Optimisation of Autonomous Threaded Fastenings Based on Non-Linear Least Square Method With GUIs

Abstract

Threaded fastenings are a common assembly method, accounting for over a quarter of all assembly operations. They are especially popular because they permit easy disassembly for maintenance, repair, relocation and recycling. Screw insertions are typically carried out manually as it is a difficult operation to automate. There is very little published research on automating threaded fastenings, and most research on automated assembly focuses on the peg-in-hole assembly problem. Non-linear least square method was designed and employed to identify torque signature signals during online threaded fastening. Creating interactive simulations and graphical user interfaces became necessary as a visualization aid. This provides help and support for the user, allowing them to concentrate on the concept they are illustrating and to put emphasis on the monitoring process rather than the mechanics of running the program. This paper presents a Graphical User Interface (GUI) tool to accommodate and support threaded fastening operations used in assembly line industries. This tool was produced as interactive software with a convenient GUI in combination with the computing and graphics capability of MATLAB. It has applied to automated monitoring of threaded fastenings based-on an analytical model and on-line parameter estimation. The monitoring problem deals with predicting the integrity of the screw insertion process based on the torque vs. insertion angle curves generated during the insertions. A Non-linear Least Square Method (NLSM) is applied for estimation of four unknown parameters during a self-tapping screw insertion to be presented. It is shown that these parameters, required by the model, can be reliably estimated on-line. Experimental results are presented to validate the estimation procedure.