7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
1/21
1
Optical Design of a Variable View Imaging System with
Combination of a Telecentric Scanner and Double Wedge
Prisms
Xiaodong Tao1, Hyungsuck Cho,
1,*and Farrokh Janabi-Sharifi
2
1Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology,
373-1, Guseong-dong, Yuseong-dong, Daejeon, Korea
2Department of Mechanical and Industrial Engineering, Ryerson University, 350 Victoria Street,
Toronto,Canada
*Corresponding author: [email protected]
Recent developments in the micro-electro-mechanical system (MEMS) and biotechnology
have presented an emerging need for observation of dynamic targets in three-dimensional
space. Unfortunately, the conventional microscopes with fixed optical parameters are
difficult to supply sufficient vision information because of occlusion, small field-of-view
(FOV) and low depth resolution. This paper introduces the design of a variable view
imaging system which can supply a flexible view with a relative large zenith angle and a
simple kinematics. Due to several performance factors, a multiobjective optimization
process is applied to achieve an appropriate design. A prototype system is developed and
used to verify the proposed design.
OCIS codes: 110.1080, 220.1080, 120.4820, 150.3040, 170.0110, 170.0180, 220.4830.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
2/21
2
1. Introduction
The small field-of-view (FOV) and fixed view direction of conventional microscope cannot
supply sufficient visual information needs for observation of dynamic targets in three-
dimensional (3D) space, particularly in the fields of the micro-electro-mechanical systems
(MEMS) and biotechnology applications. In the past, there have been several research efforts to
solve this problem. Multiple fixed microscopes or moving stages are often applied in those
applications [1,2]. Such approaches do not provide sufficient flexibility for observation of
dynamic targets. A smart optical system with adjustable optical parameters becomes a promising
solution. Potsaid has proposed an adaptive scanning optical microscope (ASOM) [3]. Through
the integration of a scanning mirror and deformable mirror, the ASOM can obtain a large FOV
with high resolution. A similar methodology is also applied in a scanning optical telescope [4].
Another system capable of achieving foveated viewing in a large FOV has been demonstrated
using a spatial light modulator (SLM)[5].
In our previous research, a proposal was made for an active optical system which can change
the view position and direction [6,7]. However, because of the coupling effect between the
scanning mirror angle and view angle, the view angle is found to vary with scanning mirror,
causing its zenith angle of view to decrease, making the kinematics complex. In order to solve
those problems, a decoupled design of the system is proposed in this paper, which employs a
telecentric lens. The basic concept of the system is presented in SPIE conference [8]. Since this
design contains a number of parameters that affect the performance of te system, optimal
election of these parameters is necessary. For this, an multiobjective optimization design process
is described in this paper in detail for the first time. Experimental results are presented to show
the validity of the proposed design.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
3/21
3
2. Concept of the variable view imaging system
The objective of the design of the variable view imaging system is to change the view state by
control of active optical components in the system. The view state includes view position and
view angles defined as Eq. 1.
T
w wV x y = (1)
where (xw,yw) is the view position in a world coordinate. and are the azimuth angle and the
zenith angle of the view.
In the robot area, by installing a camera on a robot, an active vision system can change the view
state easily. However, considering the size and weight of vision system and speed of scanning, it
is difficult to install a microscope system on a robot. In order to realize this function, the idea is
to design the vision system with active components which mimic the function of robot. Fig. 1
shows the concept of the system. To steer the view angle (, ), double wedge prisms are applied
because the compact size for deflection of light. The function of double wedge prisms is similar
to a camera installed on a pan/tilt stage. It can steer the beam in any direction in a cone. To steer
the view position (xw,yw), the telecentric scanner is applied. Its function is similar to a camera
installed on the translation stage. Therefore by integration of these two active optical components,
the proposed system can steer the both of view position (xw,yw) and orientation (, ). It can be
considered as a camera installed on a robot with four degrees of freedom as shown in Fig. 1.
