Design and Implementation of an Optimal Energy Control System for

0 downloads 0 Views 5MB Size Report
Nov 19, 2016 - Abstract: In conventional flight control design, the autopilot and the ... design of a flight control system using TECS and H∞ control theory was introduced ...... design for flight and propulsion control using total energy principles.
applied sciences Article

Design and Implementation of an Optimal Energy Control System for Fixed-Wing Unmanned Aerial Vehicles Ying-Chih Lai 1, * and Wen Ong Ting 2 1 2

*

Department of Aerospace and Systems Engineering, Feng Chia University, No. 100, Wenhwa Road, Seatwen District, Taichung 407, Taiwan Institute of Aeronautics and Astronautics Engineering, National Cheng-Kung University, No. 1, Daxue Road, East District, Tainan 701, Taiwan; [email protected] Correspondence: [email protected]; Tel.: +886-4-2451-7250 (ext. 3956)

Academic Editors: Josep M. Guerrero and Amjad Anvari-Moghaddam Received: 18 August 2016; Accepted: 15 November 2016; Published: 19 November 2016

Abstract: In conventional flight control design, the autopilot and the autothrottle systems are usually considered separately, resulting in a complex system and inefficient integration of functions. Therefore, the concept of aircraft energy control is brought up to solve the problem of coordinated control using elevator and throttle. The goal of this study is to develop an optimal energy control system (OECS), based on the concept of optimal energy for fixed-wing unmanned aerial vehicles (UAVs). The energy of an aircraft is characterized by two parameters, which are specific energy distribution rate, driven by elevator, and total specific energy rate, driven by throttle. In this study, a system identification method was employed to obtain the energy model of a small UAV. The proposed approach consists of energy distribution loop and total energy loop. Energy distribution loop is designed based on linear-quadratic-Gaussian (LQG) regulator and is responsible for regulating specific energy distribution rate to zero. On the other hand, the total energy loop, based on simple gain scheduling method, is responsible for driving the error of total specific energy rate to zero. The implementation of OECS was successfully validated in the hard-in-the-loop (HIL) simulation of the applied UAV. Keywords: total energy control system; hard-in-the-loop simulation; unmanned aerial vehicle; system identification; linear-quadratic-Gaussian regulator

1. Introduction Driven by advanced computing technologies, open-source systems, investigations and applications of unmanned aerial vehicles (UAVs) have progressed at an astounding speed in the past decades. To extend the applications of UAVs in wide area or long range, the fuel efficiency of the propulsion system plays a key role in this topic. In general, a fixed-wing aircraft has higher aerodynamic efficiency than a rotorcraft, which implies that it has better endurance and range than the rotorcraft. It can provide the advantage of longer flight durations at higher speeds thus enabling larger survey areas per given flight. Both the range and endurance of an aircraft depend on rate of fuel consumption of the engine. Therefore, the flight control design plays an important role in the optimization of the fuel consumption for an aircraft or UAV. In conventional flight control design, the autopilot and the autothrottle systems are usually considered separately [1]. The autothrottle unit is mainly used to adjust the engine throttle valve to achieve the desired flight speed, and the autopilot unit is designed to regulate the control surfaces for attitude and altitude maneuvers. However, since the dynamic model of an aircraft is nonlinear and uncertain, interaction between autopilot and autothrottle systems actually exists, and dynamic coupling between speed and altitude always occurs. Appl. Sci. 2016, 6, 369; doi:10.3390/app6110369

www.mdpi.com/journal/applsci

Appl. Sci. 2016, 6, 369

2 of 24

According to the study proposed by Denker [2], a pilot who tries to control airspeed and altitude separately winds up controlling one or the other rather poorly. Often times, the airspeed gets too low, whereupon the wing stalls and the pilot abruptly loses control. There is a coupling effect between airspeed and altitude, and controlling either one will affect another. In conventional flight control system, airspeed is controlled by throttle and altitude is controlled by elevator. This method will cause the undesirable flight path and airspeed coupling, and loss of airspeed control when thrust is limited. Therefore, by utilizing the relationship between airspeed and altitude, a control system was introduced by utilizing the energy of an aircraft. The first energy-based flight control system, named total energy control system (TECS), was developed by Lambregts [3,4]. TECS is a generic control technique that integrates all longitudinal flight-path and speed control functions, which does not depend on a particular aircraft type [5]. For fixed-wing aircraft, the mechanism and applications of TECS were used in guidance [6,7] and flight control [5,8,9]. TECS based controllers have been successfully implemented on many aircrafts, such as the National Aeronautics and Space Administration (NASA) B737 technology demonstration plane [10], a general aviation aircraft [11], and a motor glider [12]. For helicopter, the design of a flight control system using TECS and H∞ control theory was introduced in the study [1]. Recently, studies based on TECS for UAVs have attracted more and more attention. However, compared to manned aircrafts, there are relatively few studies focused on the fuel efficiency of UAVs [13,14]. To enhance the performance and usability of UAVs, there is a need for UAVs to increase endurance and/or reduce the amount of fuel carried. Many studies of UAVs focused on the problem of determining the optimal trajectory required to improve the fuel consumption [15–17]. Therefore, the design of the flight controllers emphasized on how to keep on the desired trajectory. For example, Deittert et al. used a Linear Quadratic Regulator (LQR) to stabilize the UAV on its nominal trajectory whilst rejecting disturbances and modeling errors [16]. However, for a typical flight profile of an aircraft, the level flight, climbing, and descending maneuvers are performed all the time after the aircraft leaves the ground. Therefore, the longitudinal flight control plays a significant role in fuel efficiency of an aircraft. A nonlinear description and implementation of TECS and a Lyapunov-based update law to estimate the aerodynamic drag coefficient of a UAV, assuming that the structure of the aerodynamic drag is known, were presented in [14]. Moreover, a longitudinal flight control system for UAVs based on TECS using 1 adaptive control theory was proposed in [13]. The fuel efficiency of UAVs is a significant issue that will limit their applications in wide area and long range. A popular solution is to adopt new propulsion systems, such as fuel cell or solar power, but these new systems are expensive and unstable due to the weight and space constraints of UAVs. Therefore, we decided to provide an approach based on flight control theory and flight dynamics. The proposed approach has the ability of the optimization of the fuel consumption and will not increase the cost and weight of UAVs. The goal of this study is to provide an optimal flight control system based TECS for fixed-wing UAVs. We propose an optimal energy control system (OECS) by integrating TECS and LQG regulator [18] to improve the fuel efficiency of UAVs. The successful flight demonstration of a UAV, named Spoonbill, proved that the LQG regulator can be effectively used in the flight control of small UAVs [19]. However, the longitudinal flight controller used previously on Spoonbill UAV for controlling the flight altitude was an altitude-hold controller. As such, it is unable to maintain a required climb/descent rate of the Spoonbill UAV. To overcome this problem, this study presents the development of an energy-based flight control system, in conjunction with optimal control technique, LQG regulator. Finally, the developed control scheme is verified in the hard-in-the-loop (HIL) simulation of Spoonbill UAV. The design steps of the proposed OECS are as follows. First, the energy equations of the model of an aircraft are derived to obtain the total specific energy rate, driven by throttle, and the specific energy distribution rate, driven by elevator. Then, the energy distribution state-space model is obtained by using small-disturbance theory and Taylor’s series expansion. Second, the system identification process is executed to identify the unknown system parameters of the discrete time state-space model by using a set of experimental input and output data. Third, the proposed OECS is designed to form

Appl. Sci. 2016, 6, 369

3 of 24

the aircraft longitudinal flight control system and to control the airspeed and altitude of an aircraft. The control scheme of inner and outer loops is used. In inner loop, the energy distribution loop is designed based on LQG regulator and responsible to regulate specific energy distribution rate to zero. In the outer loop, total energy loop, a gain scheduling method is adopted and it is responsible for driving the error of total specific energy rate to zero. Gain scheduling approach is applied in this study since it has had great success in a variety of commercial as well as military applications [20]. Finally, the implementation of OECS is validated in the HIL simulation of the target UAV. The remainder of this paper is organized as follows. Section 2 derives the energy equations and state-space model for an aircraft. Section 3 presents the process of determining adequate mathematical model by using the system identification method. Section 4 presents the design and implementation of the proposed OECS. Section 5 shows the simulation results using the HIL system of Spoonbill UAV. Finally, Section 6 gives the conclusions of this study. 2. Aircraft Energy Equations and Dynamic Model The concept of TECS design is to compute the total energy state and desired state of an aircraft as represented by flight path, speed, and associated target. The total energy error is controlled by thrust, and elevator controls the energy distribution error between flight path and airspeed. Therefore, it is important to derive the energy equations of an aircraft, especially for the total specific energy rate error and the specific energy rate distribution error. Then, the aircraft distribution energy model can be derived by postulating from the simplified equations of motion and energy equations with the addition of the jerk of aircraft. 2.1. Aircraft Energy Equations The derivation of energy equations is based on Lambregts [4]. The total energy of an aircraft can be expressed as the sum of the potential energy and kinetic energy as follows: E = EKE + EPE E=

(1)

1 mV 2 + mgh 2

(2)

where g is the acceleration of gravity, h is altitude, m is aircraft weight, V is the longitudinal velocity, EKE is kinetic energy, and EPE is potential energy. Assuming constant weight, Equation (2) can be rewritten as:  E = mg

1 V2 +h 2 g

 (3)

Similar to the derivation of aircraft equations of motion, by using the small-disturbance theory, the energy of aircraft is linearized at steady flight conditions, as shown below: E0 + ∆E = mg

1 (V0 + ∆V )2 + (h0 + ∆h) 2 g

! (4)

where V0 is trimmed airspeed. Assume that (∆V )2 = 0, and differentiate Equation (4) with respect to time, yielding: " ! # d d 1 V0 2 2V0 ∆V (5) + + (h0 + ∆h) ( E0 + ∆E) = mg dt dt 2 g g ! . . . V0 ∆V ∆ E = mg + ∆h (6) g

Appl. Sci. 2016, 6, 369

4 of 24

Dividing by the trimmed airspeed, V0 , yielding: .

