DEVELOPMENT OF AN INTELLIGENT ROBOTIC GRIPPER FOR OBJECT MANIPUTATION USING ADAPTIVE CONTROLLER

Subject Area: Control System Engineering

This study presents development of an intelligent robot gripper for précised object manipulation using adaptive controller. The aim was to counter the impact of slip and position displacement constraints on object during grabbing, thus making the gripper inefficient. The methodology employed is the experimental and simulation approach. The methods used are experimental investigation of gripper performance developed with Dahl controller, data collection at Robotics and Artificial Intelligence in Nigeria (R.A.I.N) company, development of the gripper position algorithm using the output from proximity sensor, modelling of the adaptive controller using machine learning, development of the intelligent gripper system. The system was implemented with Simulink and tested. The performance of the adaptive controller when examined with Mean square error (MSE) was 1.094e-10Mu; the Regression (R) performance was 1. The implication was that the error which occurred during the training of the adaptive controller was tolerable and the controller based on the R values showed it can detect position changes and apply necessary force to hold the object. The step response of the adaptive controller is 10ms as against 45ms in the characterized Dahl controller, thus giving a percentage improvement of 23.33%. The adaptive controller when implemented on the gripper system and evaluated showed that it was able to apply the desired control force necessary for object grabbing and manipulation.