In our previous research, a scanning mirror and double wedge prisms were used for view steering
[6,7], where the scanning mirror angle was coupled with the zenith angle of the view. The zenith
and azimuth angles changes with the change of the scanning mirror angle. There are several
issues for the old design. The first issue is a small zenith angle. When the target locates near the
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
4/21
4
optical axis, the motion of scanning mirror decrease the zenith angle of view because the
coupling effect between the scanning mirror angle and view orientation. The second issue is the
complex kinematics relationship. Each variable in the view state is related to all variables in joint
space as V=[fy(),fx(),f(),f()]T, where =[sx, sy, p1, p2]
T is the vector of joint variables
for scanning mirror and wedge prisms, respectively. fy(),fx(),f() and f() are the functions
to calculate the four variables in the view state as shown in Eq. 1. The calculation of joint
variables for a desired view state becomes difficult. And a precise alignment and calibration are
required to obtain an accurate kinematics equations. To decouple the rotation angle of scanning
mirror and view angles, a telecentric lens group is designed and introduced to the system. The
system aperture is projected on the focal point of the telecentric lens. It makes the entrance pupil
infinitely. Therefore in the object side, the chief rays are always parallel to the optical axis.
The layout of the whole system is shown in Fig.2. The 3 rd and 4th lenses integrate a deformable
mirror into the system. The deformable mirror is used for correction of aberration induced by
wedge prisms. The 4th and 5th lenses relay the active aperture of the deformable mirror to the
system aperture. The system aperture is projected onto the scanning mirror, which is located at
the focal point of the telecentric lens. It makes an entrance pupil infinitely. Therefore the rays are
always parallel to the optical axis in the object side. The azimuth angel and zenith angle of view
are only related to the rotation angle of the prisms.
In order to determine view state given the joint variables, a rigorous analysis of the view
position and angle is made by the ray tracing method. The forward kinematics of the system can
be defined as follows,
1 2 1 2 1 2 1 2( , , , ) ( , , , ) ( , ) ( , )
T
x sx sy p p y sx sy p p p p p pV f f f f = (2)
http://en.wikipedia.org/wiki/Entrance_pupilhttp://en.wikipedia.org/wiki/Entrance_pupil7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
5/21
5
where V(xw,yw, , ) is the vector for view state. And (sx, sy, p1,p2 ) includes the joint variables
for scanning mirror and wedge prisms. The detail derivation can be found in [7]. The inverse
kinematics can be achieved by using a damped least-squares (DLS) method, which can also be
refer to [7].
3. Optimal design of the view steering part
The view steering part include scanning mirror, telecentric lens and wedge prisms. The
performances of the system, such as scanning area, maximum zenith angle, wavefront error are
all related to this part. In this section, the system performance is defined. Because several system
performance factors are involved, a muliobjective optimization process is applied to give an
appropriate design.
A. System performance for the system design
System performance can be identified from the analysis of the functions of the system. The main
function of the proposed system is to avoid the occlusion by changing the view angle. The
effects of system parameters for the occlusion case can be shown in Fig. 3, where the image of a
cubic with a depth D is captured by the proposed system. In the top view, the corner point B is
occluded by the point A as shown in the figure. With a zenith angle , both point A and B can be
observed in the image plane. Because of the telecentric lens group used in the system, the scaled
orthographic projection model can be applied in the system. The distance between A and B in the
image plane is defined as following equation.
AB od Md=
(3)
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
6/21
6
whereMis the magnification of the system. d0 is projected length of A and B along optical axis,
which is defined as
0 sin( )d D = (4)
In order to avoid occlusion, the distance between A and B need to be observed in the image
plane. Therefore the condition for the proposed system to avoid occlusion is as follows,
sin( )o
r D < (5)
where ro is the optical resolution of the system. We can find that with high resolution r and larger
zenith angle , the system can easily avoid occlusion. Therefore the maximum zenith angle and
the resolution are two important performance factors for occlusion avoidance.
Another function of the proposed system is to increase the field of view by steering the view
position. In the whole scanning area, the area A2 with full azimuth angle with maximum zenith
angle is most important as shown in Fig. 4. In this area, the system can achieve the view with an
azimuth angle from 0 to 2 and a zenith angle from 0 to maximum zenith angle.
As an optical system, image quality is also an important performance factors. Because the
wedge prisms are used in the system, a relative large wavefront error is induced into the system.
