Stratospheric airship height control method and system

Abstract

The invention relates to a stratospheric airship height control method and system, and belongs to the technical field of airship height control, and the method comprises the steps: building a Markov decision process model for airship height control, selecting a plurality of motions from a motion space according to a Q-learning algorithm and a state transition probability based on a current state;determining all actions selected from the initial state to the termination state of the airship, taking all the selected actions as action sequences corresponding to the termination state, finally selecting the action sequence with the maximum target value in all the action sequences as an optimal sequence by utilizing a target optimization function, and controlling the airship to move according to the optimal sequence. Furthermore, the relevant kinetic model parameters of the stratospheric airship do not need to be obtained any more, the real-time state of the stratospheric airship and the set reference state serve as control input of the control system, and a control loop is simplified.

Description

technical field [0001] The invention relates to the technical field of airship height control, in particular to a method and system for controlling the height of a stratospheric airship. Background technique [0002] The stratospheric airship is a typical low-speed near-space vehicle. In the air-stationing stage, the energy depends on the day and night closed-loop system composed of solar cells and batteries, and the power depends on the vector propulsion device. The purpose is to control the horizontal trajectory and altitude of the airship, so that Achieving monthly-level long-duration regional stays, it has great application potential in the fields of high-resolution observation of the earth, communication relay, reconnaissance and surveillance, and environmental monitoring, and is known as a "stratospheric satellite." [0003] Reliable altitude control is one of the key technologies for airships to stay in the stratosphere and stay in the air for a long time. , The stra...

AI Technical Summary

Problems solved by technology

Scholars at home and abroad have carried out a series of studies on the altitude control of stratospheric airships and similar aerostats. Traditional controls include conventional linear PID control, nonlinear gain scheduling control, backstepping control, etc., and can achieve altitude control under certain conditions. effective control, but to varying degrees, there are problems such as difficulty in parameter selection and difficulty in self-adaptation

With the improvement of computer performance and the rapid development of machine learning algorithms and platforms, artificial intelligence decision-making and control have gradually penetrated into various application scenarios, including the Monte-Carlo learning method using Gaussian processes to process continuous state action spaces, Gaussian process dynamics However, these intelligent control methods generally have the characteristics of large randomness and poor convergence characteristics.

Images