Development of a Customized Autopilot for Unmanned Helicopter Model Using Genetic Algorithm via the Application of Different Guidance Strategies

Paper: ASAT-14-008-US
14th
International Conference on
AEROSPACE SCIENCES & AVIATION TECHNOLOGY,
ASAT - 14 – May 24 - 26, 2011, Email: asat@mtc.edu.eg
Military Technical College, Kobry Elkobbah, Cairo, Egypt
Tel: +(202) 24025292 –24036138, Fax: +(202) 22621908
1
Development of a Customized Autopilot for Unmanned Helicopter
Model Using Genetic Algorithm via the Application of Different
Guidance Strategies
A. Hosny*
, M. Hosny†
, and H. Ez-Eldeen‡
Abstract: The objective of this paper is to develop a customized autopilot system that enables
a helicopter model to carry out an autonomous flight using on-board microcontroller. The
main goal of this project is to provide a comprehensive controller design methodology,
Modeling, simulation, guidance and verification for an unmanned helicopter model. The
autopilot system was designed to demonstrate autonomous maneuvers such as flying over the
planned waypoints with constant forward speed and considerable steep maneuvers. For the
controller design, the nonlinear dynamic model of the Remote Control helicopter was built by
employing Lumped Parameter approach comprising of four different subsystems such as
actuator dynamics, rotary wing dynamics, force and moment generation process and rigid
body dynamics. The nonlinear helicopter mathematical model was then linearized using small
perturbation theory for stability analysis and linear feedback control system design. The linear
state feedback for the stabilization and control of the helicopter was derived using Pole
Placement Method. The overall dynamic system control with output feedback was computed
using Genetic Algorithm. Series of Matlab-Simulink models and guidance algorithms were
presented in this work to simulate and verify the autopilot system performance. The proposed
autopilot has shown acceptable capability of stabilizing and controlling the helicopter during
tracking the desired waypoints. This paper is presenting a detailed comparison study for two
different guidance strategies. The first strategy is concerning the difference between the
desired heading or elevation referred to the next waypoint and the actual heading or elevation
of the unmanned helicopter model. The second strategy is concerning the relative distance
between the actual and the desired trajectories. In other words the first method is tracking the
waypoints while the second one is tracking the trajectory. In this work a comparison study
was conducted through the mentioned strategies simulation to show the significant differences
in the output performance. Some performance indexes were presented to evaluate the system
performance errors and the control effort needed for both strategies using the same desired
trajectory and the same waypoints.
Keywords: UAH unmanned aerial helicopter, RC Remote Control, PPM Pole Placement
Method, GA Genetic Algorithm, DTCS Desired Trajectory Current Segment, WPTM
Waypoint Tracking Method. TTM Trajectory Tracking Method, WPTM Waypoint Tracking
Method, Jerk, FFR Fixed Frame of Reference, BFR Body Frame of Reference.
*
Egyptian Armed Forces, Egypt, ahmed160771@yahoo.com
†
Egyptian Armed Forces, Egypt, mbmomtaz@hotmail.com
‡
Egyptian Armed Forces, Egypt, hesham.ez@gmail.com

