Vehicle Dynamic State Estimation Using Smartphone Embedded Sensors

The access to the information on the vehicle motion state is of great significance for the vehicle stability control and the development of active safety products. However, the vehicle state parameter extraction is primarily accessed by attaching special sensors to the vehicles, which usually requires modification of redesign of the vehicle with high cost. Smartphone integrate gyro, orientation sensor, GPS and some other sensors thus providing a new way for vehicle dynamic parameter estimation. Therefore, we choose smartphone as our working platform, and data acquisition is fulfilled on a variety of mobile sensors embedded in an Android platform phone. Combining the features of these sensors and the properties of vehicle kinematics, a Kalman filter based sensor fusion approach is proposed to perform the vehicle state estimation. The parameters of vehicle heading angle and sideslip angle are extracted using fusion of data from the gyro, GPS and orientation sensor. Experiments carried on real vehicle show that the estimation results generated by fusing the gyro and orientation sensor are better than that of the gyro and GPS, but the two fusion approaches can complement each other in different contexts used. The main contribution of our work is that we provide a new attempt for accessing the vehicle dynamic parameters using off–the-shelf sensors. Copyright © 2013 IFSA.