Professional Documents
Culture Documents
Gaynor1
Topology Optimization Group,
Civil Engineering Department,
Johns Hopkins University,
3400 N. Charles Street,
Baltimore, MD 21218 Multiple-Material Topology
e-mail: agaynor1@jhu.edu
Nicholas A. Meisel
Optimization of Compliant
Design, Research, and Education for Additive
Manufacturing Systems Laboratory, Mechanisms Created Via
Virginia Tech,
Randolph Hall,
460 Old Turner Street,
PolyJet Three-Dimensional
Blacksburg, VA 24061
e-mail: meiselna@vt.edu Printing
Christopher B. Williams Compliant mechanisms are able to transfer motion, force, and energy using a monolithic
Mem. ASME
structure without discrete hinge elements. The geometric design freedoms and multimate
Design, Research, and Education for
rial capability offered by the PolyJet 3D printing process enables the fabrication o f com
Additive Manufacturing Systems Laboratory,
pliant mechanisms with optimized topology. The inclusion of multiple materials in the
Virginia Tech,
topology optimization process has the potential to eliminate the narrow, weak, hingelike
Randolph Hall,
sections that are often present in single-material compliant mechanisms and also allow
460 Old Turner Street,
for greater magnitude deflections. In this paper, the authors propose a design and fabri
Blacksburg, VA 24061
cation process for the realization of 3-phase, multiple-material compliant mechanisms.
e-mail: cbwill@vt.edu
The process is tested on a 2D compliant force inverter. Experimental and numerical per
formance o f the resulting 3-phase inverter is compared against a standard 2-phase
James K. Guest design. [DOI: 10.1115/1.4028439]
Mem. ASME
Topology Optimization Group, Keywords: design for AM, topology optimization, PolyJet, 3D printing, multiple materials,
Civil Engineering Department, compliant mechanisms, material jetting, robust design
Johns Hopkins University,
3400 N. Charles Street,
Baltimore, MD 21218
e-mail: jkguest@jhu.edu
1 Additive Manufacture (AM) of Multimaterial make a strong case for the need of multimaterial compliant mech
Compliant Mechanisms anisms in the medical field [3]. By including both a stiff and flexi
ble material phase in the design of contact-aided compliant
Howell defines compliant mechanisms as those which utilize mechanism forceps for natural orifice translumenal endoscopic
the deformation of flexible members to successfully transfer surgery, the authors were able to achieve larger total jaw openings
motion, force, and energy [1]. This is in direct contrast to tradi and blocked forces. This improved mechanism performance has
tional mechanisms that rely on movable joints in order to perform the potential to directly impact the success rate of the surgery.
their function. Compliant mechanisms are encountered on a daily However, Aguirre and Frecker’s design was limited by their intui
basis in the forms of binder clips, paper clips, and various compli tive understanding of how forceps should look. This paper takes a
ant latches. In addition to the various man-made examples, nature more systematic design approach based on topology optimization
also makes use of compliant mechanisms, with many living to leverage multimaterial AM processes.
organisms displaying parts that are both strong and flexible [2]. PolyJet 3D printing is one of the only AM processes capable of
Advantages of compliant mechanisms include part consolidation utilizing stiff and flexible material phases within a single build,
and improved mechanism robustness. However, as the design of making it uniquely qualified for manufacturing complex, multi
compliant mechanisms increases in complexity, traditional manu material compliant mechanisms. PolyJet 3D printing is an AM
facturing methods become infeasible. This drives the authors’ material jetting process, wherein droplets of liquid photopolymer
overall goal of integrating design optimization with AM methods, are deposited directly onto an elevator substrate via a series of
with a particular focus herein on the design and fabrication of inkjet printheads [4]. As the material is deposited, two ultraviolet
compliant mechanisms. lamps cure the photopolymer in multiple passes. Each subsequent
While there are many examples of single-material compliant layer is jetted on top of the previous one. A representation of this
mechanisms present in everyday life, man-made, multimaterial process can be seen in Fig. 1.
