Unmanned ground vehicle‐unmanned aerial vehicle relative navigation robust adaptive localization algorithm

The unmanned aerial vehicle (UAV) and the ground (UGV) can complete complex tasks through information sharing ensure mission execution capability of multiple carrier platforms. At same time, cooperative navigation use interaction between multi-platform sensors to improve relative positioning accuracy entire system. Aiming at problem deviation system model due gross errors in sensor measurement data or strong manoeuvrability environments, a robust adaptive UGV-UAV algorithm is proposed. In this paper, studied based on inertial unit (IMU)/Vision. Based analyzing kinematics model, leader (UGV)-follow established. implementation algorithm, non-linear Kalman filter (extended [EKF]) are combined dynamically adjust state parameters. Finally, mathematical simulation accompanying landing process scene carried out. position, velocity attitude calculated by EKF, Robust-EKF Robust-Adaptive-EKF algorithms compared analyzed simulating two cases where noise obeys Gaussian distribution non-Gaussian distribution. results show that under conditions, improved about three times with EKF almost reach conditions. self-adaptive proposed paper has adaptability uncertainty modelling higher stronger robustness.