![]() ![]() Combined with the advantages and disadvantages of LADRC, Adaptive Control and Neural Network, the coupling force between each channel, gust crosswind disturbance and additional resistance of offshore platform jitter in the flight state of the quadrotor are optimized. Radial Basis Function Neural Networks (RBFNN) is a well−performing forward network with best approximation, simple training, fast learning convergence and the ability to overcome local minima problems. Linear Active Disturbance Rejection Control (LADRC) is a product of a tracking differentiator (TD), a linear extended state observer (LESO) and an error feedback control law (PD) and is a control technique for estimating compensation for uncertainty. In this paper, a robust control scheme based on linear active disturbance rejection is proposed to solve the problem that the quadrotor is susceptible to various disturbances during the take−off process of non−horizontal planes and strong disturbances. The quadrotor aircraft has the characteristics of simple structure, high attitude maintenance performance and strong maneuverability, and is widely used in air surveillance, post−disaster search and rescue, target tracking and military industry.
0 Comments
Leave a Reply. |