compliant mechanisms are rare. This is because manufacturing The PolyJet process offers a high resolution print, with a layer
complexity increases significantly with the introduction of addi thickness of 16-30 pm and an in-plane resolution of 42 pm. In
tional material phases. The potential benefit, however, may like addition, the PolyJet process offers one significant and unique
wise increase significantly. For example, Aguirre and Frecker advantage among modem AM process: the PolyJet process is
capable of depositing two different materials on a pixel-by-pixel
'Corresponding author. basis. One material is a rigid, white plasticlike material (Ver-
Contributed by the Manufacturing Engineering Division of ASME for publication
in the J ournal of M anufacturing S cience and E ngineering. Manuscript received oWhite+), while the other is an elastomeric, flexible black mate
April 15, 2014; final manuscript received August 17, 2014; published online October rial (TangoBlack+). The two materials can be combined in
24, 2014. Assoc. Editor: Joseph Beaman. various ratios to create nine gradient material blends with
Journal of Manufacturing Science and Engineering DECEMBER 2014, Vol. 136 / 061015-1
Copyright© 2014 by ASME
Print Block Travel d irectio n (X)
Print Block
M odel M aterials
Support M aterial
B u ild T ra y
properties ranging along the continuum of the two extremes. By that minimizes (or maximizes) the defined objective function
including multiple material phases such as these in the design of within a prescribed set of design constraints. This results in the
compliant mechanisms, the maximum deflection of the mecha effective and efficient use of material within the part. The use of
nism can potentially be improved, while potentially decreasing the topology optimization approach as applied to the design of
the likelihood of fatigue failure at the structure’s jointlike compliant mechanisms can be traced back to work by Sigmund,
sections. as well as by Frecker et al. [5,6].
As Sec. 1.2 will show, the field of topology optimization in AM
1.1 Introduction to Compliant Mechanism Design and is incredibly varied, with different researchers using different FE
Topology Optimization. In general, the compliant mechanism representations and optimization algorithms according to the con
design process can be separated into a series of key decisions that text of the particular problem, as well as personal preference.
the designer must make. Each one of these decisions serves to
lead the designer towards a final design methodology. The deci 1.2 Topology Optimization in AM. While little to no work
sions include the general approach to be used (kinematics-based has yet been done regarding the manufacturing of optimized, mul
or optimization-based), the finite element (FE) representation of timaterial compliant mechanisms via AM (to be discussed further
the design space (continuum, discrete, or hybrid), and the appro in the Sec. 1.4), several researchers have investigated the use of
priate optimization algorithm (gradient-based or stochastic). A AM as a means of realizing topology-optimized parts, including
decision tree that represents these key decisions in the design pro small scale material microstructures (e.g., Refs. [7-9]). The “free
cess is shown in Fig. 2. complexity” inherent in the AM process makes it ideal for the
The first decision is whether or not to pursue a kinematics based realization of final optimized parts. While there are several topol
approach or a topology optimization approach. For the kinematics ogy optimization groups looking at manufacturing processes, the
approach, the designer equates the desired compliant mechanism following section seeks to elucidate the larger hubs specializing in
design to more traditional rigid-link kinematics design. This manufacturing research that have also pursued design
approach relies heavily on the designer’s intuition and preconcep optimization.
tions regarding the final compliant system. In this way, it does not At Loughborough University, work has been performed to
fully leverage the design freedom allowed by AM and will not be assist in the design of optimized artifacts while specifically con
pursued herein. sidering the necessary manufacturing constraints provided by
For the topology optimization approach, the general compliant AM. Brackett and coauthors recently offered an overview of some
mechanism design domain is defined (with applied forces, sup of the largest perceived opportunities in this sector, including the
ports, and desired responses) and material is systematically dis importance of mesh resolution, support material constraints, and
tributed (added or removed) throughout the domain in a manner adaptations of the solid isotropic material with penalization
C o m plian t
M echanism
Design
Journal of Manufacturing Science and Engineering DECEMBER 2014, Vol. 136 / 061015-3
intractable for large-scale optimization problems such as 1.5 Context. The study presented in this paper demonstrates
continuum-based topology optimization. a start-to-finish process for the realization of optimized, multima
terial compliant mechanisms. This represents an important first
step in unlocking the design potential of the multimaterial PolyJet
1.4 Manufacturing of Multimaterial Compliant Mechanisms. process. The authors determine an appropriate compliant mecha
While literature has offered some discussion regarding how to nism design process, based on the decision tree presented earlier
optimize the design of multiple material compliant mechanisms, in Fig. 2, in Sec. 1.1. A SIMP and projection-based optimization
there has been little content detailing their actual fabrication. The method (Sec. 2.2) is applied to the design of a compliant force in
few instances of literature pertaining to the fabrication of multiple verter, a well-known compliant mechanism case-study. Results
material compliant mechanisms will be discussed herein, but it is from experimentally testing the printed multimaterial optimized
important to note that none of the objects fabricated have been structures are provided in Sec. 3. Concluding remarks are offered
subjected to structural optimization. Following a review of the lit in Sec. 4.