2
1. Helicopter Model
Helicopter dynamics obey the Newton-Euler equations for rigid body in translational and
rotational motions. The helicopter dynamics can be studied by employing lumped parameter
approach which presents that the helicopter model shown in Fig. 1 as a composition of following
components; main rotor, tail rotor, fuselage, horizontal bar and vertical bar. Figure 2 illustrates
typical arrangement of component forces and moments generated in helicopter simulation model,
[1, 2].
Fig. 1 Raptor 90 (15 cc Engine)
Forces equations:
𝒖𝒖̇ = −(𝒘𝒘𝒘𝒘 − 𝒗𝒗𝒗𝒗) +
𝑿𝑿
𝒎𝒎
− 𝒈𝒈𝒈𝒈𝒈𝒈 𝒈𝒈 𝒈𝒈
(1)
𝒗𝒗̇ = −(𝒖𝒖𝒖𝒖 − 𝒘𝒘𝒘𝒘) +
𝒀𝒀
𝒎𝒎
+ 𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈 𝒈𝒈 𝒈𝒈
(2)
𝒘𝒘̇ = −(𝒗𝒗𝒗𝒗 − 𝒖𝒖𝒖𝒖) +
𝒁𝒁
𝒎𝒎
+ 𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈
(3)
Moment equations:
𝑰𝑰𝒙𝒙𝒙𝒙 𝒑𝒑̇ = �𝑰𝑰𝒚𝒚𝒚𝒚 − 𝑰𝑰𝒛𝒛𝒛𝒛�𝒒𝒒𝒒𝒒 + 𝑰𝑰𝒙𝒙𝒙𝒙(𝒓𝒓̇ + 𝒑𝒑𝒑𝒑) + 𝑳𝑳 (4)
𝑰𝑰𝒚𝒚𝒚𝒚 𝒒𝒒̇ = (𝑰𝑰𝒛𝒛𝒛𝒛 − 𝑰𝑰𝒙𝒙𝒙𝒙)𝒓𝒓𝒓𝒓 + 𝑰𝑰𝒙𝒙𝒙𝒙(𝒓𝒓𝟐𝟐
− 𝒑𝒑𝟐𝟐) + 𝑴𝑴 (5)
𝑰𝑰𝒛𝒛𝒛𝒛 𝒓𝒓̇ = �𝑰𝑰𝒙𝒙𝒙𝒙 − 𝑰𝑰𝒚𝒚𝒚𝒚�𝒑𝒑𝒑𝒑 + 𝑰𝑰𝒙𝒙𝒙𝒙(𝒑𝒑̇ − 𝒒𝒒𝒒𝒒) + 𝑵𝑵 (6)
Kinematic equations:
𝝋𝝋̇ = 𝒑𝒑 + 𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒 + 𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓 (7)
𝜽𝜽̇ = 𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒 − 𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓𝒓 (8)
𝜳𝜳̇ = 𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒𝒒 + 𝒓𝒓𝒓𝒓𝒓𝒓𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔 (9)
The physical helicopter parameters used for the model (Raptor 90) are given in Table 1.

3
Table 1 Parameters of Raptor 90 helicopter for simulation model
Parameter Description
ρ = 1.225 kg/m3
Atmosphere density
m = 7.70 kg Helicopter mass
Ixx = 0.192 kg m2
Rolling moment of inertia
Iyy = 0.34 kg m2
Pitching moment of inertia
Izz = 0.280 kg m2
Yawing moment of inertia
Ωnom = 162 rad/s Nominal main rotor speed
RM = 0.775 m Main rotor radius
RCR = 0.370 m Stabilizer bar radius
CM = 0.058 m Main rotor chord
CCR = 0.06 m Stabilizer bar chord
aM = 5.5 rad-1
Main rotor blade lift curve slope
𝐶𝐶𝐷𝐷𝐷𝐷
𝑀𝑀
=0.024 Main rotor blade zero lift drag coefficient
𝑪𝑪𝑻𝑻𝑻𝑻𝑻𝑻𝑻𝑻
𝑴𝑴
=0.00168 Main rotor max thrust coefficient
Iβ = 0.038 kg m2
Main rotor blade flapping inertia
RT = 0.13 m Tail rotor radius
CT = 0.029 m Tail rotor chord
aT = 5.0 rad-1
Tail rotor blade lift curve slope
𝑪𝑪𝑫𝑫𝑫𝑫
𝑻𝑻
=0.024 Tail rotor blade zero lift drag coefficient
𝑪𝑪𝑻𝑻𝑻𝑻𝑻𝑻𝑻𝑻
𝑻𝑻
=0.0922 Tail rotor max thrust coefficient
nT = 4.66 Gear ratio of tail rotor to main rotor
nes = 9.0 Gear ratio of engine shaft to main rotor
𝜹𝜹𝑻𝑻
𝒕𝒕𝒕𝒕𝒕𝒕𝒕𝒕
= 0.1 rad Tail rotor pitch trim offset
SV= 0.012 m2
Effective vertical fin area
SH =0.01 m2
Effective horizontal fin area
𝐶𝐶𝐿𝐿∝
𝑉𝑉
= 2.0 rad-1
Vertical fin lift curve slope
𝐶𝐶𝐿𝐿∝
𝐻𝐻
= 3.0 rad-1
Horizontal tail lift curve slope
𝑆𝑆𝑥𝑥
𝐹𝐹
= 0.1 m2
Frontal fuselage drag area
𝑺𝑺𝒚𝒚
𝑭𝑭
= 0.22 m2
Side fuselage drag area
𝑆𝑆𝑧𝑧
𝐹𝐹
= 0.15 m2
Vertical fuselage drag area
hM = 0.235 m Main rotor hub height above CG
lM = 0.015 m Main rotor hub behind CG
lT = 0.91 m Tail rotor hub location behind CG
hT = 0.08 m Tail rotor height above CG
lH= 0.71 m Stabilizer location behind CG
kMR =0.3333 Amount of commanded swash plate tilt
kCR =1.1429
Geometry coefficient of the mechanical
linkage of control rotor and swash plate

