Mobile Robot Localization in Indoor Environments Using Fuzzy Adaptive Unscented Kalman Filter and Random Tree Routing Algorithm with Fast Exploration

In the field of mobile robot navigation, challenges such as nonlinear conditions, uncertainties, and the development ‎of advanced methods have necessitated accurate position estimation. In this paper, fuzzy based adaptive ‎unscented Kalman filter (FAUKF) is evaluted to enhance the state estimation performance for mobile robot ‎localization. In proposed method, the FAUKF algorithm effectively removes the noise uncertainty by adaptively ‎adjusting the covariance of the measurement noise according to the adaptation law. The output of the Mamdani ‎Fuzzy Inference System (FIS) acts as an observer and improves the matching law. The results of this research show ‎the accuracy of FAUKF algorithm position estimation compared to traditional UKF methods. Also, this study ‎presents a pioneering navigation strategy for mobile robots, which integrates random tree routing algorithm with ‎rapid exploration (RRT*) for optimal path design in indoor environments. The goal of RRT* integration is to create ‎optimal routes taking into account safety and environmental constraints. By combining the Kalman filter ‎prediction and updating steps, this method reduces the accumulation of uncertainty during the localization process ‎and facilitates accurate localization and path planning towards the target.‎The simulation results confirm the effectiveness of this method in keeping the uncertainty levels in localization ‎constant over time. The presented adaptive method enables efficient navigation in complex environments. Path ‎planning is a critical element in robotics applications and the RRT* based approach presented in this paper provides ‎a comprehensive solution to create optimal and efficient paths.

کلیدواژه ها:

Fuzzy ، ‎ Random tree routing ‎ Robot localization ، ‎ Unscented Kalman filter