Dexterous robotic finger with joint mobility modulation for adaptive grasping

Abstract

Adaptive grasping is an essential feature in robotic hands for improving dexterity. The paper proposes a novel approach for enhancing dexterity in hand prostheses by implementing joint-mobility modulation in a robotic finger which uses an underactuated tendon-driven mechanism. The proposed finger design includes a locking mechanism for mobility modulation, enabling both precision and power grasping modes. A mathematical model was developed to calculate design parameters of the robotic finger. Physical experiments were conducted to compare joint kinematics between the human and robotic fingers for selected grasping patterns. Results show that the proposed mechanism can produce said grasping patterns using controlled mobility restriction of joints. The motion trajectory of the robotic finger has an average correlation of 89% with the human finger. There is potential for the proposed mechanism to be used in developing a dexterous robotic hand.