Design and implementation of advanced auto calibrating line following sensor for coloured surfaces with a white line

Abstract

This paper discusses the design and implementation of a 7 channel line sensor for line following robot having advanced functionalities such as software based automatic sensor calibration and individual threshold per sensor channel. Unlike traditional line sensors which work on black surfaces with a white line or vice versa, this sensor is designed to differentiate white line on any coloured surface. Thus this sensor can be implemented in robotics applications or in industries where different section's floors are assigned with different colours and robot navigates using line following. The system was implemented using light emitting diodes, phototransistor and an onboard microcontroller Arduino Mega 2560 which communicates with any navigation control system using serial communication. Moving average filters are implemented per channel to remove the fluctuations in the readings due to vibration of the sensor during locomotion. It then gives the error feedback or the offset of the white line from the centre, to the system that is corrected using Proportional-Integral-Derivative algorithm. It also takes care of non scaled readings of the line sensors due to ambient light by having separate threshold values for individual sensors making each sensor independent. The main aim of this paper is to highlight the use and need of a line following sensor capable of differentiating any background colours with white line and at different light conditions. Performance metrics were measured and compared to show tradeoffs between cost and performance.