4
Fig. 2 Typical Arrangement of Component Forces and Moments Generation in
6-DOF Helicopter Simulation Model
The linearized matrices for hover condition excluding the control rotor (Six DOF model eight
states) are given in Table 2 and Table 3. Each column and row is marked with the states and
inputs that are being referred to the state space model. Table 4 states the stable and unstable
eigenvalues according to the different modes for six DOF in hovering and low speed flight
conditions. Figure 3 illustrates the different poles on the corresponding Pole-Zero Map.
𝑥𝑥̇ = 𝐴𝐴𝐴𝐴 + 𝐵𝐵𝐵𝐵 (10)
𝑦𝑦 = 𝐶𝐶𝐶𝐶 (11)
Table 2 Analytically obtained A matrix in hover with no control rotor
u w Q θ v p φ r
u -0.0070825 0 0.00093895 -9.81 -0.0009389 -0.0292 0 0
w 0 -0.8159 0 0 0 0 0 -0.12
q 0.0377 -0.2775 -0.6718 0 -0.28599 0.1558 0 0.265
θ 0 0 1 0 0 0 0 0
v 0.00093895 0 0.0144 0 -0.06808 0.122823 9.81 0.055
p 0.0094513 0 -0.2942 0 -0.1377 -1.286 0 0.19
φ 0 0 0 0 0 1 0 0
r 0 -1.5246 0 0 1.528 0.122 0 -5.1868
Table 3 Analytically obtained B matrix in hover with no control rotor
δCol δLon δLat δPed
u 5.2981 1.5591 -0.1816 0
w -128.777 0 0 0
q -72.0367 -8.3082 0.9678 9.07
θ 0 0 0 0
v -31.9088 0.0605 -0.5196 5.055
p -321.1883 1.8281 -15.6933 17.322
φ 0 0 0 0
r 178.2831 0 0 17.322

5
Table 4 Eigenvalues and modes for 6-DOF in Hovering/Low Speed Flight Condition
Mode Eigenvalues Damping Frequency (rad/s)
Longitudinal Oscillation 0.169 ± 0.392i -0.395 0.427
Lateral Oscillation -0.0279 ± 0.765i 0.0365 0.765
Heave -0.68 ± 0.142i 0.979 0.695
Roll Subsidence -1.68 1 1.68
Yaw -5.28 1 5.28
Fig. 3 Poles of Coupled Longitudinal and Lateral Motion for
6-DOF with no Control Rotor
2. Application of Pole Placement Method to 6-DOF
(8 States Linear Model)
The objective of applying Pole Placement Method (PPM) to 6-DOF model is to shift the
unstable Poles (the above marked ones) to the Left region of the root Locus (stable region).
Consequently an initial stable reference model is established and ready to be tuned by
applying GA functions to optimize the net performance index. The default poles will be
placed in new position as the following (-0.8±1.095i, -0.85±1.096i, -57, -10, -15, -10)
according to the stated flying qualities in Fig. 4, [5]:

6
Fig. 4 Limits on pitch (roll) oscillations – hover and low speed according to
Aeronautical Design Standard for military helicopter (ADS-33C)
(US Army Aviation Systems Command, 1989)
2. Application of GA Functions to optimize the PID Controller Parameters
The main objective of the proposed GA procedure is to optimize (minimize) the output
performance index (maximize fitness function) of the whole integrated control system used in
Fig. 5. An application of GA code combined with SIMULINK control model for the proposed
Helicopter model is implemented to minimize the performance index (J), [3].
The equation describing the net output performance (J) is stated below:
∫
∞
⋅⋅+⋅⋅+⋅⋅+⋅⋅=
0
4
2
3
2
2
2
1
2
])_()_()_()_([ dtsimresimesimwesimueJ WtWtWtWt ϕ (12)
where
e(u_sim) is the forward speed error
e(w_sim) is the normal speed error
e(𝜑𝜑_sim) is the bank angle error
e(r_sim) is the yaw rate error
W1,W2, W3 and W4 are the performance weights that depend on the required performance
priority.
The previous index can be obtained from the nonlinear simulation of the pitch control
model consequently GA can use it as the fitness function.
2.1 Optimizing PID Parameters
2.1.1 Using binary system coding
Coding: using 10-bit binary genes to express DIP KKK ,, . For example (KP) from bit 1
(0)0000000000 , to bit 10 is (1023)1111111111 .Then string DIP KKK ,, to 30-bit binary
cluster. 000010001011101110000000110111:x expresses a chromosome，the former 10-bit
express KP, the second portion express KI and the third one express the KD.
Decoding: Cut one string of 30-bit binary string to three 10-bit binary string，then converts
them to decimal system values y1, y2 and y3, [3, 4].

7
Fig. 5 Proposed Control System for Computing PID Optimum Gains
2.1.2 Evaluation of fitness function
Fitness function is the main criterion of the GA algorithm, as it represents how much the
system is optimum and stable. The following equation describes the relation between the
fitness function and the performance index.
1
))_((),,( −
= simeJKKKf DIP (13)
2.1.3 Design operators
Proportion selection operator，single point crossover operator，basic bit mutation operator.
2.1.4 Parameters of GA
Population size is 04=M , generation 100=G , crossover probability 0.60c =P ，mutation
probability 0.10m =P . Adopting the above steps, after 100 steps iteration, Fig. 6, the best
performance index will be reached giving the optimum gains as shown in Table 5.