.

∆E = mg V0

∆V + ∆γ g

! (7)

where γ is flight path angle. Thus, at a given airspeed, the rate of change of aircraft energy is dependent only upon the change . of longitudinal acceleration, ∆V, and the change of flight path angle, ∆γ. Since the total energy is determined by the thrust of engine, the thrust can be derived from aircraft longitudinal equation of motion, . m∆V = ∆T − ∆D − mg sin∆γ (8) Assuming that ∆γ is small, therefore sin∆γ = ∆γ, then .

∆V mg ∆γ + g

!

= ∆T − ∆D

(9)

Thus, the aircraft rate of change of energy is proportional to the difference between thrust, T, and drag, D. By rearranging the equation above, the required thrust change as: .

∆V ∆T = mg ∆γ + g

!

.

.

∆E ∆ Es + ∆D = mg + ∆D + ∆D = V0 V0

(10)

.

where ∆ Es is total specific energy rate. Assume that during the steady flight condition, the change of drag is small and negligible. The change of thrust is proportional to the sum of change of longitudinal acceleration and change of flight path angle: . ! ∆V ∆T ∝ ∆γ + (11) g Conversely, at a specific thrust level, it is possible to trade the change of flight path angle for change of longitudinal acceleration and vice versa using the elevator control only. Finally, the change of total specific energy rate error, driven by throttle, is derived as .

∆ Es = ∆γ + .

∆along g

.

.

∆ Es = Es − Es0 .

.

.

(12) (13)

.

where ∆ Es = Es − Es0 and Es0 = 0 is the total specific energy rate of an aircraft at steady flight condition. In addition, the change of specific energy rate distribution error, driven by elevator, is derived as .

∆ Ls = −∆γ +

∆along g

(14)

.

where ∆ Ls is the change of specific energy rate distribution error and along is longitudinal acceleration .

along flight path. The detailed derivation of ∆ Ls refers to the study in [14]. 2.2. Energy Distribution State-Space Model By assuming that aircraft is a rigid body with fixed mass distribution and constant mass, the aircraft equations of motion can be obtained by Newton’s second law. In addition, the energy equations

Appl. Sci. 2016, 6, 369

5 of 24

are obtained from law of conservation of energy. The equations of motion and energy equations are both simplified by using small-disturbance theory and Taylor’s series expansion. The aircraft distribution energy model is postulated from the simplified equations of motion and energy equation with the addition of the jerk of aircraft. The energy distribution state-space model can be expressed in the following equation:          

.

∆V . ∆α . ∆q . ∆θ . ∆ax . ∆ az

 XV ZV   V0     ZV M α.   M + V V0 =   0       A xV  AzV 

Xα V0 Zα V0 Z M Mα + αV0

0 

. α

Zq V0

1+

1+  M α. + Mq

Zα V0

− gsinθ0

0

gsinθ0 − gM α. sinθ0 V0

0

0

0

0

0

0

1

0

0

A xα V0 Azα V0

0

0

0

A xV

0

0

AzV.

.

A xα. V0 Azα. V0



          

∆V ∆α ∆q ∆θ ∆a x ∆az



Xδ e Zδe   V0      Z e Mα   Mδe + δV 0 +    0     A xδe Azδe 

      ∆δe    

(15)

The previous part shows that the linearized aircraft dynamics may be structured and expressed in the form of linear state-space equations. To implement the flight control system of the target UAV in real system, the equivalent state-space equations in discrete time domain were obtained by using system identification process. The used discrete-time mathematical model with no physical meaning, and the details of system identification process and discrete-time model are presented in Section 3. 2.3. Total Energy Model For total energy model, assuming that the rate of work (power) done by the engine, PE , is . correspond to the total specific energy rate error, Es , with no power loss. In addition, assuming that the torque, τ, is constant, we will have . Es ∝ rpm (16) where rpm is the revolutions per minute of engine. According to Hsieh [21], engine rotational speed . is a nonlinear function of airspeed and throttle area, which implies that Es is a nonlinear function of airspeed and throttle:   .

Es = f V, δT , V 2 , δT 2 , V 3 , δT 3

(17)

Assume that the change of airspeed is small, and then Equation (17) can be linearized. This result provides that the change of throttle is small around a trim throttle setting. .

∆ Es = µ∆δT

(18)

where µ is a constant. 3. Aircraft System Identification Aircraft system identification is the process of determining the unknown parameters of the adequate mathematical model [22], and the parameters have to be determined indirectly from the measured data. The flow chart of system identification process is presented in Figure 1. It begins with aircraft model postulation based on a priori knowledge about aircraft dynamic, aerodynamic and energy. After that, experiments are designed to measure the necessary output variables that are postulated. The experiment consists of deciding the operating flight conditions of the aircraft and designing of aircraft optimal control input to excite the dynamic motion of the aircraft. Next, the measured data undergo data compatibility analysis to compare with the reconstructed aircraft responses based on known rigid body kinematics. With the available input–output data, the parameters of the model can be estimated. The final step will be model validation with different set of input–output data. The identified model must demonstrate that its parameters have physically reasonable values and acceptable accuracy and that the model has good prediction capability on comparable aircraft maneuvers [23].

Appl. Sci. 2016, 6, 369

6 of 24

Appl. Sci. 2016, 6, 369 

6 of 24 

Experiment Design (Input Design)

Model Postulation

Measured Data Data Compatibility Analysis Input-output Data Parameter / State Estimation

Model Validation

Final Model

 

Figure 1. Flow chart of system identification process. Figure 1. Flow chart of system identification process. 

3.1. Input Design 3.1. Input Design  The objective of input design for aircraft system identification is to excite the aircraft so that the The objective of input design for aircraft system identification is to excite the aircraft so that the  data contain sufficient information for accurate modeling, subjected to the practical constraints of data  contain sufficient  information  for accurate  modeling,  subjected  to  the  practical  constraints  of  the experiment [23]. The common longitudinal flight maneuvers for aircraft system identification the  experiment  [23].  The  common  longitudinal  flight  maneuvers  for  aircraft  system  identification  consist of short-period maneuver, phugoid maneuver, thrust variation, et cetera, which will be able to consist of short‐period maneuver, phugoid maneuver, thrust variation, et cetera, which will be able  excite the longitudinal motion of aircraft, that is, short-period mode or phugoid motion. According to  excite  the  longitudinal  motion  of  aircraft,  that  is,  short‐period  or  phugoid  motion.  to Jategaonkar [22], short-period motion is a fast responding longitudinal mode  mode, provides the most According to Jategaonkar [22], short‐period motion is a fast responding longitudinal mode, provides  information to enable the estimation of derivatives pertaining to the vertical and pitching motion. Since the most information to enable the estimation of derivatives pertaining to the vertical and pitching  phugoid mode is the energy exchange, it is more suitable than short-period mode to excite the energy characteristic of the aircraft, which is the function of pitch angle, angle of attack, and longitudinal motion. Since phugoid mode is the energy exchange, it is more suitable than short‐period mode to  acceleration. The methods of utilizing the flight maneuvers in input design are called optimal input excite the energy characteristic of the aircraft, which is the function of pitch angle, angle of attack,  where the aircraft system the approach of a priori knowledge about the are  and design, longitudinal  acceleration.  The identification methods  of  uses utilizing  the  flight  maneuvers  in  input  design  dynamic system response and the input is tailored accordingly. One of the popular optimal inputs is called optimal input design, where the aircraft system identification uses the approach of a priori  multistep input design where the range of aircraft natural frequencies are investigated, then followed knowledge  about  the  dynamic  system  response  and  the  input  is  tailored  accordingly.  One  of  the  by designing of suitable multistep input to cover the desired frequency range. popular optimal inputs is multistep input design where the range of aircraft natural frequencies are  A more common multistep input is doublet input, as shown in Figure 2. The doublet input is a investigated, then followed by designing of suitable multistep input to cover the desired frequency  two-sided pulse, resulting with a symmetrical signal, which has higher energy and wider frequency range.    bandwidth compared to pulse input. The time step of doublet input can be approximated by:

A more common multistep input is doublet input, as shown in Figure 2. The doublet input is a  1 two‐sided pulse, resulting with a symmetrical signal, which has higher energy and wider frequency  × Tosc (19) ∆tDBLT = 2.7 bandwidth compared to pulse input. The time step of doublet input can be approximated by: 

Energy

where ∆tDBLT is the time step of doublet input and Tosc 1 is the period of aircraft oscillation. = of system ×   ∆ DBLT The input design is a part of the iterative process identification, until an adequate model (19) 2.7 is developed and finally the model is successfully with the implementation of the controller. In this where  ∆ DBLT is the time step of doublet input and    is the period of aircraft oscillation.  study, the  energy distribution model consists of elevator input and total energy model consists of throttle input. For aircraft longitudinal motion, without significant influence from lateral dynamics, we can assume that energy distribution model is independent of total energy model. The suitable input  t = 0.6 s 1 for elevator is a doublet signal with ∆tDBLT = 1 s. The maximum amplitude  t = 0.8 s is eight degree and the s  t = 1.2 minimum amplitude is −10 degree, as shown in Figure 3. The maximum positive elevator deflection 0.5 is smaller to prevent the aircraft from excessive pitch down. The multi-steps input in the Figure 3 was suggested by Klein [23]. It is chosen due to the rich frequency spectrum covered in the input excitation 0 quality of dynamic response data. which would result in better 0 2 4 6 8 10 12 Frequency, (rad/s)

Amplitude

0.5 0 -0.5 0

1

2

3

4

5 Time, (s)

6

7

8

9

10

 

A more common multistep input is doublet input, as shown in Figure 2. The doublet input is a  two‐sided pulse, resulting with a symmetrical signal, which has higher energy and wider frequency  bandwidth compared to pulse input. The time step of doublet input can be approximated by:  ∆ Appl. Sci. 2016, 6, 369

where  ∆

DBLT

=

1 × 2.7

DBLT   is the time step of doublet input and 

(19)

 

7 of 24

  is the period of aircraft oscillation. 

