PID-PSO DC motor position controller design for ankle rehabilitation system

• Main Author: Azizi, Muhammad Azizul Raziq
• Format: Thesis
• Language: English; English
• Published: 2023
• Subjects: T Technology (General); TJ Mechanical engineering and machinery
• Online Access: http://eprints.utem.edu.my/id/eprint/28277/; https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=123909

In rehabilitation system application, precise output responses are important for position control on the mechanism’s joint to avoid injury that occurs during physiotherapy. Hence, position control of DC motors has attracted considerable research with applied control system algorithms. This research aims to determine the mathematical modeling gain of the third-order transfer function for the DC motor that represents the features parameter of the Ankle Rehabilitation System. The transfer function is a model in Matlab software to validate the performance of the control system through simulation compared with real-time experiments. Next, the control algorithms are proposed to design and implement the Proportional-Integral-Derivative (PID) with Particle Swarm Optimization (PSO) controller technique for optimal Proportional (Kp), Integral (Ki) and Derivative (Kd) gains. The control algorithms also aim to be analyzed using an incremental rotary encoder sensor device as closed-loop feedback for dorsiflexion and plantarflexion movement. This rotary encoder sensor device converts rotary motion into electrical signals or pulse signals to count per revolution of the gearbox output shaft. The H-Bridge module is used for bi-directional motor control with pulse-width modulation (PWM) from the Arduino microcontroller. The control pulse-width modulation is calculated and realized by tuning the value of Proportional (Kp), Integral (Ki) and Derivative (Kd) with soft computing optimization techniques PSO controller. This proposed approach to develop optimal controller tuning parameters for proper computational performances position control efficiency and stable convergence characteristics. The simulation result of the PID-PSO controller with variables Kp= 6.542, Ki= 0.103 and Kd= 0.255 provide good performance with the rise time (TR) is 0.0659sec, settling time (Ts) is 0.1183sec and maintain the steady-state error with zero overshoot. This gain tuning of Kp, Ki and Kd from the simulation was also implemented in real-time hardware for validation producing effectiveness for the controller to improvise the Ankle Rehabilitation System position control analysis. The statistical trajectory tracking error is evaluated using mean square error (MSE) and root mean square error (RMSE) achieving a small value. The percentage improvement for simultion from PID controller to PID-PSO shows the MSE made is almost 91% while the RMSE is 71%. The real-time experiment performance also have high percentage improvement with MSE is 97% and RMSE is 84%. It concludes that the PID-PSO controller effectiveness control strategies of DC motor can accurately track the sinusoidal setpoint rotational angle movement of the Ankle Rehabilitation System.