In this design, a deformable mirror is applied to correct this aberration. However the wavefront
error correction is limited due to the limited stroke of the actuator in the deformable mirror. In
the proposed system, a micromachined membrane deformable mirror (37ch, D15mm, OKO) is
applied. The limitation of wavefront correction for this mirror is 3.5 (wavelength =0.660 m)
according to [10]. Therefore in system design, the wavefront error should be kept inside this
limitation. The details calculation of these system performances are described as follows.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
7/21
7
Maximum zenith angle, max
When two wedges have the same rotation angle, the zenith angle achieves its maximum value as
shown in Fig. 5(a). Therefore for a given v, the maximum zenith angle is defined as
max (0,0)v
f = (6)
Therefore max is a function of the vertex angle of the wedge prisms defined as follows,
maxmax( )
vF = (7)
Fig. 5 (b) shows the final result of the relationship. Larger vertex angle of the wedge prisms can
provide a larger zenith angle.
Radius of view area, r2
The configuration of the system for calculation of radius of A2 is shown in Fig. 6(a). The
radius of the prism is defined as rp. The view will located at the boundary of the A2 when
s px r= , 0sy = , 1 0p = ,
2 0p = . Therefore the radius of A2 for a given v can be obtained
from following equation,
2( ,0,0 ,0 )
vx pr f r
= (8)
r2is a function of the radius and vertex angle of wedge prism, defined as follows,
2 2( , )
r p vr F r
=
.(9)
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
8/21
8
This relationship is shown in Fig. 6(b). Comparing to the vertex angle v of prisms , the radius
of the prism gives more effects on the change of r2.
Optical Resolution
The Optical resolution is the ability of a microscope to allow one to distinguish between small
objects. The optical resolution ro is depend on the numerical aperture as shown in Eq. 9.[3]
0.61o
rNA
=
(10)
where is the wavelength. NA is the numerical aperture, which characterizes the range of angles
over which the system can accept. It can be defined as following equation.
sin( / 2)NA n = (11)
where n is the refraction index, is the angle of the maximum cone of light that can enter the
objective lens.
Peak-Valley Aberration, WPV
The aberration of the system is variable depending on the wedge prisms angle and scanning
mirror angle. It can be corrected by a deformable mirror which was applied in the system.
However, because the correction limit of the deformable mirror, the peak-valley wavefront error
should small than the correction range of the deformable mirror. For a specific design, the peak-
valley wavefront is depending on the configuration of the prisms and mirror. The worst
configuration with the maximum peak-valley wavefront error can be obtained from Zemax
software as shown in Fig. 7(a), where xs=rp, ys=0, p1=0,p2=180. The maximum peak-valley
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
9/21
9
wavefront error is a function of vertex angle of prisms, radius of the prism and numerical
aperture defined as follows,
( , , )PV pvW p v
W F r NA= (12)
It is difficult to achieve this relationship analytically. From Zemax software, the warefront
errors for different rp, v and NA can be obtained easily as shown in Fig. 7(b). However, this
discrete result is not accurate enough for the system design. Therefore a Backpropagation (BP)
networks are built for function approximation [11]. In this design, a standard BP networks with 2
hidden layers is applied. Each hidden layer has 10 neurons. The Tan-sigmoid transfer functions
are applied in the networks. The data from the Zemax software was used for training data.
