摘要
For the missile with blended aero-fin and lateral impulsive thrust, a blended control autopilot is designed, which comprises an optimal controller and a control allocation module. The combined optimal/classical approach is applied to designing the optimal controller to determine the virtual controls, and the control allocation module is used to distribute the desired vitlual controls onto the redundant control effectors. The autopilot holds some attractive characteristics, such as simple structure, good tracking performance and robustness; moreover the actual constraints of the control effectors can be taken into account. Based on this blended control autopilot, it is found that the conflict between stability and fast tracking performance is serious when using the total acceleration as feedback. In order to avoid this problem, the transient factors in total acceleration are eliminated, so the acceleration caused only by angle of attack is used as feedback, and obvious improvement is shown. Finally, how to get reasonable acceleration feedback is discussed, and conclusion is presented that after passing the low-pass filter, the total acceleration can also be used as feedback, and satisfied tracking performance can be obtained.
For the missile with blended aero-fin and lateral impulsive thrust, a blended control autopilot is designed, which comprises an optimal controller and a control allocation module. The combined optimal/classical approach is applied to designing the optimal controller to determine the virtual controls, and the control allocation module is used to distribute the desired vitlual controls onto the redundant control effectors. The autopilot holds some attractive characteristics, such as simple structure, good tracking performance and robustness; moreover the actual constraints of the control effectors can be taken into account. Based on this blended control autopilot, it is found that the conflict between stability and fast tracking performance is serious when using the total acceleration as feedback. In order to avoid this problem, the transient factors in total acceleration are eliminated, so the acceleration caused only by angle of attack is used as feedback, and obvious improvement is shown. Finally, how to get reasonable acceleration feedback is discussed, and conclusion is presented that after passing the low-pass filter, the total acceleration can also be used as feedback, and satisfied tracking performance can be obtained.
