Curve Path Tracking Method of Independent Two In-Wheel Motor Driven Vehicle

Abstract

This paper presents a novel curved path tracking method for the independent two in-wheel BLDC motor-driven surveillance vehicles. In the autonomous self-driving vehicle type robot, the absolute position and the relative position of the vehicle may be estimated by a GPS (Geometric Positioning System), which is a pre-fixed absolute position, and an installed position sensor. However, GPS cannot be used indoors, and when a location sensor is used, additional costs are incurred and the complexity of system increases. If these sensors are not used, the position of the vehicle depends on the distance traveled by the rotation of the motor. However, the estimated vehicle position based on the integrated motor position is not same as the actual position due to the irregularities of the driving path and the slipping of the wheels. Accordingly, a serious error occurs in the stop position of the vehicle after tracking the curved path in the target position according to the driving path condition. In order to solve this problem, a real-time position error correction method using a direction angle error estimated based on a simple IMU (Inertial Measurement Unit) is introduced. The proposed method calculates the required real-time azimuth of the vehicle on a predetermined driving path. Then, the real-time position error during the driving route can be compensated for as a directional angle error between the required directional angle and the directional angle estimated by the IMU sensor. The performance of the proposed method was verified through experiments.