Appl. Sci. 2016, 6, 369 

7 of 24   t = 0.6 s

1

e (degree)

Amplitude

Energy

 t = 0.8 s The  input  design  is  a  part  of  the  iterative  process  of  system  identification,  until  an  adequate   t = 1.2 s model is developed and finally the model is successfully with the implementation of the controller.  0.5 In  this  study,  the  energy  distribution  model  consists  of  elevator  input  and  total  energy  model  consists of throttle input. For aircraft longitudinal motion, without significant influence from lateral  0 Appl. Sci. 2016, 6, 369  7 of 24  0 2 4 6 8 10 12 dynamics, we can assume that energy distribution model is independent of total energy model. The  Frequency, (rad/s) suitable input for elevator is a  signal with  ∆ DBLT = 1 s. The maximum amplitude is eight  The  input  design  is  a  part doublet  of  the  iterative  process  of   system  identification,  until  an  adequate  degree  and  the  minimum  amplitude  is  −10  degree,  as  shown  in  Figure  3.  The  maximum  positive  0.5 model is developed and finally the model is successfully with the implementation of the controller.  elevator  deflection  is  smaller  to  prevent  the  aircraft  from  excessive  pitch and  down.  The  multi‐steps  In  this  study,  the  energy  distribution  model  consists  of  elevator  input  total  energy  model  0 input in the Figure  3  was  suggested by Klein [23]. It is chosen due to the rich  frequency spectrum  consists of throttle input. For aircraft longitudinal motion, without significant influence from lateral  covered in the input excitation which would result in better quality of dynamic response data.  -0.5 dynamics, we can assume that energy distribution model is independent of total energy model. The  The input of throttle is a pulse signal with  ∆ = ∆3 s, as shown in Figure 4. The determination of  suitable input for elevator is a  doublet  signal with    = 1 s. The maximum amplitude is eight  DBLT 0 1 2 3 4 5 6 7 8 9 10 the  duration  and  magnitude  for  this  input  was  based  on  the  flight  simulations  and  Time, (s) degree  and  the  minimum  amplitude  is  −10  degree,  as  shown  in results  Figure  of  3.  The    maximum  positive  experiments. Moreover, in  throttle response,  there  is  a lag  time  constant,  which  is The  typically  of  the  elevator  deflection  is  smaller  to  prevent  the  aircraft  from  excessive  pitch  down.  multi‐steps  Figure 2. Doublet input with different time steps.  Figure 2. Doublet input with different time steps. order of 2 to 3 s. This is the reason why the duration of 3 s is chosen for the input design of throttle.  input in the Figure  3  was  suggested by Klein [23]. It is chosen due to the rich  frequency spectrum  covered in the input excitation which would result in better quality of dynamic response data.  The input of throttle is a pulse signal with  ∆ =  3 s, as shown in Figure 4. The determination of  10 the  duration  and  magnitude  for  this  input  was  based  on  the  results  of  flight  simulations  and  0 throttle response,  there  is  a lag  time  constant,  which  is  typically  of  the  experiments. Moreover, in  order of 2 to 3 s. This is the reason why the duration of 3 s is chosen for the input design of throttle.  -10

0.5

1

1.5 2 Time (s)

2.5

3

3.5

 

e (degree)

0 10

Figure 3. Elevator doublet with time step of 1 s as input.  0 Figure 3. Elevator doublet with time step of 1 s as input. -10

T (pwm)

The input of throttle is a20pulse signal with ∆t = 3 s, as shown in Figure 4. The determination of the 0 0.5 2.5 3 3.5 duration and magnitude for this input was1based1.5on the2results of flight simulations and experiments. Time (s)   10 Moreover, in throttle response, there is a lag time constant, which is typically of the order of 2 to 3 s. Figure 3. Elevator doublet with time step of 1 s as input.  This is the reason why the duration of 3 s is chosen for the input design of throttle. 0

T (pwm)

0 20

1

2

3 4 Time (Sec)

5

6

7

 

Figure 4. Thrust pulse with time step of 3 s as input. 

10

3.2. Prediction Error Method (PEM)  0 0 1 3 parameter  4 5estimation.  6 7 In  this  study,  PEM  was  applied  to 2perform  According  to  Ljung  [24],  Time (Sec)   to  quite  arbitrary  PEMs  are  a  broad  family  of  parameter  estimation  methods  that  can  be  applied  model  parameterizations  and  these  methods  have  a  close  kinship  with  the  maximum  likelihood  Figure 4. Thrust pulse with time step of 3 s as input.  Figure 4. Thrust pulse with time step of 3 s as input. method.  The  state‐space  form  of  energy  distribution  is  then  readily  available  to  be  identified  by  3.2. Prediction Error Method (PEM)  using  the  PEM  function  in  MATLAB  System  Identification  Toolbox.  The  objective  of  system  3.2. Prediction Error Method (PEM) identification is to identify the discrete time state‐space model in the following form:  In  this  study,  PEM  was  applied  to  perform  parameter  estimation.  According  to  Ljung  [24], 

In thisare  study, PEMfamily  was applied to perform parameter estimation. According to Ljung [24],(20) PEMs PEMs  a  broad  of  parameter  estimation  ( + 1) = ( ) +methods  ( ) +that (can  )  be  applied  to  quite  arbitrary  are amodel  broadparameterizations  family of parameter estimation methods that can be applied to quite arbitrary model and  these  methods  have  a  close  kinship  with  the  maximum  likelihood  parameterizations and these form  methods have adistribution  with the maximum likelihood method. (close ) = kinship ( )+ ( )  readily  (21) method.  The  state‐space  of  energy  is  then  available  to  be  identified  by The state-space form of energy is System  then readily available to be identified by using the PEM using  the  PEM  function distribution in  MATLAB  Identification  Toolbox.  The  objective  of  system  where k is the number of time‐step.  Note  that x, y, u, w, and v are the state vector, output vector,  function in MATLAB System Identification Toolbox. The objective of system identification is to identify identification is to identify the discrete time state‐space model in the following form:  input vector, process noise vector, and measurement noise vector, respectively. The remaining A, B,  the discrete time state-space model in the following form: C,  and K are the so‐called system matrices, which determine the  (20) ( + 1) = ( ) + ( ) + ( dynamics of the aircraft motion,  )  and they are known to be constant  with respect to a particular trimmed flight condition.  Equation  x (k + 1) = Ax k) +(Bu ( ) (= ) +(k )( +) Kw (k ) (21)(20) (21) is the output equation, and C is output matrix which represents the rationale of a linear relation  between  the  state  variables  and  the  output.  The used state vector, output vector, and input vector  where k is the number of time‐step.  Note  y (k )that x, y, u, w, and v are the state vector, output vector,  = Cx (k) + v (k) (21) are presented as follows:  input vector, process noise vector, and measurement noise vector, respectively. The remaining A, B, 

where is the number of time-step. Note that x, y, u, w, and v are the state vector, output vector, input C, kand K are the so‐called system matrices, which determine the  dynamics of the aircraft motion,  vector, process noise vector, and measurement noise vector, respectively. The remaining Equation  A, B, C, and and they are known to be constant  with respect to a particular trimmed flight condition.  (21) is the output equation, and C is output matrix which represents the rationale of a linear relation  between  the  state  variables  and  the  output.  The used state vector, output vector, and input vector  are presented as follows: 

Appl. Sci. 2016, 6, 369

8 of 24

K are the so-called system matrices, which determine the dynamics of the aircraft motion, and they are known to be constant with respect to a particular trimmed flight condition. Equation (21) is the output equation, and C is output matrix which represents the rationale of a linear relation between the state variables and the output. The used state vector, output vector, and input vector are presented as follows: x = [∆V ∆α ∆q ∆θ ∆ax ∆az ] T (22) h . iT y = ∆q ∆ L (23) u = [∆δe ]

(24) .

.

The rationale of a linear relation between the state variables and the ∆ L is based on that the ∆ L is the function of the longitudinal acceleration and the flight path angle, and the flight path angle is the function of the pitch angle and angle of attack. In Equation (23), ∆q was selected to be the output state . instead of ∆θ since ∆q is equal to ∆θ and can be directly measured from the inertial sensor. The predictor used to compute the one-step-ahead prediction for an underlying system description is the Kalman filter provided that w and v are Gaussian process. The predictor form of state-space representation can be given by: xˆ (k + 1, ξ ) = A (ξ ) xˆ (k, ξ ) + B (ξ ) u (k, ξ ) + K (ξ ) [y (k) − C (ξ ) xˆ (k, ξ )] yˆ (k|ξ ) = C (ξ ) xˆ (k, ξ )

(25)

where xˆ and yˆ are the predicted states and predicted outputs, respectively, and ξ is a vector of parameters that typically correspond to unknown values of physical coefficients. The prediction error, ε, is given by: ε (k, ξ ) = y (k) − yˆ (k|ξ ) (26) One of the approaches to qualify how small the ε is to form a scalar value norm or criterion function that measures the size of N. Consider a set of N data from an aircraft is given by: Z N = [y (1) , u (1) , y (2) , u (2) , . . . , y ( N ) , u ( N )]

(27)