erature, the authors conclude that there is no prior work where
multiple material compliant mechanisms have been designed,
optimized, and subsequently fabricated. 2 Process for Design and Manufacturing of 3-Phase
One of the more prevalent examples of the manufacturing of Compliant Mechanisms
multiple-material compliant mechanisms is from Bailey and Raja-
This section discusses the optimization approach that was
gopalan. They discuss the design and manufacturing of a biomi-
implemented to design optimized compliant mechanisms. Section
metic leg that operates under the principle of heterogeneous
2.2 discusses the multivariate SIMP optimization method, and
material compliance [36,37]. While the final design is not driven
how it is applied to multiple material optimization, hi addition,
by the concept of optimization, the authors specifically address
Sec. 2.1 will discuss the logic behind the selection of this
the process of multimaterial. They adapt the process of shape dep
approach.
osition manufacturing (SDM) to allow for the creation of flexible
joints while maintaining stiff members for the rest of the leg
shape. SDM involves the deposition of material in layers, fol 2.1 Determination of Compliant Mechanism Design Pro
lowed by machining in order to form the material layer into the cess Suitable for PolyJet Printing. As has already been men
desired shape (in this way it is like a combination of AM and tra tioned in Fig. 2, the design of compliant mechanisms can be
ditional CNC machining). Because the process offers continuous divided into a hierarchal decision tree. For the first decision, we
access to the part interior, specialized subpieces can be embedded have already determined the use of the kinematics approach does
during creation. In this case, the authors embedded separate flexi not sufficiently leverage the potential of AM, so we instead follow
ble joints in their biomimetic leg. a topology optimization path. The next decision is dependent on
Several authors have also investigated the use of multimaterial how the designer wishes to represent the FE discretization in the
molding (MMM) for the creation of multiple material compliant design space. The discrete element representation, such as that
mechanisms [38-40]. MMM is a process whereby the various used in truss and frame topology optimization with the ground
materials in the final part are created volumetrically, as opposed structure approach, has the potential to drastically reduce the com
to the layer-by-layer methods of both AM and SDM. While there putational intensity of the optimization routine, due to the (rela
are several variations on the process, the general MMM flow tively) low number of degrees of freedom. However, this comes at
involves the creation of a one material phase being molded sepa the cost of resolution and design freedom, as the design domain
rately and then being inserted into a mold for the second stage ma has already been restricted by in selecting the ground structure. A
terial phase. Filling this second stage mold will embed the first continuum representation, on the other hand, offers the potential
material phase within the part. for a more free-form representation of topology (depending on the
For the fabrication of small-scale multiple material compliant chosen mesh size). It is worth noting that a hybrid representation
mechanisms, there are two examples that are derivations of the might be able to balance the speed of the discrete representation
MMM process. Rajkowski proposes a prototyping process that with the resolution of the continuum method. While such hybrid
uses a curable rigid polymer as well as a curable, flexible silicone approaches generally exist in literature, such as a truss-continuum
as the two material phases [41], By placing the material phases models simultaneously optimized to place steel and concrete
down in bulk and using a mask to cure only the desired sections materials [43-45], the authors are unaware of any hybrid repre
of the part, the author offers a quick, inexpensive solution for the sentations being used in conjunction with multiple material AM at
fabrication of multiple-material mechanisms on the millimeter this time.
scale. Vogtmann proposes a process whereby the negative space The authors have instead chosen to pursue a continuum repre
for the flexible material phase is cut from a bulk piece of the rigid sentation, due in part to the quality of its resolution as well as the
phase [42]. The flexible material is deposited, cured, and planed, way in which a continuum representation aligns with the PolyJet
before the desired mechanism profile is cut from the bulk process’ method of printing. When printing, the PolyJet process
material. utilizes a series of multicolored bitmaps that are sent to the
While the above processes have been shown to successfully printer. Each bitmap represents a single slice of the printed part,
create multiple material compliant mechanisms, they all also have with multiple colors used in each slice to denote the material to be
limitations when considering complexity and distributed compli deposited. While the ability does not currently exist, the authors
ance of the final pieces. The examples presented are relatively hope to eventually be able to use the image outputs from 2D to
geometrically simple when compared to traditional results of pology optimization as a direct bitmap slice input to the printer. In
multiple-material optimization, and thus were all manufacturable. this way, translating the topology optimization output to a Stand
However, these processes do not scale well. As the complexity of ard Tesselation Language (STL) file will become unnecessary and
topology and multimaterial distribution increases, the processes the process of manufacturing optimized multimaterial compliant
will require significantly more user interaction and time invest mechanisms will become more streamlined.
ment to create the necessary mechanisms. In addition, the pre The final decision to be made when considering the design de
sented examples all rely on the principle of lumped compliance, cision tree in Fig. 2 is whether to solve the chosen formulation
where the flexible material phase is implemented at the location with a gradient-based optimization algorithm or stochastic search
that would traditionally be represented by a revolute joint. These optimization algorithm. Stochastic algorithms, such as genetic
processes would be ill-prepared to manufacture mechanisms algorithms and particle-swarm optimization, randomly sample the
based on distributed compliance, where the flexible material design space and are thus capable of handling discrete formula
phases would be more interspersed among the rigid material. tions and facilitating escape from low performance local minima.
