Using Artificial Neural Networks for Closed Loop Control of a Hydraulic Prosthesis for a Human Elbow

We address control of a hydraulic prosthesis for human elbow, a problem in which it is essential to obtain quick simulation results to appreciate the system dynamic. The forward kinematics problem for a prosthesis developed at Politecnico di Bari is solved using artificial neural networks as an effective and simple method to solve the problem in real time and limit computations. We show the method effectiveness designing two PID regulators that control the arm thanks to the neural computation of the forward kinematics.