The corresponding norm:   1 V N ξ, Z N = N

N

∑ ||ε (k, ξ ) ||2

(28)

k =1

Thus, the essence of PEM is the estimation of ξ from the model parameterization and ZN so that the norm VN is minimized. 3.3. Aircraft Model Evaluation Method In order to choose the best model among the identified models, it is required to apply a model evaluation method to evaluate the performance of the models. The used method in this study is adopted from Lee [18], which is variance accounted for (VAF) and norm. Given a set of N data, VAF is defined by the equation below:  VAF =

1−

var (yi − yˆ i ) var (yi )



× 100%, i = 1, 2, 3, . . . , N

(29)

where var(.) is the variance of the variable in the bracket, yi is the measured output, and yˆ i is the predicted output of the identified model. VAF is expressed in percentage where 100% means the outputs of identified model fit the measured output perfectly. However, there is a possibility to encounter with negative percentage, simply indicating that the outputs of the identified model do not fit the measured output. The norm is given by:

Appl. Sci. 2016, 6, 369

9 of 24

 NRM =

yˆ i − yi yi

T 

yˆ i − yi yi

 , i = 1, 2, 3, . . . , N

(30)

where a smaller value of norm will correspond to a better identified model. There are total of 55 longitudinal flight data collected, performed by using elevator doublet input, Appl. Sci. 2016, 6, 369  9 of 24  with adequate elevator deflection angles. The flight data are named on the form as shown below for thebelow for the ease of data management and analysis, as an example ft05run11 referred to the fifth  ease of data management and analysis, as an example ft05run11 referred to the fifth flight test and theflight test and the eleventh run. The aircraft distribution energy models were identified by imposing  eleventh run. The aircraft distribution energy models were identified by imposing the performance requirements as shown below: the performance requirements as shown below:  VAF ≥ 85% and NRM ≤ 15 VAF ≥ 85% and NRM ≤ 15  The values of VAF have to be greater or equal to 85% and the norms must be smaller or equal to The values of VAF have to be greater or equal to 85% and the norms must be smaller or equal to  15.15. There are total of 13 identified aircraft energy distribution models along with the trim conditions  There are total of 13 identified aircraft energy distribution models along with the trim conditions of theof flight, VAF and norm. example of an identified model ft11run10 that fulfillsthat  the fulfills  performance the  flight,  VAF  and  An norm.  An  example  of  an  identified  model  ft11run10  the  requirements is shown in Figure 5. Identified models will be short-listed, and cross-validation performance  requirements  is  shown  in  Figure  5.  Identified  models  will  be  short‐listed,  and  is performed to further evaluate the performance of the model. Figure 6 presents the cross-validation cross‐validation is performed to further evaluate the performance of the model. Figure 6 presents the  result of model ft10run07 and data ft10run15. Cross-validation of model ft10run07 with other flight cross‐validation result of model ft10run07 and data ft10run15. Cross‐validation of model ft10run07  data is shown in Table 1. The discrete-time state-space of ft10run07state‐space  model, identified at trimmodel,  airspeed with  other  flight  data  is  shown  in  Table  1.  The  discrete‐time  of  ft10run07  of identified at trim airspeed of 27.44 m/s and altitude of 150.8 m, is given by Equations (20)–(24). From  27.44 m/s and altitude of 150.8 m, is given by Equations (20)–(24). From Figure 5, the VAF value of .   are  92.31%  and both 94.52%,  respectively,  which  both  values  the  VAF  value  of  ∆ and  ∆qFigure  and Ls5, are 92.31% and 94.52%, respectively, in which values satisfy thein  minimum requirement satisfy the minimum requirement of VAF at 85% based on the our experience.    of VAF at 85% based on the our experience. Model: ft11run10 Data: ft11run10 1.5 Model Measurement

q, (rad/s)

1 0.5 0 -0.5 -1

0

0.5

1

1.5

2

2.5

3

2.5

3

 Ldot

0.5

0 Model Measurement -0.5

0

0.5

1

1.5

2

q Ldot

Error

0.5 0 -0.5 0

0.5

1

1.5

2

2.5

3

0

0.5

1

1.5 Time, (Sec)

2

2.5

3

e, (degree)

10 5 0 -5 -10

Figure 5. An example of identified model ft11run10.  Figure 5. An example of identified model ft11run10.

 

Appl. Sci. 2016, 6, 369

10 of 24

Appl. Sci. 2016, 6, 369 

10 of 24  Model: ft10run07 Data: ft10run15 1.5 Model Measurement

q, (rad/s)

1 0.5 0 -0.5 -1

0

0.5

1

1.5

2

2.5

3

3.5

0.6 Model Measurement

Ldot

0.4 0.2 0 -0.2 -0.4

0

0.5

1

1.5

2

2.5

3

3.5

 

1

q

Error

0.5

 Ldot

0 -0.5 -1

0

0.5

1

1.5

2

2.5

3

3.5

0

0.5

1

1.5 2 Time, (Sec)

2.5

3

3.5

e, (degree)

10 5 0 -5 -10

 

Figure 6. Model ft10run07 cross‐validated with data ft10run15.  Figure 6. Model ft10run07 cross-validated with data ft10run15. Table  1.  Cross‐validation  model  ft10run07 with with  other  variance  accounted  for; for; Table 1. Cross-validation of of  model ft10run07 other flight  flightdata.  data.VAF,  VAF, variance accounted NRM, norm.  NRM, norm. Model   

Model

V (m/s)

VAF (%) NRM ∆VAF (%) ∆ ∆ ∆ 151.0  66.99  85.37  6.09  8.58  ∆q ∆L 153.4  67.12  93.57  7.22  7.48  66.9978.12  91.99  85.37 150.5  9.27  9.84  67.1275.54  92.44  93.57 150.7  7.99  7.21  78.12 ‐ ‐ 91.99‐ ‐ 150.8  75.5457.98  86.56  92.44 151.5  3.85  4.77  150.6 - 54.52  91.90  - 5.36  5.85  57.9865.05  94.13  86.56 150.2  5.33  5.61  54.5289.56  94.82  91.90 5.37  6.48  151.0  65.0558.92  87.65  94.13 150.7  5.90  3.88  89.5671.75  84.84  94.82 150.8  8.63  7.51  58.9276.16  91.03  87.65 151.0  2.97  6.07  71.7589.49  90.87  84.84 156.7  4.68  4.05 

V (m/s)  h (m) 

h (m)27.03  ft09run06  ft09run07  26.50  151.029.33  ft10run01  153.428.43  ft10run02  150.527.44  ft10run07  150.728.35  ft10run14  150.829.10  ft10run15  151.528.73  ft10run17  150.628.80  ft10run21  150.227.98  ft11run07  151.027.65  ft11run10  150.728.40  ft11run25  150.829.25  ft11run28 

NRM ∆q

∆L

ft09run06 27.03 6.09 8.58 ft09run07 26.50 7.22 7.48 ft10run01 29.33 9.27 9.84 ft10run02 28.43 7.99 7.21 ft10run07 27.44 ft10run14 28.35 3.85 4.77 ft10run15 29.10 5.36 5.85 ft10run17 28.73 5.33 5.61 ft10run21 28.80 5.37 6.48 ft11run07 27.98 5.90 3.88 ft11run10 27.65 8.63 7.51 ft11run25 28.40 151.0 76.16 91.03 2.97 6.07 ft11run28 29.25 156.7 89.49 90.87 4.68 4.05 Since model ft10run07 has better performance than others, it was adopted to be the model for 

control design. The identified discrete‐time system matrices of model ft10run07 are shown below: 

Since model ft10run07 has better performance than others, it was adopted to be the model for control design. The identified discrete-time system matrices of model ft10run07 are shown below:

Appl. Sci. 2016, 6, 369

11 of 24

     A=   

Appl. Sci. 2016, 6, 369 

 0.8817  0.1233   0.1390 A  0.3292  0.0125   0.1440

−0.1550 0.3277 −0.1030 −0.1448 0.3950 −0.4245 −0.5860 0.3802 0.8934 0.3057 0.1676 −0.0735 −0.5987 −0.5262 −0.1264 −0.0310 0.4682 −0.5089   −0.0036  0.0049     −0.0289    B=   0.0523     0.0315  −0.0173 0.3277 0.1030 0.0179  0.0036     0.3950 0.4245 0.0044 0.0658  −0.5807  0.0049       −0.1267 0.3802 0.8934 0.2023 −0.1459  0.0289     B     −0.1427 0.1676 0.0735  0.4707  −1.3377  0.0523  K=  0.5262 0.1264 0.6399  −0.5222  0.0315   −0.1539      −0.0718 0.4682 0.5089 0.5017  −0.1018  0.0173 −0.3711 −1.7641

0.8817 −0.1233 −0.1390 −0.3292 0.0125 −0.1440

0.3071 0.1550 0.5893 0.1448 0.1834 0.5860 0.7071 0.3057 0.0863 0.5987 0.4595 0.0310

−0.3071 0.5893 0.1834 −0.7071 −0.0863 −0.4595

0.0179 0.0658 −0.2023 0.4707 0.6399 0.5017

C=

       

11 of 24 

 0.0044  0.1267   0.1427 K  0.1539  0.0718   0.3711

1.8588 0.2620 0.6264 0.4888 0.3138   0.6142 −0.6142 −1.8588 0.2620 0.6264 −0.4888 0.3138 C 0.6276 0.1134 0.0499 0.0579 0.1528  1.2955 −1.2955 0.6276 0.1134 −0.0499 −0.0579 0.1528

"



0.5807 0.1459   1.3377  0.5222 0.1018  1.7641

#

3.4. Total Energy Model 3.4. Total Energy Model  In of In  Equation Equation  (18), (18),  the the total total energy energy is is proportional proportional to to the the deflection deflection  of throttle, throttle, ∆δ ∆δTT..  The The  main main  . objective this section is tois find out the relationship between ∆ Es and∆∆δT  ,and  described by the parameter ∆δ objective for for  this  section  to  find  out  the  relationship  between  ,  described  by  the  T . µ. The ∆ Es is obtained applying the ∆δT as input with time step of 3 s. An example of Pulse Width is obtained by applying the  ∆δT   aas input with a time step of 3 s. An example  parameter  μ. The  ∆  by Modulation (PWM) input of ∆δT with a magnitude of 15 µs is shown in Figure We assume that the of Pulse Width Modulation (PWM) input of  ∆δT   with a magnitude of 15  μs  7. is shown in Figure 7.  width of the input is proportional to the deflection of throttle. For each input with different magnitudes, We assume that the width of the input is proportional to the deflection of throttle. For each input  . with different magnitudes, five data were collected to obtain the average value of  ∆ , as listed in  five data were collected to obtain the average value of ∆ Es , as listed in Table 2. The average values of . Table  2.  The  average  values  of  measured  ∆   are  plotted  against  the  ∆δ ,  and  the  curve  is  fitted  measured ∆ Es are plotted against the ∆δT , and the curve is fitted with a third as T degree polynomial, with a third degree polynomial, as shown in Figure 8. The slope of the fitted curve at zero PWM in  shown in Figure 8. The slope of the fitted curve at zero PWM in Figure 8 is corresponding to the value Figure 8 is corresponding to the value of 0.0006522.  of 0.0006522.

