Design of a Real Time Image Processing Based Industrial Pick and Place Robot with Path Learning

Abstract

Our project goal is to research, analyze, and build a robot that performs real-time image processing based on object selection and placement from one place to another. As we all know, the demand for industrial products is increasing day by day, so we developed this robot in such a way that it can work nonstop and more efficiently. Our robot is based on automation technology that is dynamically picking and putting things in their place automatically. It has advanced sensors that sort products by color and keep them on designated shelves or conveyor belts. This color detection smart camera sensor, called Pixy2 Cam, can work independently. It has a built-in microprocessor, and it works through a hue-based color filtering algorithm. This camera calculates the hue and saturation of each RGB pixel from the image sensor and uses these as a primary filtering parameter. This camera only needs to be trained with color through a software program called Pixymon. The detected object data was sent to the microcontroller (ATmega328p), and the microcontroller commanded the motor to move and followed the line through the IR sensor as programmed. The Arduino IDE software is used to program the microcontroller. The entire programming code is sent to the Arduino UNO microcontroller. Now a day’s traditional robotic arm can only pick and place the product only specific location. But our proposed robot can detect the product wherever it is. As a result, this technology has the potential to improve the efficiency and accuracy of the work.