Tracking Fast-Maneuvering Objects with IMM and a Three-Coordinate Radar
Maciej Sierżęga, Jakub Miler, Hubert Boratyński, Mateusz Głodowski, Kamil Jankowski, Wojciech Strehlau, Piotr Kaczmarek · 2025
The use of algorithms for tracking moving objects in 3D space is an important aspect of applications related to autonomous vehicles, safety, and navigation. Popular methods for tracking moving objects are based on Kalman filtering. In cases where the type of object movement may change over time, algorithms such as IMM (Interacting Multiple Model), incorporating multiple Kalman filters, each based on a different motion model, are employed. This paper evaluates the performance of the IMM algorithm in estimating an object's position using data from a 3D radar. The results show the viability of this algorithm in real scenarios.