RPM

7050 7000 6950

0

1

2

3

4

5

6

0

1

2

3

4

5

6

0

1

2

3 Time (Sec)

4

5

6

Edot

0.01 0.005

T (pwm)

0 20 10 0

.

 

Figure 7. The response of the ∆ Es  with an increment of throttle input. with an increment of throttle input.  Figure 7. The response of the 

Table 2. The average value of  ∆   from positive throttle input. PWM, Pulse Width Modulation.  ∆ Flight Tests ft33run10  ft33run11  ft33run12 

( )= V (m/s)  h (m)  ∆ 30.0  150.5  0.0050  30.2  152.6  0.0045  30.0  150.7  0.0050 

Appl. Sci. 2016, 6, 369

12 of 24 .

Table 2. The average value of ∆ Es from positive throttle input. PWM, Pulse Width Modulation. ∆δT (PW M ) = 15 .

Flight Tests

V (m/s)

h (m)

∆Es

ft33run10 ft33run11 ft33run12 ft33run13 ft33run14

30.0 30.2 30.0 30.0 30.1

150.5 152.6 150.7 151.0 153.0

0.0050 0.0045 0.0050 0.0050 0.0050

Average

-

-

0.0049

Appl. Sci. 2016, 6, 369 

12 of 24 

0.02 Measured data Curve fit data

0.01

0

-0.01

E dot

-0.02

-0.03

-0.04

-0.05

-0.06

-0.07 -80

-60

-40

-20

0

20

40

60

 PWM

 

  and  Figure 8. Curve fit relation of  Figure 8. Curve fit relation of ∆∆E and ∆δ∆δ. T .  .

s

T

4. Optimal Energy Control System  4. Optimal Energy Control System 4.1. OECS Design  4.1. OECS Design In this study, an OECS has been proposed to form the aircraft longitudinal flight control system In this study, an OECS has been proposed to form the aircraft longitudinal flight control system  and controlthe  theairspeed  airspeed and  and altitude  altitude of There are two parameters, specific energy and  to tocontrol  of the the aircraft. aircraft.  There are  two  parameters,  specific  energy  . . distributionrate,  rate, Ls, , and specific energy rate,rate,  Es , utilized to characterize the energy an aircraft distribution  and total total  specific  energy  ,  utilized  to  characterize  the ofenergy  of  an  or UAV. The block diagram of the proposed OECS is shown in Figure 9. The block diagram of the aircraft or UAV. The block diagram of the proposed OECS is shown in Figure 9. The block diagram  total energy loop is used to perform the tracking of aircraft altitude. Altitude tracking begins with the of the total energy loop is used to perform the tracking of aircraft altitude. Altitude tracking begins  commanded altitude, hc . The difference between the commanded altitude and the current altitude with the commanded altitude,  ℎ . The difference between the commanded altitude and the current  of the aircraft, h, is the altitude error, h which will be scheduled by gains, Kh . The values . The  of e altitude of the aircraft, h, is the altitude error,  ℎin turn which in turn will be scheduled by gains,  K were decided according to the tests of flight simulations. Table 3 shows the gains used in the gain h values of    were decided according to the tests of flight simulations. Table 3 shows the gains used  scheduling. This simple gain scheduling is necessary to guarantee the throttle response is small, and in the gain scheduling. This simple gain scheduling is necessary to guarantee the throttle response is  thus meet the linearized total energy model, which stated that the increment of total specific energy . small,  and  thus  meet  the  linearized  total  energy  model,  which  stated  that  the  increment  of  total  rate, E , is proportional to ∆δT . The result of gain scheduling is the climb rate or descent rate of an specific  eenergy  rate,  ,  is  proportional  to  ∆δT .  The  result  of  gain  scheduling  is  the  climb  rate  or  aircraft, which in turn will be divided by airspeed to obtain the flight path angle error, γe . By assuming descent rate of an aircraft, which in turn will be divided by airspeed to obtain the flight path angle  that the airspeed is almost constant, thus the longitudinal acceleration will be neglected. error,  γ . By assuming that the airspeed is almost constant, thus the longitudinal acceleration will be  neglected.  Energy distribution and pitch stabilization loop ∆

Throttle 1

LQG Regulator +

1

d

Aircraft Dynamics

∆ ∆ ,∆

values of    were decided according to the tests of flight simulations. Table 3 shows the gains used  in the gain scheduling. This simple gain scheduling is necessary to guarantee the throttle response is  small,  and  thus  meet  the  linearized  total  energy  model,  which  stated  that  the  increment  of  total  specific  energy  rate,  ,  is  proportional  to  ∆δT .  The  result  of  gain  scheduling  is  the  climb  rate  or  descent rate of an aircraft, which in turn will be divided by airspeed to obtain the flight path angle  Appl. Sci. 2016, error, 6,γ369 . By assuming that the airspeed is almost constant, thus the longitudinal acceleration will be  13 of 24 neglected.  Energy distribution and pitch stabilization loop ∆

Throttle 1

LQG Regulator +

1

d

Aircraft Dynamics

∆ ∆ ,∆

1

V



ℎ ℎ

h -

Total energy loop

 

Figure  9.  The block  diagram  of  the  proposed  OECS.  OECS,  optimal  energy  control  system;  LQG,  Figure 9. The block diagram of the proposed OECS. OECS, optimal energy control system; linear‐quadratic‐Gaussian. 

LQG, linear-quadratic-Gaussian.

Table 3. The gains used in the gain scheduling. No.

he = |hc − h|

Kh

1 2 3 4

0 ≤ he ≤ 2 2 < he ≤ 5 5 < he ≤ 10 he > 10

0.130 0.045 0.025 0.008 .

The inner loop is energy distribution and pitch stabilization loop where the Ls is regulated to zero. The inner loop, controlled by elevator, is responsible to distribute the energy of the aircraft between airspeed and altitude. The outer loop is total energy loop, which corresponds to the total energy of the aircraft. Total energy loop, controlled by throttle, is responsible to drive the total specific energy rate . error ∆ Es to zero. 4.2. LQG Regulator .

The objective of LQG regulator is to regulate Ls and q to zero so that the aircraft’s energy is distributed among airspeed and altitude and the aircraft will achieve pitch-axis stabilization. This is . completed by feeding the measured Ls and q into the LQG regulator and the corresponding elevator command is computed. The design of LQG regulator is facilitated by the embedded functions in MATLAB. Eventually, a closed loop energy distribution state space model can be formed by using a series of functions in MATLAB. The process begins with the determination of optimal feedback gain of the regulator using the command dlqr. LQG regulator state-space model can be formed with the combination of Kalman estimator state-space model and LQR optimal feedback gain. For more details of LQG regulator design for a small UAV refer to studies [18,19]. 5. Simulation Results and Discussion In order to examine the feasibility and performance of the developed OECS, the simulations of OECS were conducted in the HIL system of Spoonbill UAV. Three flight maneuvers, level flight, descending, and climbing, were adopted to perform the simulations. Moreover, the influence of the wind effect was considered to validate the reliability of the developed system. The largest magnitude of wind was set to 8 m/s and the wind direction was set to be a headwind in the simulations. 5.1. Hardware-in-the-Loop System of Spoonbill UAV In order to validate the proposed OECS, a HIL system of Spoonbill UAV, as shown in Figure 10, was designed by integrating the onboard computer of Spoonbill UAV, a self-developed graphic user interface (GUI) program, and the X-Plane flight simulator, developed by Laminar Research. X-Plane can communicate with other applications via User Datagram Protocol (UDP). The GUI of the program,

Appl. Sci. 2016, 6, 369

14 of 24

written using C++Builder (2007, Scotts Valley, CA, USA, 2007), is used to form the communication between the onboard computer and X-Plane Simulator. In the onboard computer, a flight program was used to integrate and save the flight data, and carry out the flight control at the rate of 20 Hz. Figure 11 shows the architecture of Spoonbill HIL system, and Figure 12 presents the communication between X-Plane and onboard computer with the help of GUI program. Appl. Sci. 2016, 6, 369 

14 of 24 

Appl. Sci. 2016, 6, 369 

14 of 24 

Appl. Sci. 2016, 6, 369 

14 of 24 

  Figure 10. Spoonbill unmanned aerial vehicle (UAV) in X‐Plane flight simulator.   

