Unscented Kalman Filter for Estimation of Vehicle Velocity in Real Time

Vehicle velocity estimation systems can be utilized in a variety of automotive applications. Focusing Primarily on practicality, this paper presents a new method for estimating vehicle velocity in real time using low-cost sensor fusion combining a global positioning system (GPS) and an inertial navigation system (INS). In this paper, an unscented Kalman filter (UKF) with excellent estimation accuracy and robustness against model nonlinearity is developed for vehicle velocity estimation. It is a practical solution that can be easily implemented in mass-produced vehicles due to its high availability, high estimation accuracy and high robustness against model nonlinearity. The estimation performance of the proposed UKF is verified through experimental results using a test vehicle. Finally, the effectiveness of the proposed estimation algorithm can be confirmed through a comparative study.