8
Fig. 6 Performance Index J after 100 iterations
Table 5 PID Optimum Gains for the Integrated Control System
k1 k2 k3 k4 kp0 kd0 ki0 kp1
-0.7634 -0.1711 -0.9081 -0.9902 0.0190 0 -1.8034 4.6676
kd1 ki1 kp2 kd2 ki2 kp3 kd3 Ki3
0 4.6774 4.9022 0 2.9062 2.9120 0 3.0459
3. Integrated Control System
Building up the PID controller using the previous optimum gains stated in Table 5 (inner loop
controller). Then a proposed guidance algorithm is established and fed with the flying model
position from the GPS and the waypoints data (number of waypoints, locations and
sequence). Applying vector analysis with mathematical aids, the relative heading and
elevation with respect to the next waypoint will be calculated. Consequently the required
change in heading and elevation will be fed to the PID controller (outer loop) after applying
manual tuning to the out loop gains. Figure 7 demonstrates the current and the relative
heading with respect to the next waypoint in the sequence. The guidance algorithm maintains
the actual flight path not to fly in continuous loops around a definite waypoint by using some
switching limiters such as acceptable circular error around the proposed waypoints and
switching orthogonal liners. Figure 8 shows the integrated control system including
(Helicopter Model, Guidance Algorithm and PID Controller) with the input commands and
the output performance. The guidance algorithm was implemented using Matlab code, and all
the simulation results were conducted using Simulink.
4. Guidance Approach
4.1 TTM Trajectory Tracking Method
This approach is concerning the relative distances such as the lateral and the vertical distances
with respect to the desired trajectory current segment DTCS body axis Fig. 9. In this method
(in case of lateral control) inner and outer loops will be required, the inner loop will control
the yaw rate and the outer loop will be fed by the lateral relative distance with respect to the
desired trajectory current segment DTCS body axis while the feedback signal will represent
the change in the heading as the absolute heading is not useful in this case. Both the inner and
the out loop gains will be tuned by several trials after system integration. This approach leads
to minimize the relative distances between the actual flight path and the desired trajectory.

9
Fig. 7 Relative Heading using Waypoint Tracking Strategy
Fig. 8 Integrated Control System Including (Helicopter Model,
Guidance Algorithm, PID Controller)

10
4.2 WPTM Waypoint Tracking Method
This approach is concerning the relative heading and elevation with respect to waypoints,
Fig. 10. Applying some vector analysis the required change in heading or elevation will be
easily calculated, consequently they will be fed to the yaw rate inner loop after tuning the
command input gain according to the performance criteria, [1].
Fig. 9 Lateral Distances using
Trajectory Tracking Strategy
Fig. 10 Relative Heading using
Waypoint Tracking Strategy
The following equations explain the transformation procedure starting from the actual flying
model position with respect to fixed frame of reference to body coordinate of the DTCS with
estimating a small initial conditions of (0.0001 for example) to avoid the singularities during
simulation calculation, Fig. 11, [2,3].
X_T =Actual Flying Model Longitude (X component) relative to the fixed frame of reference
Y_T =Actual Flying Model Latitude (Y component) relative to the fixed frame of reference
Z_T =Actual Flying Model Altitude (Z component) relative to the fixed frame of reference
X_F =The Start Point of the Desired Trajectory Current Segment DTCS (X component)
relative to the fixed frame of reference
Y_F = The Start Point of the Desired Trajectory Current Segment DTCS (Y component)
Z_F = The Start Point of the Desired Trajectory Current Segment DTCS (Z component)
XBO = Flying Model Actual Longitude (X component) relative to DTCS body axes at the
fixed frame of reference
YBO = Flying Model Actual Latitude (Y component) relative to DTCS body axes at the fixed
frame of reference
ZBO = Flying Model Actual Altitude (Z component) relative to DTCS body axes at the fixed
frame of reference
XB = Flying Model Actual Longitude (X component) relative to DTCS body axes
YB = Flying Model Actual Latitude (Y component relative) to DTCS t body axes
ZB = Flying Model Actual Altitude (Z component) relative to DTCS body axes
θ , Ф, ψ = Rotation around X,Y, Z axes
A = cos θ cos ψ
B = (sin Ф sin θ cos ψ – cos Ф sin ψ)
C = (cos Ф sin θ cos ψ+ sin Ф sin ψ)

