Active compliance control strategies for multifingered robot hand

• المؤلف الرئيسي: Sadun, Amirul Shafiq
• التنسيق: أطروحة
• اللغة: الإنجليزية; الإنجليزية; الإنجليزية
• منشور في: 2018
• الموضوعات: TJ212-225 Control engineering systems. Automatic machinery (General)
• الوصول للمادة أونلاين: http://eprints.uthm.edu.my/160/

Safety issues have to be enhanced when the robot hand is grasping objects of different shapes, sizes and stiffness. The inability to control the grasping force and finger stiffness can lead to unsafe grasping environment. Although many researches have been conducted to resolve the grasping issues, particularly for the object with different shape, size and stiffness, the grasping control still requires further improvement. Hence, the primary aim of this work is to assess and improve the safety of the robot hand. One of the methods that allows a safe grasping is by employing an active compliance control via the force and impedance control. The implementation of force control considers the proportional–integral–derivative (PID) controller. Meanwhile, the implementation of impedance control employs the integral slidingmode controller (ISMC) and adaptive controller. A series of experiments and simulations is used to demonstrate the fundamental principles of robot grasping. Objects with different shape, size and stiffness are tested using a 3-Finger Adaptive Robot Gripper. The work introduces the Modbus remote terminal unit [RTU] protocol, a low-cost force sensor and the Arduino IO Package for a real-time hardware setup. It is found that, the results of the force control via PID controller are feasible to maintain the grasped object at certain positions, depending on the desired grasping force (i.e., 1N and 8N). Meanwhile, the implementation of impedance control via ISMC and adaptive controller yields multiple stiffness levels for the robot fingers and able to reduce collision between the fingers and the object. However, it was found that the adaptive controller produces better impedance control results as compared to the ISMC, with a 33% efficiency improvement. This work lays important foundations for long-term related research, particularly in the field of active compliance control that can be beneficial to human–robot interaction (HRI).