Journal of Manufacturing Science and Engineering DECEMBER 2014, Vol. 136 / 061015-5
of the compliant phase within a member (as before), and at pres situation where two solid elements are connected only at a comer
ent the algorithm requires the materials to have equal increments node. One node hinges allow for lumped compliance and the per
in stiffness (AE). We note the latter may be quite appropriate for formance of such elements is overestimated with low-order FEs.
the multi-material Polyjet process. Obviously, if a one-node hinge were printed, it would instantane
ously fail due to the stress concentration at a point. While projec
3 Case Study: Compliant Force Inverter tion methods enable control of a minimum feature size, it was
discussed in the original works that design variables could theoret
In order to demonstrate the utility of the presented optimization ically deposit two tangent circles, which would manifest in a one-
and printing method, the authors consider the well-established node hinge [48]. A number of researchers have specifically
example of a force inverter compliant mechanism. This case study tackled the one-node hinge issue in the context of the compliant
was initially demonstrated in Ref. [5] and has become one of the inverter. Sigmund [60], for example, simultaneously optimized an
benchmark problems in topology optimization.- As seen in Fig. 4 eroded and dilated version of the topology to mimic over- and
the design domain for the mechanism is square, with the displace under-etching, respectively. This led to a min-max formulation,
ments at the top and bottom points on the left side of the design with the idea that over-etching would lead to a disconnected struc
domain fixed. An input force is applied to the left hand-side of the ture, and thus zero performance, if one-node hinges were present.
space, along with an input spring constant value. A reaction force and While the method successfully eliminates one-node hinges from
spring constant are also applied to the right hand side of the space. designs, an actual “blue-print” design, which is passed to the man
The objective of the study is to maximize the work done on the output ufacturer, is not clearly identified. A number of other authors have
spring. If the ratio of kom to km is larger, greater force transfer to the tackled the issue by using Monte Carlo simulation to represent
output location is targeted. Conversely, the ratio of Iout to km is manufacturing uncertainties in the context of projection schemes
smaller, greater displacement of the output location is targeted. [61,62] and level set methods [63].
It should be noted that the analysis used in the topology optimi This paper adopts the same basic idea as Sigmund, employing a
zation was limited to the assumption of small displacements, and min-max formulation that simultaneously optimizes a larger pro
thus linear elastic analysis. This can be achieved by using a small jection and smaller projection of the same design variables. For
magnitude of the applied load. As load magnitude and resulting this paper, however, we consider a minimum length scale / min set
motion increases, literature has shown that the assumption of lin by the user to represent the expected radius of the droplet, and
ear analysis at best underestimates motion of the final topology then vary that droplet size by directly varying the radius r used in
and, at worst, may miss a failure mode [58,59]. However, the cre projection. This introduces two additional length scales, defined
ation of these optimized pieces should still offer a useful point of as
comparison between 2-phase and 3-phase results, even though the
experimental deflection values of each specimen under (relatively) rininlarge ; rinin ~F A r ^
large loads may differ from any predicted theoretical values.
rininsmail = r min A/'
3.1 Optimization: General formulation. In the case of the
where Ar is the variation in length scale. The resulting min-max
inverter problem, a common benchmark in topology optimization,
compliant mechanism optimization formulation then takes on the
the goal is to maximize negative displacement (minimize dis
form
placement) at an output port under a given load F at an input port.
