SALIENT OBJECT FOLLOWING SELF BALANCING ROBOT

Abstract

Robotic mobility technology have evolved exponentially over the past few years and not only do they find use in military and security sector, they are also fast gaining popularity in industry and consumer products. A number of techniques have been proposed to increase robotic mobility in dynamic environments. One such widely used technique based on the inverted pendulum model is used to provide greater mobility to a robotic platform. The instability of inverted pendulum systems has always been an ideal test bed for control theory experimentation. The presented document will display the procedures involved in balancing an unstable robotic platform with the aim to design a complete discrete digital control system that will provide the stability required by the system for optimum performance. The platform will be an ideal test bed for the implementations of both PID digital control and Kalman filter algorithms. Both these algorithms will provide the imperative control for the system. Therefore the presented project will examine the performance of both PID digital control and Kalman filter algorithms. PID control algorithm is used to offer system stability whereas Kalman filter, an estimator, provides fused data of the sensors (Accelerometer and gyroscope). The digital filter provides a more reliable sensor data used to calculate the tilt angle of the robot. To collect performance results for both the PID controller and Kalman filter, Test software was written. The control system performance directly depends on Kalman filter and PID controller input parameters and the outcome clearly shows how the adjustable parameters on the control system directly impacted the overall system performance. The results also indicate the performance and the need of the Kalman filter to remove sensor noise.The nearly reliable sensor data increases PID controller performance to drive the robotic platform to vertical equilibrium. Further to this, the raw noisy sensor data was compared against the accumulated results for the Kalman filter. The plots for this comparison are shown in the Kalman filter results section. PID controller output response data was also collected and plotted. The PID output response results were used in the controller tuning process.