Analyze and Control 2-Wheeler Auto Balance System

Goh, Keat Chun (2020) Analyze and Control 2-Wheeler Auto Balance System. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Goh Keat Chun Full Text.pdf Restricted to Registered users only Download (8MB)

Abstract

Auto Balancing System is available in the industry for 4-wheeler and above. However, the same piece of technology is absent in 2-wheeler. As a typical non-linear and dynamic balancing system, the 2-wheeler self-balancing vehicle has become more and more popular. It is because the design for 2-wheeler is small size, energy-saving compared to 4-wheeler and above and had a flexible rotation. However, current control algorithms concerning 2-wheeler self-balance vehicles are primarily suitable for balance vehicles under manual real-time control instead of full-auto vehicles. To achieve auto-balancing system, the relationship between DOF need to be understand how it works and how those parameters will affect each other, in other words, the parameters revolving around the application need to be considered. In this project, those parameters are steering angle, roll angle, pitch angle and yaw angle. To determine the relationship between those parameters, straight road test and bumper test experiment will be conducted. Next, sensors such as gyroscope and accelerometer will mount on the 2-wheeler to collect parameters data. The data collected will further perform noise reduction in MATLAB filtering function to improve the data quality. In MATLAB filtering function, three filtering method such as low pass filter, Butterworth low pass filter and Moving average filter are used to perform noise reduction. In addition, a mobile Kinetic Sensor Application is used to determine those parameters and will perform efficiency comparison between hardware gyro-sensor and mobile build in gyro sensor. The data collected from the experiment will be analyse and recorded down by using Arduino IDE, MATLAB and MATLAB-SIMULINK model. Last but not least, four-degree of freedom Linear half-car suspension system modelling will be implement in MATLAB-Simulink. Implementation of linear half-car suspension system is to observe the pitching motion that will affect the stabilization of the 2-wheeler. By the end of the thesis, the unnecessary DOF which cause a least effect in balancing a single-track vehicle is reduced, a complete data set was collected and analysed, one out of three filtering method will be use for the future improvement and future application development, efficiency comparison between hardware sensors and mobile software kinetic sensor and mathematical expression for pitch Modelling is implemented to further perform analysis.