One is that the AGV itself is subje

One is that the AGV itself is subject to external interference, which is caused by the following reasons: The AGV posture changes greatly due to the action of external forces;When the AGV meets obstacles or water on the ground, the wheels slip, so that the data read from the speedometer cannot truly reflect its motion state;The INERTIAL measurement unit sends a fault and reads the wrong data.Another kind of situation is that the road sign is subject to external interference, leading to the main reasons for this kind of situation are: (1) the action of external forces led to the change of the position of the road sign;(2) The observation sensor failed to read the wrong data.An improved UKF-SLAM algorithm is proposed to solve the problem of external interference affecting SLAM prediction accuracy.On the basis of UKF-SLAM algorithm, the process of interference detection and system state variance inflation is added to determine whether the predicted and observed data are consistent with the inconsistent performance of nearby observations, and it is used for the detection of external interference.When the existence of external interference is detected, the uncertainty of different types of external interference is enlarged by expanding the system state covariance matrix, which makes the system state converge to the truth value quickly and improves the prediction accuracy and robustness of the algorithm