Error Dynamics Based Dual Heuristic Dynamic Programming for Self-Learning Flight Control

Abstract

A data-driven nonlinear control approach, called error dynamics-based dual heuristic dynamic programming (ED-DHP), is proposed for air vehicle attitude control. To solve the optimal tracking control problem, the augmented system is defined by the derived error dynamics and reference trajectory so that the actor neural network can learn the feedforward and feedback control terms at the same time. During the online self-learning process, the actor neural network learns the control policy by minimizing the augmented system’s value function. The input dynamics identified by the recursive least square (RLS) and output of the critic neural network are used to update the actor neural network. In addition, the total uncertainty term of the error dynamics is also identified by RLS, which can compensate for the uncertainty caused by inaccurate modeling, parameter perturbation, and so on. The outputs of ED-DHP include the rough trim surface, feedforward and feedback terms from the actor neural network, and the compensation. Based on this control scheme, the complete knowledge of system dynamics and the reference trajectory dynamics are not needed, and offline learning is unnecessary. To verify the self-learning ability of ED-DHP, two numerical experiments are carried out based on the established morphing air vehicle model. One is sinusoidal signal tracking at a fixed operating point, and the other is guidance command tracking with a morphing process at variable operating points. The simulation results demonstrate the good performance of ED-DHP for online self-learning attitude control and validate the robustness of the proposed scheme

A Non-Global Traversal Method for Dynamic Load Rapid Localization and Identification

“…Step 2. Iteration: Apply the formula x f 0) to iterate once, obtain the new approximation x f (1) , and calculate g 1 = g x f (1) and g ′ 1 = g ′ x f (1) .…”

“…Step 3. Control: If x f (1) satisfies either |κ| < ε 1 or |g 1 | < ε 2 , terminate the iteration and use x f (1) as the desired root. Otherwise, proceed to step 4.…”

“…Modification: If the number of iterations reaches the pre-specified number s, or if g ′ 1 = 0, the method fails. If not, replace x f (1) , g 1 , g ′ 1 with x f (0) , g 0 , g ′ 0 and proceed to step 2 to continue the iteration.…”

“…Dynamic load identification technology is the second type of inverse problem in structural dynamics [1][2][3]. Dynamic load positioning and identification pertain to the technology for analyzing and processing the dynamic response of a structure to obtain the position and amplitude information of the dynamic load received by the structure during operation.…”

Special Issue on Advanced Technology of Intelligent Control and Simulation Evaluation

“…To enhance the unpredictable behavior and capacity for exploration of the grey wolf, a new parameter-learning technique is developed. In [2], a data-driven nonlinear control approach, called error-dynamics-based dual heuristic dynamic programming, is proposed for air vehicle attitude control. To solve the optimal tracking control problem, the augmented system is defined by the derived error dynamics and reference trajectory so that the actor neural network can learn the feed-forward and feedback control terms at the same time.…”