State Estimation for a Unicycle Robot Using Multi-Rate Sensor Fusion: A Comparative Study

Autonomous systems, including unicycle robots, have attracted considerable interest across multiple domains, encompassing transportation, surveillance, and search-and-rescue missions. Achieving accurate state estimation is therefore crucial for the reliable and effective operation of these autonomous systems. This paper presents a comparative study on state estimation techniques for a unicycle robot using multi-rate sensor fusion. This work encompasses the evaluation of the Kalman Filter, Unscented Kalman Filter, Extended Kalman Filter, Set-Valued Kalman Filter, and Particle Filter to estimate the position, velocity, and orientation of a unicycle robot. The performance of these filtering techniques is compared for accuracy, computational efficiency, and robustness in handling sensor measurements with different rates. The outcomes illustrate both the strengths and constraints inherent in each filtering algorithm, offering valuable perspectives on their appropriateness for performing state estimation in the context of unicycle robots, thus aiding researchers and developers in selecting the optimal method based on their specific requirements and constraints.