Robot workspace estimation and base placement optimisation techniques for the conversion of conventional work cells into autonomous flexible manufacturing systems

Abstract

Robot arms are used in modern work cells and flexible manufacturing systems (FMS) in handling work pieces and loading/unloading processes. The robot arm links may interfere with the bodies of the system components, which are considered as obstacles in the robot workspace. In order that the robot works in safe conditions, study of robot workspace in a free space and in the presence of obstacles should be investigated. The inverse problem of having an existing working space occupied with a number of CNC machines to which it is supposed to introduce a robot to serve is an interesting problem. This is a very common problem when trying to convert conventional work cells into autonomous systems. There are two main questions when studying this kind of problem: what type of robots is suitable to satisfy the existing working space? And where to place the base of this robot to efficiently serve the existing machines? The main objective of this article is trying to answer these two questions. A computational algorithm is developed to estimate the robot workspace. The optimisation of robot base placement is achieved using genetic algorithms. A comparative study of the suitability of different robots for a specified working area is also included. Finally, robot movement visualisation within a pre-defined FMS using solid edge modelling is presented to verify the proposed algorithm and simulate the robot path within the work cell.