Biotechnical system based on the wrist joint control system tuning via genetic algorithm

The article considers the control system of a biotechnical system on the example of a wrist joint. A mathematical model of the control system is given, including feedback loops and PI controllers of the errors in angles and angular velocities errors of the wrist joint in the adduction –– abduction and flexion – extension planes. The proposed mathematical model also includes muscle dynamics block and the cross-coupling connections due to the nonorthogonality of motion in the adduction – abduction and flexion – extension plane caused by the nonorthogonal localization of the force application in the wrist joint. The delays that might be caused by the neurological processing and signa transmission are also considered. The PI controllers are tuned using the optimization procedure via genetic algorithm. The minimization criterion is based on the desired transient response characteristics such as step response time, steady-state error and overshoot. The steady-state error minimization in a mandatory criterion but others are optional that may be related to different considered biotechnical system objective. Computational experiments are performed in which the tuning of PI controllers for the proposed system is implemented for various configurations corresponding to various criteria in the optimization problem. The results of these experiments showed the applicability of the considered approach to tuning of the control system of a biotechnical system using optimization procedures via a genetic algorithm as a numerical optimization method.