequate computational speed of system dynamics hinders the use of CASPR in fast analyses and real-time control on CDPRs. In this paper, a generalised model compilation method is presented to improve the computational speed of system dynamics. By generating symbolic expressions of the system dynamics and compiling the variables into closed-form functions, improvements from 11 to 23 times in computational speed are achieved. Case studies in forward kinematics, inverse dynamics and workspace analysis are performed to demonstrate the benefits of the compilation, and the potential for extending the method to other robotic systems.
Acceptance Date19/03/2018
All Author(s) ListYin Pok Chan, Ghasem Abbasnejad, Jonathan Eden, Darwin Tat Ming Lau
Name of Conference2018 IEEE International Conference on Real-time Computing and Robotics
Start Date of Conference01/08/2018
End Date of Conference05/08/2018
Place of ConferenceKandima
Country/Region of ConferenceMaldives
LanguagesEnglish-United States

Last updated on 2018-03-12 at 10:35