Research Article
Genetic Algorithm Tuned Super Twisted Sliding Mode Controller (STSMC) for Self-balancing Control of a Two-Wheel Electric Scooter
@INPROCEEDINGS{10.1007/978-3-030-93709-6_18, author={Tefera T. Yetayew and Daniel G. Tesfaye}, title={Genetic Algorithm Tuned Super Twisted Sliding Mode Controller (STSMC) for Self-balancing Control of a Two-Wheel Electric Scooter}, proceedings={Advances of Science and Technology. 9th EAI International Conference, ICAST 2021, Hybrid Event, Bahir Dar, Ethiopia, August 27--29, 2021, Proceedings, Part I}, proceedings_a={ICAST}, year={2022}, month={1}, keywords={Scooter Genetic algorithm STSMC and PID controller}, doi={10.1007/978-3-030-93709-6_18} }- Tefera T. Yetayew
Daniel G. Tesfaye
Year: 2022
Genetic Algorithm Tuned Super Twisted Sliding Mode Controller (STSMC) for Self-balancing Control of a Two-Wheel Electric Scooter
ICAST
Springer
DOI: 10.1007/978-3-030-93709-6_18
Abstract
Two wheel self-balancing electric scooter is based on inverted pendulum system and this system is a nonlinear and unstable. An inertial measurement unit (IMU) which is combination of accelerometer and gyroscope measurement is used in order to estimate and obtain the tilt angle of the scooter. Super twisted sliding mode controller (STSMC) is applied to correct the error between the desired set point and the actual tilt angle and adjust the brushless direct current (BLDC) motor speed accordingly to balance the scooter, when scooter is tilted forward, motor is move forward to catch up in order to balance the scooter and proportional integral derivative (PID) controller is used to control direction of scooter that means to turn left or right. The STSMC parameters and PID parameters are tuned using genetic algorithm (GA) and controllers performance evaluation is done using MATLAB/Simulink. The pitch and yaw angle with changes in magnitude of 0.1 rad and zero reference angle, almost the steady state error are 7.965 × 10–08and 5.677 × 10–07respectively for both controllers tuned by GA. GA tuned controllers are compared with analytically tuned controlled for initial pitch angle of 0.3 rad. The magnitude of steady-state errors at time 2 s are 7.71 × 10–07and 0.004648 respectively, which is an indication of parameters tuned using global optimization algorithms, in this case GA are more optimal than analytically tuned parameters.
