Bond Graph Modeling and LQR Control Synthesis of a Robotic Digit in Human Impaired Hand for Anthropomorphic Coordination

摘要

The voluntary movement of a human hand is a complex mechanism due to the physiological extensor and flexor movements. Anthropomorphic hands can have a wide range of applications. These applications include but are not limited to bionic hands in active prostheses, assisting surgical operations that are free from involuntary movements such as tremors and shudders and robotic grippers for remote applications. A partially impaired human hand coordination scheme involving a combination of natural and robotic digits is proposed in this paper. This research presents five digits, 21 degrees of freedom (DOF) graphical model of a human hand, where the index finger is impaired and therefore replaced with a robotic BeBionic finger. The angular displacement of the tip of the middle finger is controlled through a feedback force applied by the LQR (linear quadratic regulator) observer-based controller. The simulation result shows the control of the movement of robotic fingers of an impaired hand and the movement coordination with the central nervous system (CNS) and the natural digits. This demonstrates the efficiency of the scheme towards active prosthetics, exoskeleton and upper limb rehabilitation.