A control method of multi-layer recursive convergent neural network controller for unmanned aerial vehicles

Abstract

The invention discloses a control method of an unmanned aerial vehicle multilayer recursive convergence neural network controller. The method comprises the following steps: constructing an unmanned aerial vehicle model fusing a motor hysteresis effect; based on the unmanned aerial vehicle model, designing a height controller, a yaw angle controller, a roll angle controller, a pitch angle controller, an X controller and a Y controller of an unmanned aerial vehicle by adopting a recursive convergence neural dynamics method; inputting unmanned aerial vehicle state information collected by a control target and an unmanned aerial vehicle sensor to the controller of each unmanned aerial vehicle, and outputting a control component by the controller of each unmanned aerial vehicle; and after the output control component of the controller is converted, transmitting the output control component to an unmanned aerial vehicle motor speed regulator, and controlling the unmanned aerial vehicle to fly by the unmanned aerial vehicle motor speed regulator. Based on the recursive convergence neural dynamics method, the problem correct solution can be quickly and accurately approached in real time, and the obtained controller can well control the unmanned aerial vehicle to track the time-varying trajectory.

Description

technical field [0001] The invention relates to the technical field of unmanned aerial vehicle controllers, in particular to a control method for an unmanned aerial vehicle multi-layer recursive convergence neural network controller. Background technique [0002] Unmanned Aerial Vehicle (UAV) is an important tool in the fields of search and rescue, surveillance, surveying and mapping, and 3D modeling. Among them, quadrotor UAV has the advantages of vertical take-off and landing, hovering and high flexibility, and plays an increasingly important role in practice. increasingly important role. [0003] The most commonly used controller for quadrotor drones is the proportional-integral-derivative (PID) controller based on the deviation, which has strong practicability. Due to the role of the integral term, the steady-state error of the PID controller can converge when tracking a static target. to zero. But the integral term takes a long time to come into play, which makes the ...

AI Technical Summary

Problems solved by technology

But the integral term takes a long time to come into play, which makes the PID controller unsuitable for tracking dynamic targets

In addition, some PID-based controllers, such as fuzzy logic PID controllers, fault-tolerant PID controllers, intelligent PID controllers, etc., have delays in tracking time-varying targets since they still rely on integral terms to eliminate steady-state errors

[0004] In recent years, another potential branch of UAV controller design is the application of neural dynamics method. The traditional neural dynamics controller design methods include zero dynamics method and gradient dynamics method (referred to as ZD-GD method). , but there are defects in these applications on rotor UAVs, that is, the hysteresis effect of the motor is not considered, but the torque is directly used as the control variable, and the UAV cannot be well controlled to track the time-varying trajectory.

Images