Aiming at the problem of autonomous positioning of unmanned aerial vehicles in unstructured and complex environment, a simultaneous localization and mapping (SLAM) construction algorithm based on the fusion of stereo camera and inertial navigation device is proposed. The algorithm is divided into two parts: front end and back end. Adopt a local pose estimation method based on sparse feature points, and propose a feature management method and an image frame tracking method based on the inertial measurement unit (IMU) uncertainty model for feature matching and tracking, respectively, which can effectively improve the efficiency and accuracy of feature matching. A graph optimization method that combines stereo information and inertial device information. The unmanned aerial vehicle (UAV) data set Euroc test results show that the SLAM algorithm has good comprehensive performance and can meet the autonomous positioning needs of unmanned aerial vehicles in unstructured and complex environments.