KINEMATICS OF FIXED
WING UAV
Kinematics and Dynamics of Fixed-Wing UAVs
This chapter provides a review of basic knowledge required for
accurate mathematical modeling of flight of a fixed-wing UAV. These
include the kinematics and dynamics of motion, and the transformation
of forces andReference
moments acting on the
Frames andairplane.
Coordinate
Transformations
In order to accurately describe a body motion, it is required to define
(i) the forces and moments acting on the body and thus resulting in the
body motion and
(ii) the coordinate system that can be used as a reference for the
There are two types of forces acting on a body in free motion.
motion states definition.
First, the inertial forces and moments that depend on the
velocities and accelerations relative to an inertial reference frame;
the classical Newtonian dynamics equations hold only in the inertial
frame.
The second group consists of the aerodynamic forces and moments
resulting from interaction of the body with the surrounding
airflow and therefore relative to the air. Since the airflow might not be
The resulting motion can be conveniently described in terms of
the position, velocity, acceleration, and attitude
coordinates which comprise the states of the moving
body. Some of these states, in turn, need to be defined with
respect to a reference frame whose choice is defined by the
specifics of the UAV application. Thus, the information carried
by various reference frames is what facilitates the complete
and convenient definition of the free-body motion.
Therefore, this section starts with a definition of a
coordinate frame and a description of the coordinate frame
rotation.
The reference frames required to represent the aerodynamic
forces and moments and facilitating the solution of the
navigation states are introduced next.
Communication of the states information occurring during the
Kinematics of Moving Frames
Consider a vector p defined in two
orthogonal coordinate frames rotated with
respect to their mutual origin by angle α,
as shown
From this in Figure.
geometrical setup, it can be
demonstrated that vector p = can be
uniquely defined in both frames as
follows:
where 0 and 1 subscripts refer to the coordinates of p in the original and rotated
frames correspondingly.
Introducing the matrix notation for the linear transformation above results in a simple
form that relates the vector p components in frame to the corresponding components
in frame:
Where, the resulting rotation matrix is called a Direction Cosine
Matrix (DCM)
The DCM matrix consists of the cosine and sine functions
which are the direction cosines between the matching axes of
the new and old coordinate systems denoted in the
Following the same approach, it can be
demonstrated that for the case of right-
handed coordinate system represented by
three orthogonal axes, the same right-hand
rotation results in transformation
where for clarity, the subscript z denotes the axes of rotation. Proceeding similarly, right-
handed rotations of the coordinate frame about the y0 and x0 axis give
It is worth noting that the DCM transformation has the following easy-to-
remember properties that simplify its application (see more details in Rogers
2003):
1. The transformed vector components along the axis of rotation remain
unchanged with the rotation about that axis; elements of DCM are either 0 or
1.
2. The remaining elements of DCM are either sin or cos functions of the angle
of rotation.
3. The cos elements are on the main diagonal with sin elements on off-
diagonal.
4. The negative sin component corresponds to the component rotated
“outside” of the quadrant formed by the original frames.
5. Columns (rows) of a DCM matrix form an orthonormal set.
It is straightforward to verify that a DCM matrix corresponding to the right-
handed frames has the following properties:
and therefore it belongs to a general class of orthonormal transformation
matrices. For a sequence of rotations performed with respect to each
orthogonal axis, the resulting transformation can be obtained by a matrix
composed of three sequential rotations, called Euler angle rotations, starting
from the original frame of reference.
Formally, this transformation is
accomplished by rotating
through the ordered sequence
of Euler angles where the
numerical indexes define the
ordered sequence of rotations
and the
corresponding axis of rotations:
This matrix, which represents a transformation resulting from three
sequential Euler angle rotations, will be used throughout the chapter.
Overall, any rotation matrix has a number of properties. They are
summarized here for completeness; an interested reader is referred to
references for thorough details:
• Rotation matrices are orthogonal.
• The determinant of a rotation matrix is unity.
• Successive rotations can be represented by the ordered product of the
individual rotation matrices.
• Rotation matrices are not commutative (AB ≠ BA);
A nontrivial rotation matrix has only one eigenvalue equal to unity with
other two, being a complex conjugate pair with unity magnitude; a trivial
rotation is described by an identity matrix.
The time rate of change of the DCM matrix that defines the dynamics of
the attitude states is important in the derivation of the kinematic
equations of motion.
It enables relating the sensor (e.g., given by the rate gyros) measurements
obtained in a body-fixed frame to the time derivatives of the Euler angles
describing the attitude of a body in an inertial frame.
In general case, the time derivative of a rotation matrix that is considered
as a function of time can be obtained based on its key properties.
Let be a rotation matrix given as a function of time. Since
Please Note: The
transpose of the
transpose of A is A.
is a skew symmetric ˙
𝑅=𝑅 ×𝑆
matrix.
