Aircraft attitude control method, system and device based on reinforcement learning

Abstract

The invention belongs to the field of aircraft control, particularly relates to an aircraft attitude control method, system and device based on reinforcement learning, and aims to solve the problem ofrealizing non-overshoot control under a low-precision model. The aircraft attitude control method based on reinforcement learning comprises the following steps of: acquiring aircraft state data at at-th moment; calculating an expected attitude angle and an attitude angle tracking error based on an expected flight path of an aircraft, and constructing a network input state at the t-th moment; based on the network input state at the t-th moment and the expected attitude angle, acquiring an optimal control action through utilizing a DDPG deep reinforcement learning network; and adjusting the attitude of the aircraft based on the optimal control action. Non-overshoot control can be realized under a low-precision model, the adjustment time is short, and the steady-state error is small.

Description

technical field [0001] The invention belongs to the field of aircraft control, and in particular relates to an aircraft attitude control method, system and device based on reinforcement learning. Background technique [0002] In recent years, aircrafts have been widely promoted and used in civil fields such as agricultural plant protection, power inspection, urban security, forest fire prevention, environmental monitoring, low-altitude aerial photography, and aerial search and rescue, and in military fields such as battlefield environmental reconnaissance, transportation support, and patrol surveillance. The aircraft has the advantages of fast maneuvering speed, wide coverage, the ability to observe the ground situation from the "God's perspective", and quick response to commands. It will become more popular and have richer application scenarios in the future. [0003] Attitude control is the core of aircraft flight control, and its quality directly affects the effect of the...

AI Technical Summary

Problems solved by technology

However, there are disturbances caused by unmodeled dynamics and uncertain items during the flight, such as gusts, parameter perturbations, etc., and time-varying disturbances. The debugging of parameters is more complicated and requires constant trial and error, which leads to a decrease in the stability margin of the aircraft control, and even the risk of loss of control

Images