Method for controlling landing process of rocket

Abstract

The invention relates to a method for controlling the landing process of a rocket. The method comprises the steps: making the rocket stably land in a near-ground space in different environment statesin a manual control mode, and recording first type of flight data generated in the landing process; establishing a first artificial neural network and training the first artificial neural network by utilizing the collected first type of flight data, wherein the first artificial neural network is trained to calculate a real-time control strategy of the rocket in different environment states when the rocket arrives at each position point, the input parameters of the first artificial neural network comprise the position rn of the rocket at the current moment and the current wind speed wn, and theoutput parameters comprise a real-time control parameter value un used for controlling a power system of the rocket; and controlling the landing of the rocket according to the real-time control parameter value un output by the trained first artificial neural network. By means of the method for controlling landing process of the rocket, rocket landing can be accurately controlled in real time on the basis that various forms of external force borne by the rocket are considered.

Description

technical field [0001] The present invention relates to the field of rockets. In particular, the invention relates to a method for controlling the landing process of a rocket. Background technique [0002] A recoverable rocket is a space launch vehicle that can be reused. The technical difficulty of existing recoverable rockets lies in the control of the landing phase, because it not only needs to land accurately and softly according to the predetermined landing point, but also requires the speed to approach zero when landing. In the prior art, in order to make the lunar exploration spacecraft achieve a soft landing, control methods such as gravity turning guidance are generally used. Existing recoverable rockets near the ground are generally controlled using algorithms based on PID or convex optimization. However, the landing section often faces complex interactions such as aerodynamics and the force between the rocket and the ground, which are difficult to calculate wit...

AI Technical Summary

Problems solved by technology

But these other forms of force are not only difficult to measure, but also difficult to predict through existing calculation methods due to their large randomness

Images