Let where p0 is a vector of constant length rotated over time
with an angular velocity vector Ω.
Comparing two expressions of the absolute time
derivatives of p1
Thus, the skew symmetric matrix S (Ω) is used to represent the vector cross
product between the vectors Ω and the R(t) p0.
The matrix, S (Ω) where is represented by its components,
vector can be written in the form
= Please Note: Some text
books use for
representing angular
velocities.
Another useful general property of angular velocities is called the angular
velocities addition theorem. The theorem states that for angular velocity
vectors coordinated in a common frame, the resulting angular velocity of
the cumulative rotation is a plain sum of the contributing rotations.
Euler angles defined with respect to a stationary frame, then it is
straightforward to determine the components of the angular velocity
vector as though measured in the rotating frame.
Starting from an initial stationary frame and using two intermediate
frames whose relative angular velocities are defined by the Euler angle
rates and utilizing the angular velocities addition theorem, the
following kinematic equation can be obtained:
that defines the derivatives
of the Euler angles in terms
of the angles themselves
and the angular velocity
vector as it was
measured in the rotated
frame. These equations
define the rotational
kinematics of a rigid body;
they contribute to the final
set of 6DoF equations of
motion.
Coordinate
Frames
Deriving equations of motion of a fixed-wing UAV requires a definition
of coordinate frames where forces and moments acting on the airplane
can be conveniently defined and where the states including the
position, velocity, acceleration, and attitude can be suitably
described.
The long duration and great operational distances might require
considering the UAV flight operations with respect to the rotating
Earth. This consideration would extend the set of coordinate frames
used in long-endurance UAV applications.
The primary reason for such an extension would lie in the necessity to
resolve the inertial angular velocity of rotation of a body frame in a
“true” inertial frame where the Earth rotation can be resolved.
Consequently, the angular velocities addition theorem mentioned
above would be used to resolve the angular velocity vector of body
rotation with respect to the inertial frame as a vector sum of angular
velocities of the intermediate frames.
Therefore, in addition to the body and wind frames that define dynamics
of the body–fluid interaction, this subsection defines the following set of
coordinate frames used in various UAV applications:
1. Earth-centered inertial frame {i}
2. Earth-centered Earth-fixed frame
{e}
3. Geodetic coordinate system {λ, φ, h}
4. Tangent plane coordinate system
{u}
5. Body-carried frame {n}
6. Body-fixed frame {b}
Earth-centered
7. Wind frame {w} Earth-fixed (ECEF)
coordinate system is fixed to the Earth,
and therefore, it rotates at the Earth’s
sidereal rate with respect to the Earth-
centered inertial (ECI) frame that
represents nonrotating inertial frame.
The local geodetic {λ, φ, h} frame is usually
associated with the ECEF frame,. It has the same
origin placed at the center of the Earth. The
frame defines the orientation of a line normal to
Earth’s surface and passing through the point of
interest. The orientation of the line is defined by
two angles, λ (geodetic latitude) and φ (geodetic
longitude), with the height h above the Earth’s
surface; eventually these three parameters, along
with the components of the UAV velocity vector,
become the major navigation states.
Local Tangent Plane Coordinate System
The origin of the local tangent plane (LTP) is fixed to the surface of the Earth with two of its axes
attached to the plane tangent to the surface; The frame is usually marked with the subscript {u}
and serves the purpose of an inertial frame in most short-duration low-speed UAV applications.
The frame’s xu and yu axes are in the tangent plane and most often aligned with the north and east
directions correspondingly; the zu axis completes the right-hand coordinate system, thus pointing
down. It can be also defined as an NED frame indicating the north–east down alignment of the
coordinate axes.
When the origin of the LTP frame is defined in terms of its geodetic latitude,
longitude, and altitude above the ground surface, then the following
equations can be applied to define the kinematics of navigation states.
If re = 6,378,137:00 m, rp - 6,356,752:314m. The resulting transformation from the geodetic
{λ, φ, h} To the ECEF frame is as follows
where ϵ the eccentricity of oblate
ellipsoid is defined as
Body-Carried and
Fixed Frames
In flight dynamics, the body-attached
reference frames usually have their origin at
the center of gravity (CG) of an airplane;
therefore, these frames are moving.
