Formation Control Of Multiple Quadrotors Based On Adaptive Dynamic Programming

Quadrotor unmanned Aerial Vehicle(UAV)with its low cost and innovative structure,has a wide application in practice,such as rescue,fire control,surveillance,inspection,mapping.The cooperative task execution of multiple UAV formations has the advantages of high load capacity,diverse sensor selection,and strong cruising ability,which effectively solves the problems of small load and low fault tolerance when a single UAV performs a task.Therefore,this article focuses on the formation robustness,reconstruction,and fault-tolerant control methods of quadrotor UAVs.The main contents are as follows:(Ⅰ)To solve the formation control problem of quadrotor UAV under the influence of external disturbance,a formation robust controller design method based on adaptive dynamic programming is proposed when the disturbance upper bound is known.Firstly,considering the influence of the comprehensive disturbance of the known upper bound,a value function reflecting uncertainty,tracking error and control quantity is designed.A robust tracking control scheme based on a neural network is established using the critical neural network to approximate the optimal value function.Secondly,the inner loop reference signal is obtained by the attitude calculation,and the attitude controller based on the finite time sliding mode is designed to realize the attitude control of the UAV.Finally,simulations verify that the designed formation controller can ensure the generation and maintenance of multiple UAV formations under the influence of external interference under the condition that the interference upper bound is known.(Ⅱ)To solve the problem of multi-UAV formation reconfiguration control under the influence of comprehensive disturbance,an adaptive dynamic programming method based on a neural network is proposed.Firstly,the sliding mode method is used to design the disturbance observer to estimate the comprehensive external disturbance,and the cost function is developed based on the estimated value.Under the influence of the comprehensive disturbance,the formation control problem is transformed into the optimal stability control problem.The neural network is used to approximate the optimal cost function and solve the optimal control.At the same time,to meet the target of formation transformation,a potential energy function is used to avoid the collision,and the formation position controller is designed comprehensively.Then,perform attitude calculation for each UAV and design a finite-time attitude tracking controller to ensure rapid attitude adjustment.Finally,the formation maintenance and formation reconstruction control of UAV formation in the disturbance’s environment is realized through the simulation experiment.(Ⅲ)Aiming at the multi-UAV formation system with actuator failure,a fault-tolerant controller based on the fault observer and ADP algorithm is designed.Firstly,the designed fault observer is used to estimate the fault,and from this,the performance index function reflecting the fault,regulation and control of the actuator is constructed.At this time,the fault-tolerant control problem with actuator failure can be transformed into an optimal control problem,then by constructing a critical neural network,the strategy iterative algorithm is used to solve the HJB equation to achieve outer loop formation control and ensure the outer loop position with actuator failure is consistent and ultimately bounded.At the same time,the attitude controller is designed based on the sliding mode algorithm to achieve fast attitude tracking.Finally,the simulation verifies the stable operation of the multi-UAV formation system with actuator failure.