B. Optimized Design of the system
The objective of the optimized design is to achieve best performance of the system. In this
design, there are four system performance factors for optimization. The object functions can be
formulated as follows,
2 max, ,,
max ,max , max , minp vp v v
o pvr NAr NA
r r W
(13)
When the deformable mirror is applied the system, because the wavefront error inside a specific
range can be corrected by the deformable mirror, the Wpv can be a boundary constraint in system
design. Because the relationship between ro and NA defined in Eq. 9, ro can be replaced by NA
to decrease the number of parameters for the design. Therefore the objectives of the system
design can be simplified as following equation,
2 max, ,
max ,max ,maxp v p vr r
r NA
(14)
Subject to ( , , )pv p vW r NA <
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
10/21
10
where is the correction limit of wavefront error for the deformable mirror. In this design, is
set as 3.5 . As can be seen from Eq. 14, the optimal design becomes a multiobjective
optimization problem. The performance factors are conflicted with each other. For example, The
increase of vertex angle v of the prisms will increase the maximum zenith angle max, and also
increase the wavefront error and decrease the radius of view area, r2. The increase of rp increases
the radius of view area, r2, and also increases the wavefront error. Therefore, we can not get a
single optimized solution from Eq. 14. In order to solve this Multi-Objective optimization
problem, -constraint method is applied in this design according to [12]. This method is to select
one of the objective functions to be optimized and all the other objective functions become
constraint. This method can simplify the multiobject optimization problem to single objective
optimization. In this design, the boundary of the numerical aperture NA, radius of view area r2
can be defined based on the application. The objective of the design is to maximize the zenith
angle max. Therefore the optimization problem can be simplified as
max, ,
maximizep vNA r
(15)
Subject to2 , ,r NA pvr K NA K W <
Kr and KNA are the boundary of the r2 and NA, which can be determined based on the
application by the designer.
In order to solve the inequality-constrained optimization, a Logarithmic barrier method is
applied [34]. The boundary of view area is determined by the application. In the microassembly
application, the Kr is set as 5 mm. KNA is 0.03 in this design. The design result is shown in Table
1. As can be seen, the maximum zenith angle,max
is 22.29. In the real experiment, the in-stock
components are selected with the most similar value achieved in the optimal design. The
parameters for the real experiment setup are shown in Table 2. Based on the catalog of the wedge
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
11/21
11
prisms, the maximum vertex angle of the wedge prisms are 188, which gives a 19 degree of
zenith angle. The zenith angle in the real experiment setup is only 3.3 less than the optimal
design result.
As discussed in the Section 3.1, the NA and max are two important factor for occlusion
avoidance. By setting different KNA, the relationship between NA vs. max can be achieved from
the optimal design solutions as shown in Fig. 8. As can be seen, when NA decrease to 0.02, 30
degree of zenith angle max can be achieved. When the NA increases to 0.1, only 4 degree of
zenith angle max can be achieved. This tradeoff relationship between NA and max is caused by
the limitation of wavefront error correction for the deformable mirror. The NA and max in the
current experiment system is shown in the Fig. 8, which is near the optical design set.
4. Experiment results
In order to investigate the feasibility of the proposed optical system design, A prototype system
was constructed according to the Table 3. The 37-channel MMDM was supplied by OKO
Technologies. The detailed information for wavefront correction can be referred to [10]. The
deviation angle of one wedge prism was 10. The wedge prisms were installed on the compact
rotation stages. Four achromatic lenses with effective focal length of 150 mm were used in the
system.
A. Experiment for the real object
The ability of the proposed system to change the view without moving the object was
demonstrated by multiple views of micro gears. Three views with different configurations are
shown in Fig. 9(a) ~ (c). Fig. 9 (a) shows the view with a zenith angle =18.9 and an azimuth
angle =90, where the hole in the 2nd gear can be observed. Fig. 9 (b) shows the view with a
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
12/21
12
zenith angle =18.9 and an azimuth angle =0. The depth of the 2nd gear can be observed. Fig. 9
(c) shows the view with a zenith angle =18.9 and an azimuth angle =270, where the other
side of the 2nd can be observed. In order to compare the imaging results with the previous system
[7], the view state for the previous system is achieved in the proposed system. The max,2, for the
previous system is determined from system kinematics, which equals to 13.8[7]. The three view
states with zenith angle =13.8 is achieved as shown in Fig. 9(d)~(f). The interested features on
the side of gears on the image plane are much smaller than the current system. The heights of
gears, h1, h2, on the image plane for the previous system are 26 and 18 pixels respectively. In the
proposed system, they are 36 and 25 pixels respectively, which is more than 35 percent larger
than the coupled system. During online operation, the joint variables for a desired view state
need be calculated in a short time. The numerical solution based on the damped least-squares
(DLS) method is implemented, which is similar to the previous system [7]. Because of the
simplified kinematics, only 23 iterations are need for calculation of joint variables, which is
approximately 32ms in the current system. However, for the previous system, more than 350
iterations are needed, which will spend 490ms.