Figure 10. Spoonbill unmanned aerial vehicle (UAV) in X-Plane flight simulator. Figure 10. Spoonbill unmanned aerial vehicle (UAV) in X‐Plane flight simulator.    Onboard Computer (PC/104) Figure 10. Spoonbill unmanned aerial vehicle (UAV) in X‐Plane flight simulator.   Onboard Computer (PC/104) COM Port 5 COM Port 5

COM Port 6 Onboard Computer (PC/104) COM Port 6

COM Port 4 COM Port 4

Graphic User Interface COM Port 5 COM Port 6 COM Port 4 PWM (UDP) Wireless Graphic User Interface Decoder PWM (UDP) Wireless Decoder Graphic User Interface PWM (UDP) RC Wireless Airspeed, Servo Decoder AHRS GPS Receiver Altitude, Controller RC Airspeed, Servo rpm AHRS GPS Receiver Altitude, Controller Ground RC rpm Airspeed, Servo AHRS GPS RC Receiver Control Altitude, Controller Ground Controller Station X-Plane Environment rpm RC Control Controller Station X-Plane Environment Ground RC Control Controller Station X-Plane Environment Figure 11. Architecture of Spoonbill HIL system. HIL, hard‐in‐the‐loop; COM, communication port; 

 

 

Figure 11. Architecture of Spoonbill HIL system. HIL, hard‐in‐the‐loop; COM, communication port;    system;  Pulse  Width  Modulation; HIL RC,  radio  control;  AHRS,  attitude  and  heading  reference  Figure 11.PWM,  Architecture of Spoonbill system. HIL, hard-in-the-loop; COM, communication port; PWM,  Pulse  Width  Modulation;  RC,  radio  control;  AHRS,  attitude  and  heading  reference  system;  GPS, Global Positioning System; UDP, User Datagram Protocol.   Figure 11. Architecture of Spoonbill HIL system. HIL, hard‐in‐the‐loop; COM, communication port;  PWM, Pulse Width Modulation; RC, radio control; AHRS, attitude and heading reference system; GPS, GPS, Global Positioning System; UDP, User Datagram Protocol.   and  heading  reference  system;  PWM,  Pulse  Width  Modulation;  RC,  radio  control;  AHRS,  attitude 

Global Positioning System; UDP, User Datagram Protocol.

GPS, Global Positioning System; UDP, User Datagram Protocol. 

    Figure 12. Communication between X‐Plane and onboard computer with the help of GUI program.  Figure 12. Communication between X‐Plane and onboard computer with the help of GUI program.    GUI, graphic user interface.  GUI, graphic user interface.  Figure 12. Communication between X‐Plane and onboard computer with the help of GUI program. 

Figure 12.GUI, graphic user interface.  Communication between X-Plane and onboard computer with the help of GUI program. GUI, graphic user interface.

With the help of HIL system, the developed OECS is first simulated in the aircraft level flight  without the presence of wind. The simulation results of level flight, as shown in Figure 13, show that  the  change  of  pitch  rate,  ∆ ,  is  regulated  within  ±0.01 rad/s  and    is  regulated  at  0.01.  At  the  same time, the airspeed is maintained at around 20.8 m/s and the altitude shows agreement with the  Appl. Sci. 2016, 6, 369 15 of 24 commanded altitude,  ℎ , at 150 m.    Furthermore, aircraft level flight, as shown in Figure 14, is simulated with the presence of wind,  and the largest magnitude of wind is 8 m/s. The results present that the change of pitch rate,  ∆ , is  5.2. Airspeed and Altitude Hold regulated within  ±0.10 rad/s  and    is regulated within 0.05. Moreover, the airspeed is regulated  With the help of HIL system, the developed OECS is first simulated in the aircraft level flight at trim airspeed of 20.3 m/s with the variation of  ±0.5 m/s. The trim airspeed is dependent on the  without the presence of wind. The simulation results of level flight, as shown in Figure 13, show . identified aircraft energy distribution model. On the other hand, the altitude deviation is less than 2  that the change of pitch rate, ∆q, is regulated within ±0.01 rad/s and Ls is regulated at 0.01. At the m  from  commanded  of  120  As  a  summary  level  flight,  performance  of  the  samethe  time, the airspeedaltitude  is maintained at m.  around 20.8 m/s and for  the altitude showsthe  agreement with the OECS satisfies the airspeed and altitude is successfully held.  commanded altitude, hc , at 150 m.

 q (rad/s)

0.05 0

Energy rate

-0.05

0

5

10

15

20

25

30

0.02 0 E dot

-0.02

Airspeed (m/s)

0

5

10

15

20

Ldot

25

30

21 20.8 Vtrim

V 20.6

0

5

10

15

20

25

30

 

Altitude (m)

155 150 h 145

hc

0

5

10

15

20

25

30

0

5

10

15

20

25

30

0

5

10

15 Time (Sec)

20

25

30

e (deg.)

-9.1 -9.2 -9.3

T (pwm)

1310 1300 1290 1280

 

Figure 13. Simulation results of level flight without the presence of wind. Figure 13. Simulation results of level flight without the presence of wind. 

Furthermore, aircraft level flight, as shown in Figure 14, is simulated with the presence of wind, and the largest magnitude of wind is 8 m/s. The results present that the change of pitch rate, ∆q, is . regulated within ±0.10 rad/s and Ls is regulated within 0.05. Moreover, the airspeed is regulated at trim airspeed of 20.3 m/s with the variation of ±0.5 m/s. The trim airspeed is dependent on the identified aircraft energy distribution model. On the other hand, the altitude deviation is less than 2 m from the commanded altitude of 120 m. As a summary for level flight, the performance of the OECS satisfies the airspeed and altitude is successfully held.

 q (rad/s)

Appl. Sci. 2016, 6, 369 Appl. Sci. 2016, 6, 369 

16 of 24

16 of 24 

0.1 0 -0.1

Energy rate

0

10

20

30

40

50

60

Edot

0.1

Ldot

0 -0.1 0

10

20

30

40

Airspeed (m/s)

22

50

60 Vtrim

V

21 20 19

0

10

20

30

40

50

Altitude (m)

123

60

h

122

 

hc

121 120 119

0

10

20

30

40

50

60

0

10

20

30

40

50

60

0

10

20

30 Time (Sec)

40

50

60

e (deg.)

-8 -9 -10

T (pwm)

1310 1300 1290 1280

 

Figure 14. Simulation results of level flight with the presence of wind. Figure 14. Simulation results of level flight with the presence of wind. 

5.3. Climbing Maneuver 5.3. Climbing Maneuver  This section presents the climbing performance of the aircraft with and without the presence of This section presents the climbing performance of the aircraft with and without the presence of  wind. The climbing maneuver was performed at the trim airspeed of 20.8 m/s, without the presence of wind. The climbing maneuver was performed at the trim airspeed of 20.8 m/s, without the presence  simulation results are shown in Figure 15. The initial altitude is 150 m and the commanded of wind. wind. The The  simulation  results  are  shown  in  Figure  15.  The  initial  altitude  is  150  m  and  the  altitude is 158 m. During the aircraft climbing, the airspeed is maintained within the variation commanded altitude is 158 m. During the aircraft climbing, the airspeed is maintained within the  of ±0.2 m/s from trim airspeed. The settling time for the climbing maneuver is around 40 s. Moreover, variation of  ±0.2 m/s  from trim airspeed. The settling time for the climbing maneuver is around 40  the change of pitch rate, ∆q, is regulated within ±0.01 rad/s, which is the same as in level flight. On . s. Moreover, the change of pitch rate,  ∆ , is regulated within  ±0.01 rad/s, which is the same as in  the other hand, Ls is regulated around the neighborhood of 0.01. Note that there is a slight decrease of . level flight. On the other hand,    is regulated around the neighborhood of 0.01. Note that there is a  Ls at 18th second, which corresponds to the increase in flight path angle of the aircraft.

slight decrease of    at 18th second, which corresponds to the increase in flight path angle of the  aircraft.  Figure 16 shows the simulation results with the presence of wind and the largest magnitude of  wind is 8 m/s throughout the climbing maneuver. During the aircraft climbing, the change of pitch  rate,  ∆ , is regulated within  ± 0.15 rad/s  and    is regulated within the variation of  ±0.1. Besides  that, the airspeed is maintained at trim airspeed within the variation of  ±1 m/s. The settling time of  altitude tracking is around 65 s. 

Appl. Sci. 2016, 6, 369

17 of 24

Figure 16 shows the simulation results with the presence of wind and the largest magnitude of wind is 8 m/s throughout the climbing maneuver. During the aircraft climbing, the change of pitch . rate, ∆q, is regulated within ± 0.15 rad/s and Ls is regulated within the variation of ±0.1. Besides that, the airspeed is maintained at trim airspeed within the variation of ±1 m/s. The settling time of Appl. Sci. 2016, 6, 369  17 of 24  altitude tracking is around 65 s.

 q (rad/s)

0.05 0 -0.05

0

10

20

30

40

50

60

70

80

90

100

Energy rate

0.05 0 E dot

Airspeed (m/s)

-0.05

0

10

20

30

40

50

60

70

80

Ldot 90

100

22 Vtrim

V 21 20

0

10

20

30

40

50

60

70

80

90

100

Altitude (m)

160

 

155 hc

h

150 0

10

20

30

40

50

60

70

80

90

100

0

10

20

30

40

50

60

70

80

90

100

0

10

20

30

40

50 Time (Sec)

60

70

80

90

100

e (deg.)

-8.5 -9 -9.5

T (pwm)

1310 1300 1290 1280

Figure 15. Simulation results of climbing without the presence of wind. Figure 15. Simulation results of climbing without the presence of wind. 

 

Appl. Sci. 2016, 6, 369

18 of 24

Appl. Sci. 2016, 6, 369 

18 of 24 

q (rad/s)

0.2 0 -0.2 0

50

100

Energy rate

0.2

150 Edot

Ldot

0

Airspeed (m/s)

-0.2

0

50

100

22

150 V

Vtrim

21 20 19

0

50

100

150

Altitude (m)

150 140 130

h

120

hc

0

50

100

150

0

50

100

150

0

50

100

150

e (deg.)

