Design and control of an automated SCARA robotic arm with a pneumatic soft gripper

Abstract

The paper focuses on the design, analysis and control of an automated 3 DOF SCARA robotic manipulator with an end effector made of asymmetric flexible pneumatic bellow actuator (AFPBA). The manipulator is made for use in the poultry industry and therefore tested in its ability to detect, pick and place eggs. The links of the manipulator are made using acrylonitrile butadiene styrene (ABS) and the end effector is made using nitrile rubber. A neural network derived from the VGG-16 and YOLO architecture is then implemented to detect and localize eggs. The predicted values were then used to calculate the inverse kinematics of the manipulator.