Experimental Object Manipulation of Assistive Robotic Arm for Pick and PlaceTask

For people that need total or partial assistance to perform daily tasks, assistive robots are one of the solutions. Force control of these robots when interact with human or manipulate objects, is one the challenging problems in this area. In this paper a ROS-based force control system is implemented on a JACO assistive robot for grasping tasks. This robot is specifically designed for people with upper body disorders. However, the robot can be used for public use and used for specific tasks that require high precision. To do this, we need to know exactly how the robot's internal performance works and how it is structured. It is usually achieved by designing sophisticated control techniques that meet these criteria. Advanced control architectures such as torque computation control allow tracking of desired paths with high accuracy, however, the need to integrate robotic models remains. The work presented in this study provides a basis for applying these techniques to the JACO robotic arm. The calculation is based on the Euler-Lagrange method of calculating the internal energy. The results are then analyzed to ensure the models estimated with control schemes. Therefore, more advanced analysis and control techniques can be implemented on this robotic arm. Finally, this study can be controlled by PID with respect to the torque entered to the end effector by the object so that the robotic arm can move from the initial position to the secondary position with optimum capture and torque control of all robot joints. The experimental results showed the effectiveness of proposed method to perform grasping and manipulation scenario successfully.

کلیدواژه ها:

force control ، assistive robot ، GRASP ، JACO ، Dynamic modelling