A simple method to estimate the trajectory of a low cost mobile robotic platform using an IMU

摘要

In mobile robotics it is essential to know the location and orientation of the devices for proper operation of the algorithms. This task can be performed with different systems, such as GPS that sometimes may be inefficient indoors, or also with image-based systems using peripheral cameras. Inertial Measurement Units (IMU) with nine degrees of freedom are a fusion made up of accelerometers, gyroscopes and triaxial magnetometers and have been widely used in the location and orientation applications, with easy acquisition and low computational costs to process the obtained data. For the use in mobile robotics, they present an inexpensive alternative to the above-mentioned systems. In this paper we present an algorithm to determine the trajectory and orientation of a low-cost robotic platform that follows the track of a preset geometric figure using a custom built IMU. The algorithm enables the estimation of the size and the shape of the path. The data obtained from the IMU was processed with a 1D Kalman filter to remove the Gaussian noise from the data on each channel. The results of estimating the shape of trajectory actually represent the reference system and determining the shape dimensions results in an error percentage lower than 10?%.