Small unmanned aerial vehicle control object modeling method based on test data

Abstract

The invention discloses a small unmanned aerial vehicle control object modeling method based on test data. According to the method, through transformation of control signals, an established control object model can be applied to the process with small variation and can also be applied to the process with great variation, and an established model system is high in identification precision and good in model prediction capacity. According to the method, by the utilization of test flight data, high-cost wind tunnel tests can be avoided, and therefore early-stage development cost is lowered. According to the method, through changing of mathematical features of input signals (namely the control signals) in the test flight test data, the difference between the different input signals is amplified, therefore, the error range of an estimation model and a true model is narrowed down, and the accuracy and the prediction capacity of the estimation model are improved.

Description

technical field [0001] The invention relates to the field of control object modeling, in particular to a method for modeling a control object model of a small unmanned aerial vehicle based on test data. Background technique [0002] For the establishment of the aircraft control object model, there are two main modeling methods: one is to use the wind tunnel test data to establish the UAV control object model; the other is to use the test flight data to establish the UAV control object model through model identification. For small unmanned aerial vehicles, the cost of wind tunnel tests is huge and the accuracy is not high at low speeds, so the second method is often used for modeling in the early development stage. However, due to the many limitations of the flight test itself (flight safety and control signal problems), the general test of a new type of small aircraft, the test data generally only has a small range of data, which cannot well reflect the performance of the ai...

AI Technical Summary

Problems solved by technology

However, due to the many limitations of the flight test itself (flight safety and control signal problems), the general test of a new type of small aircraft, the test data generally only has a small range of data, which cannot well reflect the performance of the aircraft itself, which is not conducive to use The method of system identification is used to establish the control object model, and the final model is usually not suitable for a process with large changes, and the accuracy is not high, and the prediction performance is relatively poor

Images