-7 -8 -9 -10 -11

T (pwm)

1340 1320 1300 1280

Time (Sec)

 

Figure 16. Simulation results of climbing with the presence of wind. Figure 16. Simulation results of climbing with the presence of wind. 

  5.4. Descent Maneuver

 

For the descent maneuver, the simulation result without the presence of the wind is shown in . Figure 17. Both the change of pitch rate, ∆q, and Ls are regulated within ±0.01 rad/s and ±0.02, respectively. The aircraft descended from 158 to 150 m in around 40 s, while maintaining the airspeed within ±0.2 m/s from trim airspeed. For the simulation results of aircraft descend with the presence . of wind, as shown in Figure 18, the errors of ∆q, Ls , airspeed, altitude, and altitude tracking settling time are almost the same as the aircraft climb maneuver.

For the descent maneuver, the simulation result without the presence of the wind is shown in  Figure 17. Both the change of pitch rate,  ∆ , and    are regulated within  ±0.01 rad/s  and  ±0.02,  respectively.  The  aircraft  descended  from  158  to  150  m  in  around  40  s,  while  maintaining  the  airspeed within  ±0.2 m/s  from trim airspeed. For the simulation results of aircraft descend with the  presence  of  wind,  ,  airspeed,  altitude,  and 19altitude  Appl. Sci. 2016, 6, 369 as  shown  in  Figure  18,  the  errors  of  ∆ ,  of 24 tracking settling time are almost the same as the aircraft climb maneuver. 

q (rad/s)

0.05 0 -0.05

0

20

40

60

80

100

120

0.05 Energy rate

E dot

Ldot

0

Airspeed (m/s)

-0.05

0

20

40

60

80

100

120

21.5 21 20.5 20

Vtrim

V 0

20

40

60

80

100

120

Altitude (m)

160 h

155

hc

150 0

20

40

60

80

100

120

0

20

40

60

80

100

120

0

20

40

60 Time (Sec)

80

100

120

e (deg.)

-8.5 -9 -9.5

T (pwm)

1300 1290 1280

Figure 17. Simulation results of descending without the presence of wind. Figure 17. Simulation results of descending without the presence of wind. 

 

Appl. Sci. 2016, 6, 369  Appl. Sci. 2016, 6, 369

20 of 24  20 of 24

 q (rad/s)

0.2 0 -0.2 0

20

40

60

80

Energy rate

0.2

Edot

120

100

120

100

120

0

Airspeed (m/s)

-0.2

0

20

40

60

22

80 Vtrim

V

20 18

0

20

40

60

80

150

Altitude (m)

100 Ldot

h

140

hc

130 120 0

20

40

60

80

100

120

0

20

40

60

80

100

120

0

20

40

60 Time (Sec)

80

100

120

e (deg.)

-8 -9 -10

T (pwm)

-11 1320 1300 1280 1260 1240 1220

 

Figure 18. Simulation results of descending with the presence of wind.  Figure 18. Simulation results of descending with the presence of wind.

5.5. Result Comparison of OECS with Fuzzy Logic Control  5.5. Result Comparison of OECS with Fuzzy Logic Control To compare the performance and fuel efficacy of the proposed OECS with other controller, a  To compare the performance and fuel efficacy of the proposed OECS with other controller, a flight flight controller controller based based on on Fuzzy Fuzzy logic logic control control was was used used in in this this study. study. The The developed developed OECS OECS isis aa  model-based method, which means that it requires the system identification process and identified model‐based method, which means that it requires the system identification process and identified  model the control system. In contrast to the method, FuzzyFuzzy  logic control (FLC) model toto design design  the  control  system.  In  contrast  to model-based the  model‐based  method,  logic  control  is(FLC) is a knowledge‐based method, which means that no model is constructed, and the design of  a knowledge-based method, which means that no model is constructed, and the design of the FLC requires some understanding of the plant’s behavior and expert’s experience. The details of the the FLC requires some understanding of the plant’s behavior and expert’s experience. The details of  adopted FLC autopilot refer to the study in [25]. the adopted FLC autopilot refer to the study in [25].  Both of the climbing and descending performance of the target UAV without the presence Both of the climbing and descending performance of the target UAV without the presence of  of wind was evaluated. Since there is no autothrottle system in the used FLC, the throttle of wind was evaluated. Since there is no autothrottle system in the used FLC, the throttle of the FLC  the FLC was set to be constant and its value was 1300 PWM. The simulation results of climbing was set to be constant and its value was 1300 PWM. The simulation results of climbing maneuver  maneuver are shown in Figure 19. The initial altitude is 150 m and the commanded altitude is 158 m. are shown in Figure 19. The initial altitude is 150 m and the commanded altitude is 158 m. During  During the aircraft climbing, the airspeed decreases dramatically about 0.8 m/s from initial airspeed. the  aircraft  climbing,  the  airspeed  decreases  dramatically  about  0.8  m/s  from  initial  airspeed.  The  The settling time for the climbing maneuver is around 30 s, and the maximum change of pitch rate, ∆q, issettling time for the climbing maneuver is around 30 s, and the maximum change of pitch rate, ∆q,  about 1.5 rad/s. When the climbing maneuver is performed, the energy rates change dramatically. is about 1.5 rad/s. When the climbing maneuver is performed, the energy rates change dramatically.  The simulation results of descending maneuver are shown in Figure 20. The results are similar to that The simulation results of descending maneuver are shown in Figure 20. The results are similar to  of the climbing maneuver. In contrast to the OECS, the climbing and descending performance and fuel that of the climbing maneuver. In contrast to the OECS, the climbing and descending performance  efficiency of the proposed OECS are better than that of FLC. The aircraft is maintained approximately and  fuel  efficiency  the  proposed  OECS  better  of  FLC.  The  aircraft  is  maintained  at trim airspeed, andof  the required throttle is are  lower thanthan  thatthat  in FLC. Moreover, the energy rates are approximately at trim airspeed, and the required throttle is lower than that in FLC. Moreover, the 

Appl. Sci. 2016, 6, 369 Appl. Sci. 2016, 6, 369  Appl. Sci. 2016, 6, 369 

21 of 24 21 of 24  21 of 24 

Energy rate Energy rate

q (rad/s) q (rad/s)

more stable and within small values. The results show that the proposed OECS can achieve the goal of energy rates are more stable and within small values. The results show that the proposed OECS can  energy rates are more stable and within small values. The results show that the proposed OECS can  theachieve the goal of the optimization of the fuel consumption.  optimization of the fuel consumption. achieve the goal of the optimization of the fuel consumption.  2 1 0

2 1 0

-1 -1 200 200 0.2 0.2 0

210 210

220 220

230 230

240 240

250 250

260 260

Airspeed (m/s) Airspeed (m/s)

280 280

E Edotdot L Ldotdot

0

-0.2 -0.2 200 200 17 17

270 270

210 210

220 220

230 230

240 240

250 250

260 260

270 270

280 280

210 210

220 220

230 230

240 240

250 250

260 260

270 270

280 280

210 210

220 220

230 230

240 240

250 250

260 260

270 270

280 280

210 210

220 220

230 230

240 240

250 250

260 260

270 270

280 280

210 210

220 220

230 240 250 230 240 250 Time (Sec) Time (Sec)

260 260

270 270

280 280

16 16

  

Altitude (m) Altitude (m)

15 15 200 200 160 160 155 155 150 150

 e (deg)  e (deg)

145 145 200 200 10 10 5 0

5 0

 T (pwm)  T (pwm)

-5 -5 200 200 1320 1320 1300 1300 1280 1280 200 200

  

q (rad/s) q (rad/s)

Figure 19. Fuzzy logic control (FLC) simulation results of climbing without the presence of wind. Figure 19. Fuzzy logic control (FLC) simulation results of climbing without the presence of wind.  Figure 19. Fuzzy logic control (FLC) simulation results of climbing without the presence of wind.  0.2 0.2 0

0

Energy rate Energy rate

-0.2 -0.2 280 280 0.1 0.1 0

300 300

310 310

320 320

330 330

340 340

350 350

360 360

E Edotdot L Ldotdot

0

-0.1 -0.1 280 280 18 18

Airspeed (m/s) Airspeed (m/s)

290 290

290 290

300 300

310 310

320 320

330 330

340 340

350 350

360 360

290 290

300 300

310 310

320 320

330 330

340 340

350 350

360 360

17.5 17.5 17 17

16.5 16.5 280 280

Figure 20. Cont.

  

Appl. Sci. 2016, 6, 369 Appl. Sci. 2016, 6, 369 

22 of 24 22 of 24 

Altitude (m)

160 155 150 145 280

290

300

310

320

330

340

350

360

290

300

310

320

330

340

350

360

290

300

310

320 330 Time (Sec)

340

350

360

 e (deg)

-1 -1.5 -2 -2.5 280  T (pwm)

1320 1300 1280 280

 

Figure 20. FLC simulation results of descending without the presence of wind. Figure 20. FLC simulation results of descending without the presence of wind. 

