Comprehensive simulation of cooperative robotic system for advanced composite manufacturing : a case study

Yang, Manman and Yu, Leijian and Wong, Cuebong and Mineo, Carmelo and Yang, Erfu and Bomphray, Iain and Huang, Ruoyu and Brady, Scott; Shafik, Mahmoud and Case, Keith, eds. (2021) Comprehensive simulation of cooperative robotic system for advanced composite manufacturing : a case study. In: IOS Press. Advances in Transdisciplinary Engineering, 15 (Advanc). IOS Press, pp. 105-110. ISBN 9781643681993 (https://doi.org/10.3233/ATDE210021)

[thumbnail of Yang-etal-IOS-2021-Comprehensive-simulation-of-cooperative-robotic-system-for-advanced-composite-manufacturing]
Preview
 Text. Filename: Yang_etal_IOS_2021_Comprehensive_simulation_of_cooperative_robotic_system_for_advanced_composite_manufacturing.pdf
Final Published Version
License: Creative Commons Attribution-NonCommercial 4.0 logo
Download (352kB)| Preview

Abstract

Composite materials are widely used because of their light weight and high strength properties. They are typically made up of multi-directional layers of high strength fibres, connected by a resin. The manufacturing of composite parts is complex, time-consuming and prone to errors. This work investigates the use of robotics in the field of composite material manufacturing, which has not been well investigated to date (particularly in simulation). Effective autonomous material transportation, accurate localization and limited material deformation during robotic grasping are required for optimum placement and lay-up. In this paper, a simulation of a proposed cooperative robotic system, which integrates an autonomous mobile robot with a fixed-base manipulator, is presented. An approach based on machine vision is adopted to accurately track the position and orientation of the fibre plies. A simulation platform with a built-in physics engine is used to simulate material deformation under gravity and external forces. This allows realistic simulation of robotic manipulation for raw materials. The results demonstrate promising features of the proposed system. A root mean square error of 9.00 mm for the estimation of the raw material position and 0.05 degrees for the fibre orientation detection encourages further research for developing the proposed robotic manufacturing system.

CORE (COnnecting REpositories)