B. Experiment for the real application
In the microassembly experiment, a microassembly system including three translational
motions was installed near the proposed system. The preliminary experiment setup was only
intended to evaluate the performance of the system; therefore, the microassembly system was
teleoperated without automation. The task is to insert the micro electrode into the micro slot as
shown in Fig. 10 (a). The widths of slot and electrode are 300um and 250um respectively. The
electrode is held by a gripper. As shown in Fig. 10 (b). At an initial state, the view angle is set as
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
13/21
13
=0 and =0. The captured image is shown in Fig. 10 (b). The micro electrode will align with
the slot in X direction as shown in Fig. 10 (c). However, the insertion part of the electrode cannot
be observed because of occlusion. In order to align the electrode and slot in Y direction, the view
angle is steered by the proposed system, and Fig. 10 (d) shows the image when a zenith angle
is 19 and the azimuth angle is 0. With the visual information of the side surface, the electrode
is aligned with the slot in Y direction as shown in Fig. 10 (e). The final assembly task is to insert
the electrode into the slot. Fig. 10 (f) shows a final state of the assembly.
5. Conclusion
This paper presents the design of a variable view imaging system which can change the view
angle and view position by actuating active optical components in the system. The proposed
system with a telecentric lens group and double wedge prisms can supply a larger zenith angle of
view and simpler kinematics relationship compare with the previous system [7]. During the
system design, multi performance factors are identified. In order to obtain the best performance,
a multiobjective design process is applied. The experimental results showed the validity of the
proposed system to achieve a larger zenith angle at different azimuth angle. The focus of the
future research will be on the automated operation of the system.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
14/21
14
Reference
1. M. Probst, K. Vollmers, B. E. Kratochvil and B. J. Nelson, Design of an advancedmicroassembly system for the automated assembly of bio-Microrobots, presented at 5th
International Workshop on Microfactories, Besancon, France, 2527 October 2006.
2. X. Hui, R. Weibin and S. Lining, A flexible experimental system for complexmicroassembly under microscale Force and vision-based Control, Int. J. Optomechatronics
1, 81102, 2007.
3. B. Potsaid, Y. Bellouard and J. Wen, Adaptive Scanning Optical Microscope (ASOM): amultidisciplinary optical microscope design for large field of view and high resolution
imaging, Opt. Express 13, 65046518, 2005.
4. C. Scott, B. Potsaid and J. Wen, Off-Axis aberration correction for a wide field scanningtelescope, in Proc. SPIE 7266, 72660Y, 2008
5. H. Hua and S. Liu, A dual-sensor foveated imaging system, Applied Optics 47, 317-27,2008
6. X. Tao, D. H. Hong, H. S. Cho, The design of active vision system for variable viewimaging of micro objects, in Proc. SPIE 6376, 637608112, 2006.
7. X. Tao, H. S. Cho and F. Janabi-Sharifi, Active optical system for variable view imaging ofmicro objects with emphasis on kinematic analysis, Appl. Opt. 47, 4121-4132, 2008
8. X. Tao, D. H. Hong, H. S. Cho, Variable view imaging system with decoupling design,International Symposium on Optomechatronic Technologies, Proc. of SPIE, 7266, 72661U-1
- 11, 2008.
9. J. R. Meyer-Arendt, Introduction to Classical and Modern Optics, Prentice Hall, 1995.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
15/21
15
10.L. Zhu, P. C. Sun, D. U. Bartsch, W. R. Freeman and Y. Fainman, Wave-front generation ofZernike polynomial modes with a micromachined membrane deformable mirror, Appl. Opt.
38, 60196026, 1999.
11.S. Kumar, Neural Networks: A Classroom Approach, McGraw-Hill, 200512.K. Miettinen, Nonlinear Multiobjective Optimization, Kluwer Academic Publishers, 199913.H. S. Cho, Optomechatronic: Fusion of Optical and Mechatronic Engineering, CRC Press,
2005.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
16/21
16
Fig. 1. Concept of the variable view imaging system
Fig. 2. The layout of the proposed system.
