Training， testing， and evaluating radar intelligent recognition （RIR） methods usually require the construction of large datasets. Acquiring effective datasets efficiently is an urgent issue to be solved. The application scenarios of the RIR system for aerospace targets were analyzed first， and the construction idea of the RIR simulation system （RIRSS） for aerospace targets was given according to application demands. Then， the modular and component-based design methods were utilized to design the RIRSS framework for aerospace targets， and an intelligent recognition simulation system for aerospace targets was constructed， which could simulate the electromagnetic characteristics and dynamic echo of aerospace targets quickly and efficiently， with a scalable and reconfigurable capability. Simulation results and preliminary application show that this system can be used in the test and evaluation of the RIR method for aerospace targets in various scenarios.

For intercepting highly maneuvering targets， an integral sliding mode guidance law with finite time convergence is proposed based on the integral sliding mode control theory. A radial basis function （RBF） neural network disturbance observer with adaptive parameter adjustment is used to estimate the disturbance of the guidance system caused by the target maneuver. In order to further improve the estimation accuracy， the adaptive law is used to eliminate the estimation error of the neural network observer. According to Lyapunov stability theory， the stability and finite time convergence of the guidance law are proved.The numerical simulation shows that the guidance law proposed in this paper has excellent performance both in miss distance and in the control accuracy of missile and target line of sight angle.