نوع مقاله : مقاله پژوهشی
1 کارشناس ارشد / دانشگاه آزاد اسلامی، واحد گرمسار، باشگاه پژوهشگران و نخبگان جوان
2 عضو هیات علمی / دانشکده مهندسی هوافضا، دانشگاه صنعتی شریف
3 عضو هیات علمی / مجتمع برق و الکترونیک، دانشگاه صنعتی مالکاشتر
عنوان مقاله [English]
In this paper, we have utilized Simultaneous Localization and Mapping (SLAM) problem to design a navigation algorithm for an Unmanned Air Vehicle (UAV). In our scheme, a bearing-only sensor has been installed on UAV as an external sensor that measures the relative bearing angles between the vehicle and unknown landmarks on the ground. SLAM algorithm helps us to reach precise estimates for vehicle states and the position of landmarks at the same time. Extended Kalman Filter has been used as a fusion mechanism to combine the received measurements from external sensor and provided data by Inertial Navigation System (INS). Moreover, for finding a precise location for landmarks when the UAV observes them for first time, we have used a Delayed Constrained Initialization approach. When UAV reaches to the end of the considered trajectory, it flies through the trajectory for second time in which the UAV utilizes the stored data in previous flight to enhance the accuracy of state estimates. Finally, simulation results show the performance of the proposed approach in design of a navigation algorithm for UAVs.