In air combat, the identification of aerial targets serv- es as the foundation and prerequisite for situation assessment and threat estimation. Aiming at the problem of air target recognition, a data fusion recognition method based on improved Dempster-Shafer (D-S) Evidence Theory and fuzzy theory is proposed in this paper. For the target information obtained by radar and other multi-source sensor devices, the members- hip function is constructed by fuzzy set theory, and converted into the Basic Probability Assignment (BPA) function of the target to be identified. Then D-S Evidence Theory is used for conflict evidence synthesis and fusion, and the Pignistic probability decision algorithm is employed for discriminant analysis ultimately, so as to realize the identification of air targets. Simulation results show that the proposed method is correct and effective, and improves the accuracy of target recognition to some extent.
In joint operation, unmanned systems greatly promote the generation of new combat capabilities. The autonomous navigation is a key technology for achieving a new paradigm of warfare. To solve the problem of unmanned systems’ real-time precise localization in GNSS-denied environments, an initialization algorithm for monocular visual inertial navigation is proposed. To handle multi-rate measurements in visual inertial navigation system, the IMU pre-integration technique is utilized to process inertial measurements. And then the visual structure from motion is performed. Finally, the visual inertial loosely coupled method is employed to estimate the initial scale, gravity direction, velocity and gyroscope and acceleration biases, and the mathematical model is addressed in detail. Experimental results indicate that the initialization method can obtain accurate initial states in 15 seconds, which can meet the needs of unmanned systems’ cooperative application in joint operation.