The body-carried frame {n} is an orthogonal
frame originated at the CG of the UAV. All its
axes are permanently stabilized and aligned
with the LTP frame axes as it was connected
to the CG; the frame is usually utilized in
defining the navigation equations thus
assigning its subscript {n}. This frame is
connected to the LTP frame by means of a
plain translation
The body-fixed frame is an orthogonal frame
r=
defined with respect to the body carried frame. Its
or T in some textbooks.
origin is at the CG of UAV, and its axes are rigidly
connected to the body; therefore, the frame also
rotates with the body. The frame is usually marked
with the subscript {b}. The alignment of the {b}
Equilibrium conditions:
u T sin α
(T)
vc
T cos α
Assuming that a typical UAV has (D)
at least one plane of symmetry
W sin γ
(geometric and mass symmetry)
results in two of the body-fixed W cos γ
axes laying in the plane of (W)
symmetry. When the axes are
aligned along the principal axes
of inertia of the body, as will be
shown in the following chapter, If the UAV has a vertical plane of symmetry,
the dynamic equations of motion then xb and zb lie in that plane of symmetry; xb
become significantly simpler. In a points toward the direction of flight and zb
majority of fixed-wing UAV points downward and yb points right, thus
As the body moves, its attitude is defined with reference to the body-
carried frame {n} by three consecutive Euler rotations
DCM matrix transforming the body-carried {n} to the body-fixed {b}
frame can be constructed. Now the subscripts (u b) denote the rotation
from the LTP {u} to the body fixed frame; {u} and {n} frames are always
aligned
The by the definition
application of body-carried
of the rotation frame.
matrix immediately follows from the need
to describe the UAV translational motion in an inertial frame of reference
by utilizing the inertial velocity measurements taken in the body-fixed
frame at CG, r= or T denotes the position of an airplane CG with respect to the LTP
Also vector
(NED) frame attached to the Earth.
Aircraft Kinematics |
Reference axes Body-axis
Body velocity:
• Air-mass
relative speed
(airspeed):
• V is always
positive
Body rates:
Robot Dynamics: Fixed Wing UAS 05.12.2017 21 | |
Autonomous Systems Lab
Earth fixed frame
(regarded as inertial): Body fixed
frame:
Aircraft Kinematics | Coordinate e xE , e yE , e zE
e ,e ,e xB yB zB
transformation
Rotation Matrix ( to ) is parametrized with 3 successive
rotations using
the ZYX Tait-Brian Angles (specific kind of Euler Angles):
1 Yaw:
around - 2 Pitch:
around - 3 Roll:
around -
Frame Frame Frame
1 2 B
| |
Translational Kinematics
Relating the translational velocity and position and accounting for the fact
that body-carried frame {n} is stabilized with respect to the nonrotating
{u} frame results in
Rotational Kinematics
The relation between the Euler angles defining the relation between the
stabilized {n) frame and the body-fixed frame {b} can be derived as follows
using angular velocity addition theorem. They define the dynamics of Euler
angles defined in an inertial frame with respect to the rates measured in the
body-fixed frame.
Thus, the translational and
rotational kinematic
equations represent the
dynamics of translational and
rotational coordinates and
therefore are part of the final
set of equations of motion.
Wind Frame
Aerodynamic forces and moments resulting from the body–
air interaction as the airframe moves through the air depend
on the body orientation with respect to the surrounding air.
In other words, they depend on the vector representing the
wind. The velocity vector of the possibly moving air (wind)
resolved in the inertial frame {u} is denoted
• The magnitude of is called an airspeed Va,
• The velocity vector defined in LTP with respect
to the ground – ground speed vector
The orientation of the wind frame {w} defined
by the direction of with respect to the body–
fixed {b} is defined by two angles.
The angle of attack α usually positive angle
with respect to and influences longitudinal
stability of an airplane.
The sideslip angle. It is denoted by β.
Then the transformation from the body-fixed frame {b} to the wind frame {w} is given by
The inverse transformation from the wind
frame {w} to the body-fixed frame {b} is
Therefore, it is necessary to understand the
difference between the airspeed, represented
by the air velocity vector and the ground
speed , both resolved with respect to the LTP
frame.
In the presence of constant wind, these velocities are related by the equation that is often called
the wind triangle:
where is the wind velocity defined in the LTP frame.
The objective of the following development is to define the relations among these
velocities defined in three different frames, while being measured or estimated by the
sensors installed in the body-fixed and in the inertial frame .
Let the UAV velocity in the LTP (inertial) frame expressed in the body frame be
Let the wind velocity in the LTP frame expressed in body frame be
Let be its components expressed in the body frame.
Utilizing the definition of the angles of attack and sideslip relating the wind
frame to the body-fixed frame and the “wind triangle” expressed in the
body frame results in the following
This last equation relates the airspeed components of resolved in the body
frame with the airspeed and the angles of attack and sideslip. In turn, if
the wind components resolved in the body frame are known, then inverting
the last equation allows for calculation of the airspeed V a and the α and β
angles as follows.
This section developed the fundamental kinematic equations that not only define the
kinematics of states and contribute to the final set of 6DoF equations of motion but also
serve as the basis for the design of the guidance and navigation tasks.