11
D = cos θ sin ψ
E = (sin Ф sin θ sin ψ + cos Ф cos ψ)
F = (cos Ф sin θ sin ψ – sin Ф cos ψ)
G = – sin θ
M = sin Ф cos θ
N = cos Ф cos θ
)])(())([(
)])(())([(
BDEANACGMABGCDFA
BDEAZAXGMABGXDYA
ZBO
−−−−−
−−−−−
= (14)
)(
)]()[(
BDEA
CDFAZXDYA
Y BO
BO
−
−−−
= (15)
A
CZBYX
X BOBO
BO
)( −−
= (16)
)])(())([(
)])(())([(
BDEANACGMABGCDFA
BDEAAZGXMABGDXAY
Z FFFF
SHIFT
−−−−−
−−−−−
= (17)
)(
)]()[(
BDEA
CDFAZDXAY
Y SHIFTFF
SHIFT
−
−−−
= (18)
A
CZBYX
X SHIFTSHIFTF
SHIFT
)( −−
= (19)
XB = XBO – X shift (20)
YB = YBO – Y shift (21)
ZB = ZBO – Z shift (22)
Fig. 11 Calculating the Flying Model Relative Distances
with Respect to DTCS Body Axis

12
5. System Performance
In the following figures (Fig. 12 through Fig. 31), it is clear that the Trajectory Tracking
Method TTM has a great and obvious impact in minimizing the relative distance between the
actual flight path and the desired trajectory, the figures show that the flying model almost
flying stick to the desired path when TTM is used while it tracks the waypoints only when
WPTM is used regardless the relative distances between the actual and the desired path, thus
increasing the relative distances (relative errors) when WPTM is used compared to the
relative distances if TTM is used. Figure 26 shows also a considerable fluctuation with large
amplitude when TTM is used rather than when WPTM is used expressing a large control
energy consumed during TTM due to the high number of attempts performed by the flying
model to track each segment in the desired trajectory segments. Figures 30 and 31 show that
TTM would have longer endurance (5% more) than the WPTM for the same range (or
mission), consequently TTM would have less range than WPTM for a given amount of fuel.
Fig. 12 Actual and Desired Trajectories in
XY Plane with the Relative Lateral
Distances (in DTCS body axis)
for TTM
XY Plane with the Relative Lateral
for WPTM
Fig. 14 Actual and Desired Trajectories
with Actual Heading Angle For TTM
Fig. 15 Actual and Desired Trajectories
with Actual Heading Angle For WPTM

13
XZ Plane with the Relative Vertical
for TTM
XZ Plane with the Relative Vertical
for WPTM
XZ Plane with Actual Elevation
for TTM
XZ Plane with Actual Elevation
for WPTM
Fig. 20 Actual and Desired Forward
Speed for TTM
Fig. 21 Actual and Desired Forward
Speed for WPTM

14
Fig. 22 Angle of Attack for TTM Fig. 23 Angle of Attack for WPTM
Fig. 24 Side Slip Angle for TTM Fig. 25 Side Slip Angle for WPTM
Fig. 26 Yaw Rate for TTM Fig. 27 Yaw Rate for WPTM
Fig. 28 Pitch Rate for TTM Fig. 29 Pitch Rate for WPTM

15
Fig. 30 Range and Endurance for TTM Fig. 31 Range and Endurance for WPTM
6. Control System Performance
Figures 32 through 39 illustrate the control efforts done by each servo to perform the
maneuvers required for the desired flight path for both TTM and WPTM. Figure 38 indicates
an obvious fluctuation with large amplitude that would need a considerable amount of electric
energy. This number of fluctuations would reduce the servos life time, and decrease the mean
time between failures.
Fig. 32 Lateral Control Effort for TTM Fig. 33 Lateral Control Effort for WPTM
Fig. 34 Pitch Control Effort for TTM Fig. 35 Pitch Control Effort for WPTM
Fig. 36 Collective Control Effort for TTM Fig. 37 Collective Control Effort for
WPTM

