Four-rotor unmanned aerial vehicle path tracking control method based on online safety learning

Abstract

The invention relates to a four-rotor unmanned aerial vehicle path tracking control method based on online safety learning. According to the technical scheme, the method comprises the steps that firstly, a four-rotor unmanned aerial vehicle nominal model is established; secondly, a learning model is obtained according to data of the residual error g(X,U); thirdly, online obstacle avoidance is achieved through probability constraint conditions of a four-rotor unmanned aerial vehicle, and then a control system state safety domain is fused into the learning process; fourthly, the safety domain iscalculated online, the safety of the four-rotor unmanned aerial vehicle is ensured, and the tracking performance is improved; finally, a linear model is obtained through a feedback linearization method, and a model prediction controller is designed for the linear model under corresponding constraint conditions. By means of the method, environment parameters can be obtained on line, the change ofthe inspection environment can be automatically adapted to, and therefore, the robustness and response performance of a control system can be remarkably improved, and the method has the advantages ofbeing high in stability, good in tracking performance and high in safety.

Description

technical field [0001] The invention belongs to the technical field of path tracking control of a rotor unmanned aerial vehicle. Specifically, it relates to a path-following control method for quadrotor UAVs based on online safety learning. Background technique [0002] With the continuous development of robotics and aerospace technology, research on unmanned aerial vehicles has received more and more attention. As a typical unmanned aerial vehicle, the quadrotor UAV has four drive motors, and the flight attitude and position of the aircraft can be adjusted by controlling the speed of the four motors. Because of its compact structure, low cost, and safe and flexible use, it has very broad application prospects in the fields of environmental monitoring, disaster relief, and military reconnaissance. [0003] Since the quadrotor UAV has six degrees of freedom, which is more than the number of control variables, it is a typical underactuated control system. In addition, the q...

AI Technical Summary

Problems solved by technology

The introduced parameter adjustment method has various problems such as cumbersome process, low efficiency, lack of optimality, easy to fall into local minimum, large control error, etc.

The parameters can be obtained online by using the learning method, but the simple learning method cannot guarantee the stability of the system

Although the inherent rolling optimization strategy and the ability to deal with constraints of the model predictive control method have certain advantages in the control of quadrotor UAVs, the method must rely on a relatively reliable dynamic model

However, when the existing robust control methods solve the uncertainty problem, the controller is usually designed under the premise of a given uncertainty bound, which inevitably brings conservatism, and even makes the method impractical, and cannot guarantee that the quadrotor has no problem. Strict requirements for human-machine safety performance

Images