Neural adaptive control of LoFLYTE(R)

Abstract

A major goal in flight control over the past decade has been the development of reconfigurable flight control systems which can adapt their gains in real-time to compensate for aircraft damage and in-flight system failures. The purpose of this paper is to describe the controller developed for the LoFLYTEQ aircraft, which is a testbed for neural networks research. The LoFLYTEQ control system is based on the Accurate Automation Corp. Neural Adaptive Controller VACTM) which is designed to achieve this goal. The LoFLYTEQ program is an active flight test program at the Air Force Flight Test Center at Edwards Air Force Base, with the objective of demonstrating a neural network control system for a waverider vehicle. The AAC control system has two innovative components: an adaptive actuatorklight surface controller, and a leaming/adaptive stability augmentation system designed with neural network and reinforcement learning techniques. AAC NEURAL ADAPTIVE CONTROLLER (NACTM)The NACTM, whose architecture is illustrated in Figure 1, is an n* order multivariate adaptive controller derived fiom Seraji's , 2"d order single variate robotic joint controller, to which a neural outer loop has been added to improve the short and intermediate time response of the controller [Cox, et a1 (1994)] [Saeks, et a1 (1997)l. The NACTM is applicable to an essentially arbitrary n* order multivariate linear or nonlinear system that satisfies an appropriate "slow-varying" assumption. [In the continuous-time situation, this assumption is straightforward to satisfl. It turns out, however, when discretization is performed for computational reasons, the discretization itself injects "high speed" variations, and theoretically it can be shown that terms of the "wrong sign" enter the discrete equations. Theoretically, this computation-induced condition may be alleviated by using sufficiently high sampling rates.] Stability and robust tracking are guaranteed by the underlying adaptive controller. For some applications, a neural network "gain scheduler'' is used to speed up the adaptation process and to enhance the short and intermediate time response of the system. The N A P M deals equally well with linear and nonlinear systems and requires no explicit aircraft model. Rather, it implicitly models the aircraft dynamics via the observed error between the actual and reference trajectories. It can adapt in real-time to changes in the plant dynamics caused by system failures,

Modeling and Inverse Controller Design for an Unmanned Aerial Vehicle Based on the Self-Organizing Map

“…The best result with TDNN was obtained from 20 Monte-Carlo simulations with 12 PEs in the hidden layer 4 . From Table V, we can conclude that the constructed SOMbased network is a good model of the underlying dynamics because it provides larger SER and smaller NRMSE for all 3 Taking neighbor PEs means that the radius of neighborhood function is greater than 1. 4 We tried various sizes of TDNN's and found that the one with 12 PE's in the hidden layer performs best in system identification.…”

“…7. The LoFLYTE ® program is an active flight test program at the Air Force Flight Test Center at Edwards Air Force Base, with the objective of developing the technologies necessary to design, fabricate, and flight test a Mach 5 waverider aircraft [3,38]. In addition, the LoFLYTE ® UAV has been used to understand the low speed characteristics of a hypersonic shape and to demonstrate several innovative flight control technologies.…”

“…9(b), better performance in the testing set was obtained when we take the data samples from the winning PE as well as neighbor PEs in local modeling phase rather than from the winning PE alone 3 . The created models were tested by new sequences of 10 sets with 1,000 samples.…”

“…Controllers for aircraft have been designed predominantly by classical control techniques [39]. While this tradition has produced many highly reliable and effective control systems, recent years have seen a growing interest in the use of robust, nonlinear adaptive control theory for flight control [3], [22], [36]. For instance, the concept of multiple models with switching, according to a change in dynamics, has been an area of interest in control theory in order to simplify both the modeling and the controller design [ 13,17,21,33].…”

Rainbow Runner glider as a testbed for robust and adaptive control methods

“…Bertuccelli et al (2009); Rusnak et al (2011) integrate in their work simulation of their algorithms with real flight tests and bring up combined results. Cox et al (2001) claimed, that the LoFLYTETM program was being tested onboard an experimental waverider-shaped subsonic aircraft and would be used as a testbed for a neural network flight control system.…”