16
Fig. 38 Rudder Control Effort for TTM Fig. 39 Rudder Control Effort for WPTM
7. System Evaluation
The absolute relative distance integral in both (X, Y) directions (in DTCS body axis) was
chosen to be the evaluation criteria (Performance Index PI) in this work. Figures 40 and 41
show that the Trajectory Tracking Method TTM has less (Performance Index PI) when it is
compared with Waypoint Tracking Method WPTM though the (Performance Index PI) using
WPTM is slightly less than when using TTM in Z direction (with respect to DTCS body axis).
As the global performance index for both directions Y, Z (with respect to DTCS body axis)
using TTM is much less when compared with the WPTM. On the other hand when applying
the same performance index to both methods (TTM, WPTM) for yaw rate, yaw acceleration
and yaw jerk, it is clear that the TTM energy consumption during the tracking maneuvers is
much more than the energy consumed by the WPTM as it is shown in Figs. 42, 43 and 44,
[4,5].
Fig. 40 Performance Index in (Z)
Directions for TTM and WPTM
Fig. 41 Performance Index in (Y)
Directions for TTM and WPTM

17
Fig. 42 Absolute Yaw Rate Integral for both TTM and WPTM
Fig. 43 Absolute Yaw Angular Acceleration Integral for both TTM and WPTM
Fig. 44 Absolute Yaw Angular Jerk Integral for both TTM and WPTM

18
8. Conclusion
This paper has presented a customized autopilot using PID controller combined with a
guidance algorithm. Genetic algorithm was used in this work to tune the proposed stable state
space model after applying pole placement method PPM. The proposed stable poles satisfy
the flying qualities criteria. The illustrated results show an adequate system performance with
a reasonable relative distances with respect to the desired trajectory. The forward speed was
maintained constant during the whole flight according the desired speed. The servos control
efforts are satisfying the band width limits, and also the minimum and maximum limits. It is
concluded from the system performance that it is recommended to use TTM in case when the
unmanned flying model is required to track a planned trajectory with less relative errors,
while the WPTM is recommended when the precision in tracking the planned trajectory is not
an objective. However the control energy consumed by TTM is obviously more than WPTM
for the same planned trajectory (Flight Path), thus the control servos using TTM would
require more batteries than they would require when using WPTM. The application of hybrid
system that utilizes both methods advantages is strongly recommended in this case. Using
both methods will allow applying TTM during loitering when the tracking precision is
required, while applying WPTM when it is only required to pass the waypoints without
tracking the flight path passing those waypoints (when covering distances only is required
during flight).
9. References
[1] Shamsudin S. S., “The Development of Autopilot System for an Unmanned Aerial
Vehicle (UAV) Helicopter Model”, M. Sc. thesis, Universiti Teknologi, Malaysia,
Faculty of Mechanical Engineering. 2007, pp. 1-147.
[2] Hosny A. M., Chao H., “Development of Fuzzy Logic LQR Control Integration for
Aerial Refueling Autopilot”, " Proceedings of the 12th
International Conference on
Aerospace Sciences and Aviation Technology, ASAT-12," , MTC, Cairo, Egypt, May
29-31, 2007.
[3] Hosny A. M., Chao H., “Fuzzy Logic Controller Tuning Via Adaptive Genetic
Algorithm Applied to Aircraft Longitudinal Motion”, " Proceedings of the 12th
International Conference on Aerospace Sciences and Aviation Technology, ASAT-12,"
MTC, Cairo, Egypt, May 29-31, 2007.
[4] Goldberg, D., Genetic Algorithms in Search, Optimization & Machine Learning,1st
. ed.,
Vol. 1, Addison Wesley Longman, 1989.
[5] Flying Qualities of Aeronautical Design Standard for Military Helicopter (ADS-33C)
(US Army Aviation Systems Command, 1989).

Development of a Customized Autopilot for Unmanned Helicopter Model Using Genetic Algorithm via the Application of Different Guidance Strategies

More Related Content

What's hot (19)

Viewers also liked (7)

Similar to Development of a Customized Autopilot for Unmanned Helicopter Model Using Genetic Algorithm via the Application of Different Guidance Strategies (20)

More from Ahmed Momtaz Hosny, PhD (16)

Recently uploaded (20)

Development of a Customized Autopilot for Unmanned Helicopter Model Using Genetic Algorithm via the Application of Different Guidance Strategies