A self-driving car based on a combination of Median and Kalman filters

A Kalman filter is proved to be an effective tool for data processing. It was implemented widely over the last decade, mostly in robotics. This paper presents an application of sensor fusion to predict distance from a self-driving car to another object. We propose a method that uses a combination of sensors for distance estimation, which also uses a Kalman filter implementation to increase the efficiency of distance estimation. Multiple Lidars and a camera contribute to the improvement of data preprocessing for the Kalman filter. This application will help develop an advanced driver-assistance system such as a collision warning system, vehicle velocity calculation, and an advanced emergency braking system.