Development of a Microcontroller-based Wireless Writing Robotic Arm Controlled by Skeletal Tracking

Abstract

This study aims to develop a microcontroller-based robotic arm that is controlled by motion capturing camera which is capable of mapping the skeletal layout of the person and track its movements. The robotic arm aims to encourage students to learn how to write by tapping into their natural curiosity which aims to develop their muscle memory the longer they used the system. The system is composed of two parts, the hardware and software components. The hardware component involves assembling the robotic arm to suit the application. The software component is further subdivided into two parts, namely the C# code and the Arduino C code. The C# code is the one responsible for analyzing the input from the camera and computing for the data needed to control the robotic arm while the Arduino C code is the one responsible of reading the data computed by the C# code before processing it into microcontroller readable data to make the motors of the robotic arm. The researchers performed various test such as checking whether the robotic arm follows the desired 1:1 translation ratio of the human movement to robotic arm movement, and checking the camera's processed input to check the accuracy of the system. Through testing the researchers have found the system to be accurate having an average accuracy of 94.46%.