Algorithm for Human-Robot Collaborative Tasks

Phoon, Zee Him (2018) Algorithm for Human-Robot Collaborative Tasks. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Phoon Zee Him - Final Thesis.pdf Restricted to Registered users only Download (2MB)

Abstract

A typical industrial robot has high precision, repeatability, and payload characteristics and has better performance doing repetitive and dangerous task compare to a human who may make mistake and stress or fatigue easily. But the robot cannot deal with the unexpected event in the real world, but, the human is able to handle it intelligently. Hence, Human-Robot Collaboration (HRC) becomes a new trend, it can be implementing in the application of medical, assembly, palletizing and even more. The Human-Robot Collaboration technology allows robot works with a human in the close distance without conventional guarding needed to surround the robot such as fencing and can be work in more flexible than a typical industrial robot. Many researchers are putting great efforts and studied on this Human-Robot Collaboration technology especially the control techniques and algorithms because it potentially brings some commercial values and great usage in near future. The most crucial thing that influenced the collaborative robot performance is the program algorithm that is to determine the structure of robot behaviour to perform in the different type of the applications. In this project, the number of the alternative program algorithms have been developed that is collaborative robot algorithm and the speed and separation monitoring algorithm. The collaborative robot algorithm is mainly executed picking and placed task and using the special action to ask for help from human if necessary. The speed and separation monitoring algorithm is a safety algorithm to fulfil the ISO standard (ISO 10218-1, 5.10.4, ISO/TS 15066) that the operator’s position and robot’s position and speed are being supervised to prevent the physical contact between human and moving robot. Those are the main algorithms are determine the robot behaviour structure, also, with a supporting algorithm such as the vision algorithms involves colour tracking algorithm, pixel to centimeter (cm) conversion algorithm and more that used of android phone rear camera to capture input data, and the inverse kinematics algorithm based on the trigonometric mathematical principal including the law of cosine and sine to determine each of the robot axis angles needed to reach to the object position after receive the 2-dimensional coordinates (X, Y) input data. The 5 degrees-of-freedom (DOF) miniature articulated robotic arm is designed and built with the aid of 3D CAD design software and additive manufacturing, then assembly with essential electronic devices such as Arduino Mega 2560 microcontroller, SC08A servo controller and more. The series of the experiments had been conducted in this project to verify the robot performance based on the developed program algorithm that mentioned above.