Design and Implementation of a Delta-type Parallel Robot with Artificial Vision to for object tracking (#858)

Authors

Quispe Ccachuco, Marcelo Jaime

Castillo Cáceres, Cesar Pio

Llerena Sánchez, Julio César

Abstract

In the industry today it is very unusual to find that a robot performs tracking operations with the help of a camera applying artificial vision, however, there are countless applications where the combination of these concepts would be fabulous, such as in the arms industry, production, selection processes, etc. The present project consists of achieving the union of a delta-type parallel robot with an object recognition and tracking interface, which will be carried out with a camera with a resolution of 640 x 480 pixels, which will be more than enough for our application. Artificial vision will have a program that will allow us to recognize objects and discriminate them by color and size. It should be noted that, if greater precision is desired regarding the position of our object, it is possible to change the camera to one with higher resolution, since we will use the pixels as coordinate points in the camera plane. The equations that govern the kinematics of the entire system were calculated, these equations were implemented in an Arduino program, which receives data from the coordinate points in ASCII format through the serial port of the computer at a frequency of 1200 bauds. a Python program made in Visual Studio, with the aim of calculating at high speed the rotation angles of each servomotor to position our robot at the desired point. To validate the design, simulations of the forces present in each component and the torque necessary to move the joints, the robot's work area, tracking tests, connectivity tests and data transmission speed were carried out.