5.6. Summary of OECS Performance  5.6. Summary of OECS Performance Table  4  presents  a  summary  OECS  performance  different  flight  maneuvers  and  flight  Table 4 presents a summary of of  OECS performance forfor  different flight maneuvers and flight conditions. The simulation results for aircraft level level  flight,flight,  descending, and climbing maneuvers show conditions.  The  simulation  results  for  aircraft  descending,  and  climbing  maneuvers  . show that the errors of airspeed and altitude are very small in windless condition. Besides that, the  that the errors of airspeed and altitude are very small in windless condition. Besides that, the ∆q and Ls   and    are successfully regulated to a value close to zero. The errors with the presence of wind  are∆ successfully regulated to a value close to zero. The errors with the presence of wind are acceptable forare acceptable for aircraft level flight, descending, and climbing maneuvers. Moreover, the settling  aircraft level flight, descending, and climbing maneuvers. Moreover, the settling time for climbing and descending maneuvers is the same for tracking a similar altitude difference. This shows that the time for climbing and descending maneuvers is the same for tracking a similar altitude difference.  linearity of the total energy model is satisfied. Furthermore, by comparing the simulation results of This shows that the linearity of the total energy model is satisfied.  Furthermore,  by  comparing  the  OECS and FLC, it shows that the proposed OECS has better performance and fuel efficiency than that simulation results of OECS and FLC, it shows that the proposed OECS has better performance and  offuel efficiency than that of FLC.  FLC. Table 4. A summary of OECS performance for different flight maneuvers and flight conditions. Table 4. A summary of OECS performance for different flight maneuvers and flight conditions.  Condition Condition 

Maneuver  Maneuver

Windless Windless

ClimbDescend Descend Level Level Level LevelClimb

AltitudeAltitude Tracking (m)  Tracking (m) 150 Settling Settling Time (s)  Time (s) ∆q (rad/s) 0.01 ∆ (rad/s)  . 0.01 Ls   Altitude Deviation (m) 0.20 Altitude Deviation (m) Airspeed Deviation (m/s) 0.10 Airspeed Deviation (m/s)

150  150→150→158 158 158158→150 →150 ‐  40 40  40 40  0.01  0.01 0.01  0.010.01  0.01  0.01 0.01  0.010.01  0.20  0.20 0.20  0.200.20  0.20 0.20 0.10  0.20  0.20 

150  150 - ‐  0.15 0.15  0.05 0.05  2.00 2.00  0.50

0.50 

Wind Wind  Climb  Climb Descend  Descend 120→150  150→120  120→150 150→120 65  65 65 65 0.15 0.15 0.15  0.15  0.10 0.10 0.10  0.10  2.00 2.00 2.00  2.00  1.00 1.00 1.00  1.00 

6. 6. Conclusions  Conclusions   The design and implementation of the proposed OECS, which is based on the total energy of an The design and implementation of the proposed OECS, which is based on the total energy of an  aircraft, has been presented in this study. OECS is an aircraft longitudinal flight control system aircraft,  has  been  presented  in  this  study.  OECS  is  an  aircraft  longitudinal  flight  control  system  which is based on pilot-like control strategy to control the airspeed and altitude of the aircraft. which is based on pilot‐like control strategy to control the airspeed and altitude of the aircraft. The  The controller design process begins with the postulation of the aircraft model, from aircraft equations controller design process begins with the postulation of the aircraft model, from aircraft equations of  of motion and energy equations. After that, the postulated model was identified using the system motion  and  energy  equations.  After  that,  the  postulated  model  was  identified  using  the  system  identification method. The identified models were used to design the developed OECS. Finally, identification  method.  The  identified  models  were  used  to  design  the  developed  OECS.  Finally,  OECS was verified in HIL system of Spoonbill UAV. Three flight maneuvers, level flight, descending, OECS  was  were verified  in  HIL  system the of simulations. Spoonbill  UAV.  Three the flight  maneuvers,  level effect flight,  and climbing, adopted to perform In addition, influence of the wind descending, and climbing, were adopted to perform the simulations. In addition, the influence of the  was considered to validate the reliability of the developed system. The largest magnitude of wind was effect  was  considered  to show validate  the implementation reliability  of  the  developed  largest  setwind  to 8 m/s. The simulation results that the of the proposedsystem.  OECS forThe  different magnitude  of  wind  was  set  to  8  m/s.  The  simulation  results  show  that  the  implementation  the  flight maneuvers and flight conditions was successfully validated in the HIL simulation system ofof  the proposed OECS for different flight maneuvers and flight conditions was successfully validated in  the HIL simulation system of the used UAV.  When comparing the performance and fuel efficiency 

Appl. Sci. 2016, 6, 369

23 of 24

used UAV. When comparing the performance and fuel efficiency of the proposed OECS with another flight control system based on Fuzzy logic control, the results show that the proposed OECS has better performance and fuel efficiency than that of FLC. Acknowledgments: This study was supported by the Ministry of Science and Technology of Taiwan under grant number MOST 103-2632-E-035-001-MY3 and MOST 104-2622-E-035-019-CC3. Furthermore, this study was directed by Fei-Bin Hsiao, who passed away in February 2014. The authors admire and respect him for his great contributions to aerospace societies in the world. Author Contributions: Ying-Chih Lai determined the research framework, proposed the main idea, and performed mathematical modeling. Wen Ong Ting performed the implementation of the developed algorithm and simulations. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3.

4. 5. 6. 7. 8. 9. 10.

11. 12. 13.

14.

15.

16. 17.

Chen, S.-W.; Chen, P.-C.; Yang, C.-D.; Jeng, Y.-F. Total energy control system for helicopter flight/propulsion integrated controller design. J. Guid. Control Dyn. 2007, 30, 1030–1039. [CrossRef] Denker, J.S. See How It Flies. Available online: http://www.av8n.com/how/ (accessed on 18 November 2016). Lambregts, A. Integrated system design for flight and propulsion control using total energy principles. In Proceedings of the American Institute of Aeronautics and Astronautics, Aircraft Design, Systems and Technology Meeting, Fort Worth, TX, USA, 17–19 October 1983. Lambregts, A.A. Vertical flight path and speed control autopilot design using total energy principles. AIAA Pap. 1983. [CrossRef] Akmeliawati, R.; Mareels, I.M.Y. Nonlinear energy-based control method for aircraft automatic landing systems. IEEE Trans. Control Syst. Technol. 2010, 18, 871–884. [CrossRef] Kaminer, I.; O’Shaughnessy, P. Integration of four-dimensional guidance with total energy control system. J. Guid. Control Dyn. 1991, 14, 564–573. [CrossRef] Wu, S.-F.; Guo, S.-F. Optimum flight trajectory guidance based on total energy control of aircraft. J. Guid. Control Dyn. 1994, 17, 291–296. [CrossRef] Warren, A. Application of total energy control for high-performance aircraft vertical transitions. J. Guid. Control Dyn. 1991, 14, 447–452. [CrossRef] Voth, C.; Ly, U.-L. Design of a total energy control autopilot using constrained parameter optimization. J. Guid. Control Dyn. 1991, 14, 927–935. [CrossRef] Bruce, K.; Kelly, J.R.; Person, J.R.L. NASA B737 flight test results of the total energy control system. In Proceedings of the Astrodynamics Conference, Williamsburg, VA, USA, 18–20 August 1986; American Institute of Aeronautics and Astronautics: Reston, VA, USA, 1986. Rzucidlo, P. TECS/THCS based flight control system for general aviation. In Proceedings of the AIAA Modeling and Simulation Technologies Conference, Chicage, IL, USA, 10–13 August 2009. Lamp, M.; Luckner, R. The total energy control concept for a motor glider. In Advances in Aerospace Guidance, Navigation and Control; Springer: Heidelberg, Germany, 2013; pp. 483–502. Kale, A.; Singhz, G.; Patel, V. 1-adaptive-output feedback based energy control. In Proceedings of the 2015 European Control Conference (ECC), Linz, Austria, 15–17 July 2015; IEEE: New York, NY, USA, 2015; pp. 2792–2797. Brigido-González, J.; Rodríguez-Cortés, H. Experimental validation of an adaptive total energy control system strategy for the longitudinal dynamics of a fixed-wing aircraft. J. Aerosp. Eng. 2015, 29, 04015024. [CrossRef] Weiss, S.; Siegwart, R. Real-time metric state estimation for modular vision-inertial systems. In Proceedings of the 2011 IEEE International Conference on Robotics and Automation (ICRA), Shanghai, China, 9–13 May 2011; IEEE: New York, NY, USA; pp. 4531–4537. Deittert, M.; Richards, A.; Toomer, C.; Pipe, A. Dynamic soaring flight in turbulence. In Proceedings of the AIAA Guidance, Navigation and Control Conference, Chicago, IL, USA, 10–13 August 2009; pp. 2–5. Akhtar, N.; Whidborne, J.F.; Cooke, A.K. Real-Time Trajectory Generation Technique for Dynamic Soaring UAVs; Department of Aerospace Sciences, Cranfield University: Oxfordshire, UK, 2008.

Appl. Sci. 2016, 6, 369

18. 19. 20. 21. 22. 23. 24. 25.

24 of 24

Lee, C.; Chan, W.; Jan, S.; Hsiao, F. A linear-quadratic-gaussian approach for automatic flight control of fixed-wing unmanned air vehicles. Aeronaut. J. 2011, 115, 29–41. [CrossRef] Lee, C.-S.; Hsiao, F.-B.; Jan, S.-S. Design and implementation of linear-quadratic-gaussian stability augmentation autopilot for unmanned air vehicle. Aeronaut. J. 2009, 113, 275–290. [CrossRef] Rugh, W.J.; Shamma, J.S. Research on gain scheduling. Automatica 2000, 36, 1401–1425. [CrossRef] Hsiao, F.-B.; Hsieh, S.-Y.; Chan, W.-L.; Lai, Y.-C. Engine speed and velocity controller development for small unmanned aerial vehicle. J. Aircr. 2008, 45, 725–728. [CrossRef] Jategaonkar, R. Flight Vehicle System Identification: A Time Domain Methodology; AIAA: Reston, VA, USA, 2006; Volume 216. Klein, V.; Morelli, E.A. Aircraft System Identification: Theory and Practice; American Institute of Aeronautics and Astronautics: Reston, VA, USA, 2006. Ljung, L. Prediction error estimation methods. Circuits Syst. Signal Process. 2002, 21, 11–21. [CrossRef] Lai, Y.C.; Hsiao, F.B. Application of fuzzy logic controller and pseudo-attitude to the autonomous flight of an unmanned aerial vehicle. J. Chin. Inst. Eng. 2010, 33, 387–396. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).