Optimal Flight Control on Cessna X Aircraft using Differential Evolution

Abstract

Usually, setting the appropriate optimal gains for Stability Augmentation System and Control Augmentation System for aircrafts depends on the system knowledge by the engineer. When this setting depends on tuning gains such as Proportional Integrator Derivative control or weights as in Linear Quadratic Regulator method, the engineer will use the trial and error process, which is time consuming procedure. In this research, a study of modeling and control system design will be conducted for a business aircraft using heuristic algorithm. A linear model of Cessna Citation aircraft was designed. Then a Linear Quadratic Regulator technology was used to achieve desirable dynamic characteristics with respect to the flying qualities requirements on the stability augmentation system for the Cessna Citation X aircraft. The Proportional Integral controller was further used in the Control Augmentation System, the weighting matrix of the LQR method and the PI parameters were optimised by using the differential evolutions method. The heuristic algorithm here used has given very good results. This algorithm was used in this form for the first time to optimize linear quadratic regulation and proportional Integral controllers on an aircraft control, using one fitness function.

New Methodology for Optimal Flight Control using Differential Evolution Algorithms applied on the Cessna Citation X Business Aircraft – Part 2. Validation on Aircraft Research Flight Level D Simulator

“…as the LQR controller performance rely on the weighting matrices selection, then it became interesting to automate the weighting searches processes, as shown in [3], where the LQR was genetically optimized for UAV control under wind disturbance, and gave good results in both performance and robustness, and [4] the authors optimized the performance of the controller using the LQR method, with the meta-heuristic Differential Evolution, the controllers were cleared for each flight condition in the Cessna Citation X aircraft flight envelope. In [5], and [6], LQR gains were optimized by using the Genetic Algorithm and were applied on Lynx helicopter, and lateral control on Cessna Citation X business aircraft, the robustness of the controllers was assisted by the guardian map theory, the optimized controllers show a very good results, in other hand, the application of the guardian map is a very long time computation, which made the guardian map method less desirable to clear the controller for the entire flight envelope.…”

“…By using a Cessna Citation X Level D Research Aircraft Flight Simulator designed and manufactured by CAE Inc the benchmark was developed at Laboratory of Active Controls, Avionics and AeroServoElasticity LARCASE in [10]- [11]. This benchmark programmed in Matlab/Simulink was already used for new identification methods designed and developed in [12]- [13], for advanced flight control design and clearance [14]- [15], and for robust control analysis in [4]- [6]. This paper is organized as follows: First a description of the controller optimized using the differential evolution algorithm, the aircraft flight envelope is detailed, and then a brief description of the clearance criteria.…”

New Methodology for Optimal Flight Control Using Differential Evolution Algorithms Applied on the Cessna Citation X Business Aircraft – Part 1. Design and Optimization

“…Then the difference between two different vectors is weighted by a scalar selected at random to finally obtain the "donor" vector , [27] as defined in equation (9):…”

“…Two integers and are chosen arbitrarily in the exponential crossover from the interval [1, ] , where D represents the dimension, which is the number of parameters subject to optimization [27], and then the trial vector is given as follows: …”

“…In [21], [22], and [23] the authors used a PID controller based on genetic tuning. In [24][25][26][27][28][29] the Differential Evolution and Genetic Algorithm were used to optimize robust flight controllers.…”