Intelligent Vehicle Pose Estimation Based on Kalman Filtering Algorithm

In the intelligent vehicle positioning process, the ultrasonic indoor positioning system is susceptible to the surrounding, environment and thus leads to an inaccurate result. To solve this problem, in this paper, the key positioning information is extracted by using the self-developed serial communication software. 'then, the corresponding pose estimation algorithm is designed according to the structural characteristics of the intelligent vehicle and the delay effect of the positioning system. Finally, the Kalman filter is used to optimize the positioning data. In terms of experimental design, firstly, the validity of the filtering algorithm is determined by comparing the static and dynamic positioning effects before and after the optimization of the positioning data. Then, the paired position estimation of the intelligent vehicle is carried out with the optimized positioning data and compared with the experimental results before filtering. The experimental results show that the localization optimization algorithm and pose estimation algorithm described in this paper can solve the problem that the positioning of the intelligent vehicle is not accurate and the vehicle body is not coherent in the pose estimation process.