Академический Документы
Профессиональный Документы
Культура Документы
Email address:
sbektas1@gmail.com
Received: January 19, 2017; Accepted: February 4, 2017; Published: March 4, 2017
Abstract: This study aims to show how to obtain the curvature of the ellipsoid depending on azimuth angle. The curvature
topic is quite popular at an interdisciplinary level. It can be to the friends of geometry, geodesy, satellite orbits in space, in
studying all sorts of elliptical motions (e.g., planetary motions), curvature of surfaces and concerning eye-related radio-therapy
treatment, for example the anterior surface of the cornea is often represented as ellipsoidal in form. On the calculation of the
curvature, there is a famous Euler formula for rotating ellipsoid that everyone knows. Let θ be the angle, in the tangent plane,
measured clockwise from the direction of minimum curvature κ1. Then the normal curvature κn(θ) in direction θ is given by
κn(θ) = κ1 cos2θ + κ2 sin2θ = κ1 + (κ2 - κ1) cos2θ I wonder how can a formula for a triaxial ellipsoid? So we started to work.
And we finally found the formula for the triaxial ellipsoid.
Keywords: General Ellipsoid, Normal Section Curve, Principal Curvatures, Gaussian Curvature, Mean Curvature
uses ‘spheroid’ in place of rotational ellipsoid. The standard The parameter lines (u, v) and geodetic (planetographic)
equation of an ellipsoid centered at the origin of a Cartesian coordinates (Φ, λ) are orthogonal on rotational ellipsoid but
coordinate system and aligned with the axes. General are not orthogonal on triaxial ellipsoid.
ellipsoid equation as below in [3-5]. In this parametrization, the coefficients of the first
fundamental form are
E=[b2cos2 u + a2sin2 u] sin2 v (5)
F=(b2-a2) cos u sin u cos v sin v (6)
2 2 2 2 2 2 2
G=[a cos u + b sin u] cos v + c sin v (7)
I. fundamental form
a.b.c.sin 2 v
e=
(a.b.cos v) 2 + c 2 (b 2 cos 2 u + a 2 sin 2 u )sin 2 v (9)
Figure 1. Triaxial Ellipsoid.
x2 y2 z 2 f =0 (10)
+ + −1 = 0 (1)
a 2 b2 c2
a.b.c
g=
Ellipsoid equation (u,v) Gauss Parametric form (a.b.cos v) + c (b cos 2 u + a 2 sin 2 u )sin 2 v
2 2 2
(11)
x= a cos u sin v
y= b sin u sin v (2) II. fundamental form
a. y
u = arctan (4)
b.x
The Gaussian curvature and Mean curvature can be We will compute H and K in terms of the first and the
calculated from Cartesian coordinates given below formulas second fundamental form.
[14, 17]
e.g − f 2 II
K= = (17)
1 E.G − F 2 I
K= 2
x 2 y2 z2 (15) G.e − 2 F . f + E.g
a.b.c. 4 + 4 + 4 H= (18)
a b c 2( E.G − F 2 )
2.2. Principal Curvatures, Gaussian Curvature, Mean
x2 + y 2 + z 2 − a 2 − b2 − c2
H= 3/ 2
(16) Curvature
x 2 y2 z2
2 ( a.b.c ) 4 + 4 + 4
2
We will now study how the normal curvature at a point
a b c
varies when a unit tangent vector varies. In general, we will
Landscape Architecture and Regional Planning 2017; 2(2): 61-66 63
see that the normal curvature has a minimum value κ1 and a associated with that particular normal section. This
maximum value κ2,. This was shown by Euler in 1760. The curvature κn (α) is called the normal curvature of the
quantity surface at point Po in the direction α.
Then the normal curvature at point Po is given by
K = κ1.κ2 called the Gaussian curvature (19)
e.du 2 + 2. f .du.dv + g .dv 2
and the quantity κ n (du , dv) = (23)
E.du 2 + 2.F .du.dv + G.dv 2
H = (κ1 + κ2)/2 called the mean curvature, (20)
where E, F, G, e, f, g are the fundamental coefficients of the
play a very important role in the theory of surfaces. first and second order.
