针对夜间环境建立了舰载机人工进近着舰模型,模型包括了舰载机、甲板运动、舰尾流、着舰指挥员及飞行员模型,飞行员模型基于MPC方法建立,能够描述飞行员在控制输入和速率约束下的控制策略,且在约束边界内建立的飞行员模型与LQG飞行员模型等效,经仿真验证,建立的飞行员模型在频域0.1Hz-10Hz内符合人类特性。基于建立的飞行员模型完成了夜间环境的飞行仿真,仿真结果表明:夜间环境影响飞行员对角度、角速度及横纵偏差的观测精度,相较于日间环境飞行员的纵向航迹偏差散布增大,横向航迹偏差散布略微增大,且在靠近舰船时有躲避舰尾的趋势;通过重复仿真实验验证舰机人环大系统的合理性,结果表明在1/2英里、1/4英里、1/8英里及舰尾处的着舰散布趋势与美军实验一致,夜间复飞率为28%,日间为12%,符合实际经验,验证了建立的人工着舰模型可以用于分析复杂环境下的舰载机着舰安全。
A carrier-based aircraft model was established for night-time environments, which includes models for the carrier-based aircraft, deck motion, ship wake, landing signal officer, and pilot. The pilot model was developed based on the Model Predictive Control (MPC) method, capable of describing the pilot's control strategy under control input and rate constraints. The pilot model established within the boundary constraints is equivalent to the Linear Quad-ratic Gaussian (LQG) pilot model. Through simulation verification, the established pilot model exhibits human-like characteristics within the frequency range of 0.1Hz to 10Hz.Based on the established pilot model, flight simulations in night-time environments were conducted. The simulation results indicate that the night-time environment affects the pilot's observation accuracy of angles, angular velocities, and lateral and longitudinal deviations. Compared to daytime conditions, the pilot's longitudinal trajectory deviation dispersion increases, while the lateral trajectory devi-ation dispersion slightly increases. There is also a tendency to avoid the ship's wake when approaching the vessel. Through repeated simulation experiments, the rationality of the carrier-aircraft-human system was validated. The results show that the landing dispersion trends at 1/2 mile, 1/4 mile, 1/8 mile, and at the ship's wake are consistent with U.S. military experiments. The night-time go-around rate is 28%, while the daytime rate is 12%, aligning with practical experience. This validates that the established artificial landing model can be used to analyze carrier-based aircraft landing safety in complex environments.