Path Tracking and Position Control of Nonholonomic Differential Drive Wheeled Mobile Robot

Abstract

Differential drive wheeled mobile robot (DDWMR) is one example of a robot with a constrained movement, Multiple Input Multiple Output (MIMO), and nonlinear system. Designing a low resource position and heading controller using linear MIMO methods such as LQR became a problem because of the linearization of robot dynamics at zero value. One of the solutions is to design a MIMO controller using a Single Input Single Output (SISO) controller. This work design a controller using PID for DDWMR Jetbot and selects the best feedback gain using different scenarios. The designed controller manipulates both motors by using calculated control signal to achieve a complex task such as path tracking with robot position in x-Axis, y-Axis, and heading angle as the feedback. The priority between position and heading angle can be adjusted by changing three feedback gains. The controller was tested, and the best gain was selected using Integral Absolute Error (IAE) metrics in a path tracking task with four different path shapes. The proposed methods can track square, circle, and two types of infinity shape paths, with the less well-formed shape being the four edges square path.

An Efficient Approach for Line-Following Automated Guided Vehicles Based on Fuzzy Inference Mechanism

“…With its popularity and efficiency, the PID (Proportional Integral Derivative) controller has been widely applied to linefollowing AGV studies [1][2][3][4][5]. In addition, control methods such as MSE controller (mean-square error controller) [6], visual controller [7], fuzzy controller [8][9][10][11], neural network [12], rotary controller [13], sliding mode controller [14,15] and adaptive controller [16,17] have also been used for AGV navigation.…”