Multi-Center Collaborative Localization Method Based on Extended Kalman Filter in Dynamic Environments

In dynamic environments such as search and rescue missions, traditional positioning methods are difficult to meet the requirements of high accuracy and high stability due to factors such as signal shielding, multipath effects, non-line-of-sight (NLOS) propagation, and sensor error accumulation. In response to these challenges, this paper proposes a multi-center collaborative positioning method based on ultra-wideband (UWB) and extended Kalman filter (EKF) to improve the positioning accuracy and robustness of mobile robot groups in complex environments. This method uses UWB to measure relative distance and azimuth, constructs relative position constraints between multiple robots, and uses the DKF algorithm for state estimation on this basis. Different from the traditional filtering method that relies on single-vehicle observations, this paper makes full use of the coupling information between vehicles and spreads the benefits of any measurement in the vehicle group to the entire vehicle group, thereby improving the positioning accuracy while reducing the influence of the sensor error of a single vehicle. In addition, to verify the effectiveness of the method, this paper builds a group of mobile robots based on the ROS platform and conducts experiments in actual scenarios. The experimental results show that the positioning error of the proposed method can always converge within 0.6m, significantly improving the positioning accuracy of the vehicle group.