Fig. 3. Avoid occlusion in the proposed system
Fig. 4. View area A2 with full azimuth angle with maximum zenith angle
Fig. 5. (a) The configuration of the wedge prisms with maximum zenith angle (b) The
relationship between the vertex angle v (degree) of the wedge prism and max (degree)
Fig. 6. (a) The configuration of the system for calculation ofr2(b)The relationship between
2
, ,p vr r
Fig. 7. (a) The worst configuration with the maximum peak-valley (b) The warefront errors for
different rp, v and NA
Fig. 8. NA vs. max in the optimized design
Fig.9. Multiple views of micro gears: (a)~(c) captured images in three different azimuth angle
with =18.9 and (d)~(f) =13.8
Fig. 10. Microassembly of a micro electrode and a slot, (a) microassembly task, and (b)~(f)
captured images during the assembly.
Table 1 The optimal design result
Table 2 Specification of the experiment setup
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
17/21
17
Pan/tilt
stageCamera
Scanning
lens
Scanning
mirror
Translation stage
Wedge
prisms
Fig. 1. Concept of the variable view imaging system
camera
telecentric
lens
scanning
mirror
wedge
prisms
deformable
mirror 2nd lens
3rd lens
4th lens
5th lens
6th lens
system
aperture
Xw
Zw
Ow
Yw
Zv
Xv
YvOv
(xw,yw)
view direction
Object plane
Fig. 2. The layout of the proposed system.
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
18/21
18
A
A
Telecentric lens
groupA
B
B
Image plane
Occlusion
Without occlusion
do
D
di
Fig. 3. Avoid occlusion in the proposed system
Zv
Xv
Yv
Ov
max,2
=0~2
Fig. 4. View area A2 with full azimuth angle with maximum zenith angle
v
max
0 10 20 30 40 50 600
10
20
30
40
50
60
70
80
90
max
v
(a) (b)
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
19/21
19
Fig. 5. (a) The configuration of the wedge prisms with maximum zenith angle (b) The
relationship between the vertex angle v (degree) of the wedge prism and max (degree)
pri sms
rp
2r2A 5
10
15
20
25
0
10
20
30
-10
0
10
20
30r2(mm)
Fig. 6. (a) The configuration of the system for calculation ofr2(b)The relationship between
2, ,
p vr r
1st wedge prism
2nd wedge prism
Object plane0.1
0.20.3
0.40.5
510
1520
250
10
20
30
40
50
60
70
80
90
rp v
Wpv
() NA=0.1
NA=0.09
NA=0.01
(degree)(mm)
(a) (b)
Fig. 7. (a) The worst configuration with the maximum peak-valley (b) The warefront errors for
different rp, v and NA
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
20/21
20
0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.15
10
15
20
25
30max
NA
Curr ent exper i ment
syst em
Fig. 8. NA vs. max in the optimized design
h1
h2
500m 500m 500m
2nd gear
1st gear
h2
(a) =18.9, =90
(b)=18.9
, =0
(c)=18.9
, =270
h1
h2
500m 500m 500m
h2
(d) =13.8, =90 (e)=13.8, =0 (f)=13.8, =270
Fig.9. Multiple views of micro gears: (a)~(c) captured images in three different azimuth angle
with =18.9 and (d)~(f) =13.8
7/28/2019 2010_AO_3D Scan_telecentric Scanner and Double Wedges
21/21
500um
Gripper
Electrode
Slot
500um
(a) (b) (c)
500um 500um 500um
(d) (e) (f)
Fig. 10. Microassembly of a micro electrode and a slot, (a) microassembly task, and (b)~(f)
captured images during the assembly.
Table 1 The optimal design result
MaximumZenith angle,max
22.29
Scanner area, r2 5.56mmNA 0.0302Wavefront error,WPV 3.498, =685nmVertex angle, v 24.3Radius of prisms, rp 15.15 mm
Table 2 Specification of the experiment setup
Maximum Zenith angle, max 19
Scanner area, r2 4.35 mm
NA 0.033Wavefront error,WPV 2.4, =685nm
Vertex angle, v 18.13
Radius of prisms, rp 12.5 mm