1 1 Formula (39) above can be re-written in the following way
R1 = = Maximum radii of curvature (21)
κ1 H − H2 −K dv dv
2
e + 2. f . + g .
dv du du
κn = (24)
1 1 du
2
R2 = = Minimum radii of curvature (22) dv
E + 2.F . + G.
dv
κ2 H + H −K 2 du du
The formula for the radius of curvature at arbitrary simply by dividing the numerator and denominator by du2. In
azimuth points up that the fact that the fundamental this form it is obvious that κn is a function of the ratio dv/du.
mathematical quantity is the inverse of these radii, which are If we let cot α = dv / du then (24) becomes
simply called curvatures
e + 2. f .cot α + g .cot 2 α
κ n (α ) = (25)
2.3. Normal Section of a Surface E + 2.F .cot α + G.cot 2 α
Let us construct a normal to a surface at a point Po. Then where
the curve that is described on the surface by any plane
passing through the normal (i.e. containing the normal) is E + F tan θ
cot α = (26)
called a normal section of the surface (Fig. 1). In other words W tan θ
a normal section is a plane section formed by a plane
containing a normal to the surface [15, 12, 1]. A surface may be curved in many ways and consequently
one might think that the dependence of the curvature κ on the
2.4. Curvature of a Surface at a Point angle α might be arbitrary. In fact this is not so. The
following theorem is due to Euler.
Let us construct a unit normal and a tangent plane at
given point Po on surface and consider the curves that are 2.5. Euler’s Theorem
formed on the surface by planes passing through Po
containing the normal i.e. the various normal sections Let θ be the angle, in the tangent plane, measured
passing through point Po. Each normal section passing clockwise from the direction of minimum curvature κ1. Then
through Po possesses a particular curvature at point Po. We the normal curvature κn(θ) in direction θ is given by
can specify a particular normal section by use of a polar
κn(θ) = κ1 cos2θ + κ2 sin2θ = κ1 + (κ2 - κ1) cos2θ (27)
coordinate system constructed on the tangent plane, origin
at point Po, polar axis as some arbitrarily chosen initial ray κn(θ) curvature at azimuth θ
in the tangent plane, and an angle α measured clockwise For spheroid (rotational ellipsoid):
from the polar axis to the plane of the normal section (Fig.
2). The curvature at point Po in direction α is thus given as
the function κn (α). For each value of α there is a curvature
1
N = R1 = Radius of Curvature in Prime Vertical (Max. radii of curvature) (28)
κ1
1
M = R2 = Radius of Curvature in Meridian (Min. radii of curvature) (29)
κ2
3. Numerical Example
Find the curvature of normal section curve at Po point
which contains Po surface of normal and passes P1 point on a
triaxial ellipsoid
θ Angle is a azimuth angle Po-P1 direction and Cartesian
coordinates of Po and P1 point are given below
x2 y 2 z 2
+ + − 1 = 0 (Ellipsoid equation)
25 16 9
(a= 5, b= 4, c= 3) semi-axis
Figure 2. (X,Y,Z) Cartesian and (Φ, λ, h) Geodetic coordinates on Triaxial
ellipsoid.
Cartesian coordinates (x,y,z)
Po (3.000 2.500 1.4981) P1 (2.6189 2.4125 1.8047)
First, we need to find the azimuth angle between the two Geodetic coordinates [18]
points known as the Cartesian coordinates. Let’s assume that Φo = 40.194814370 o λo = 52.47573738o ho = 0
Landscape Architecture and Regional Planning 2017; 2(2): 61-66 65
[11] Harris W. F. Curvature of ellipsoids and other surfaces. [16] Moritz, H., Advanced Physical Geodesy, Herbert Wichmann
Ophthalmic Physiol Opt 2006; 26: 497–501. Verlag Karlsruhe. 1980.
[12] James R. C., Mathematics Dictionary, 5th Edition Number, [17] Zhang Hongxin, Feng Jieqing, Preliminary Mathematics of
Springer Netherlands, ISBN 978-0-412-99041-0, 1992. Geometric Modeling (4), State Key Lab of CAD&CG,
http://www.cad.zju.edu.cn/home/zhx/GM/003/00-sg.pdf,
[13] Klein, P., "On the Ellipsoid and Plane Intersection Equation," 2006.
Applied Mathematics, Vol. 3 No. 11, pp. 1634-1640. doi:
10.4236/am.2012.311226, 2012. [18] URL-1
http://www.mathworks.com/matlabcentral/fileexchange/46248
[14] Lipschutz M, Schaum’s Outlines – Differential Geometry, -converter-cartesian-coordinates-to-geodetic-coordinates.
Schaum’s/McGraw-Hill, 1969, ISBN 0–07–037985–8.
[19] URL-2
[15] Ligas M., Cartesian to geodetic coordinates conversion on a http://www.mathworks.com/matlabcentral/fileexchange/52958
triaxial ellipsoid, J. Geod., 86, 249-256. 2012. -intersection-ellipsoid-and-a-plane.