兩輪式自我平衡車之控制器設計與實現

Abstract

本論文主要是研究PID控制器、LQR控制器與滑動控制器應用於兩輪移動車之平衡上。由於兩輪移動車本身為不穩定系統，需要外加控制機制才能使其平衡。本論文針對一實體兩輪移動車，規劃與實現相關控制電路，並將PID控制器、LQR控制器與滑動控制器實現於微控制器，使兩輪移動車即使有外界干擾產生，亦可有效達成平衡。首先使用三軸加速度計與陀螺儀感測車身目前所處的角度與角速度，使用其角度誤差值與角速度誤差，透過PID控制器、LQR控制器與滑動控制器輸出適當的PWM Duty，並驅動馬達轉動使得車身平衡。最後比較其LQR控制器、PID控制器與滑動控制器的平衡效率與抗外部干擾之效能。並使用自行創意設計的手部感測方法，使得兩輪移動車的行駛效能更加全面，而兩輪移動車的實驗證明了我們所提出的方法的是可行的。

This thesis is to study the balance of two-wheeled robots by LQR (Linear Quadratic Regulator) control, PID (Proportional Integral Derivative) control and sliding mode control. A two-wheeled robot needs extra control mechanism to reach the balance because the system itself is unstable. This study implements the related control circuit to a real two-wheeled robot. By realizing the LQR control, PID control and sliding mode control into a microcontroller, the two-wheeled robot can still keep the balance even when the external interference is added. At first, by using the three-axis accelerometer and gyroscope to measure the angle and angular velocity of the body, the angle error and angular velocity error are generated. Then, the LQR algorithm, the PID algorithm and sliding mode algorithm generate the appropriate PWM (Pulse Width Modulation) duty to drive the motors such that the two-wheeled robot reaches the balance. Finally, we compare the balance efficiency and the efficiency of resisting the external interference for the LQR control, PID control and sliding mode control. And using a self-designed hand sensing method makes the driving performance more comprehensive. The experimental results of the two-wheeled robot show that the method proposed in this thesis is feasible.