This is expressed mathematically in general as follows: mm max {LTd(,.mlnsmall), LTd(rminlaigc)}
min LTd
^ ( / minsmall) = F
subject to K (0)d = f
K ^ ( ^ ) ( ^ e ) ) d ('-n,i„,„gt) = F (5)
£ /» W < v (3) subject to y e{<j>) / < y
Z J N ' '' min
eeO. e€Cl
Q<<t>i <i v/en o <</>, < i v i e n
where d are the nodal displacements, the unit vector L extracts While the formulation in Eq. (5) is nearly identical to Sigmund
the output port degree-of-freedom, K is the global stiffness ma [60], there is a subtle difference in achieving the geometric pertur
trix, V is the allowable volume of material, ve is the elemental vol
bation: Sigmund’s dilate and erode variations actually simulate
ume, and c)> is the independent design variable vector, describedover-depositing and over-etching, which may represent different
below. All examples were solved using a uniform distribution of manufacturing processes and lead to different concavities of the
material as the initial guess. material interface, while Eq. (5) simulates only the deposition pro
cess and the idea of an inkjet droplet being larger and smaller than
3.2 Robust Topology Optimization Formulation. When anticipated. Though subtle, we feel the latter more accurately
using topology optimization to design compliant mechanisms it is reflects the AM process. A continuation scheme on the [3 Heavi
well-known that solutions may contain one-node hinges, a side parameter is used to achieve a quality solution. The P for rmin_
large is started at 0 and increased by 1 each continuation step.
Alternatively, the p for the rminsmall is held fixed at a magnitude of
2. Although using a continuation method on beta may not be
required [57], we have found that the continuation method
improves convergence in this application. The results found in
r in
__frnAnAl
Design "out this paper performed 11 continuation steps with 60 MMA optimi
|W A * zation iterations for each continuation step. As our focus is on the
multimaterial aspect of these designs, the finer details of this ro
D om ain k 0ut
bust topology optimization formulation and algorithm tuning are
not explored here.
Fig. 6 3-phase inverter result found using the robust com bina Fig. 7 3-phase inverter result found using the robust, m ulti
torial SIM P approach (2:1 stiffness ratio) phase SIMP approach
Journal of Manufacturing Science and Engineering DECEMBER 2014, Vol. 136 / 061015-7
using small values for the fl Heaviside parameter associated with
the projection scheme. This fading effect can be mitigated by sim
ply using larger values of /? (see [57] for full discussion), though
this was not done here.
To demonstrate the ultimate potential of the PolyJet process’s
array of materials, an additional optimization was performed
using a stiffness ratio of 20:1 between the two nonvoid candidate
materials. This ratio is intended to more closely resemble the stiff
ness difference between the stiff VeroWhite+ material and
TangoBlack+, the most elastomeric material offered by the object
process. The optimized topology is shown in Fig. 10 using the ro
bust, multiphase SIMP approach.
The TangoBlack+ and VeroWhite-1- inverter achieved a deflec
tion of 11.58 mm with only 2.75 kg of applied load, as shown in
F ig . 8 C o m p lia n t s p e c im e n s w ith lo a d and c a n tile v e r Fig. 11. This is almost nine times larger in displacement and 3
a t t a c h m e n ts times less in load, or an improvement in efficiency of approxi
mately 30. It is important to note that modulus of elasticity infor
mation for TangoBlack+ has yet to be published by the
manufacturer or by independent researchers, and so the perform
ance of the printed specimen has the potential to differ signifi
cantly from the performance predicted by the optimization
algorithm (since the stiffness ratio is purely an estimation). How
ever, it nevertheless demonstrates the dramatic displacement
improvements that might be achieved when using the most elasto
meric material for the PolyJet process.
It is also interesting to note the double curvature present in the
deformation seen in Fig. 11. This is forced through the robust
F ig . 1 0 3 - p h a s e in v e r te r r e s u lt f o u n d u s in g ro b u s t, m u lt ip h a s e
S IM P a p p r o a c h (2 0 :1 s t if f n e s s r a tio )
(c)
F ig . 9 D e fle c tio n o f (a ) 2 - p h a s e in v e r te r (F ig . 5 ), ( b ) 3 -p h a s e
c o m b in a to r ia l S IM P in v e r te r ( F ig . 6 ), a n d (c ) 3 -p h a s e m u ltip h a s e
S IM P in v e r te r (F ig . 7 ) (a ll u n d e r 9 .6 5 k g a p p lie d lo a d )
there were no springs applied to the output and input ports. Look
ing at the multiphase SIMP (Fig. 7) and combinatorial SIMP (Fig.
6) solutions, we see a thin compliant, border around all stiff F ig . 11 D e f le c t io n o f 3 - p h a s e in v e r te r w it h T a n g o B la c k +
regions. This border is likely not optimal, but instead an artifact of m a te r ia l ( u n d e r 2 .7 5 k g o f a p p lie d lo a d )
Journal of Manufacturing Science and Engineering DECEMBER 2014, Vol. 136 / 061015-9
[28] Maheshwaraa, U., Bourell, D., and Seepersad, C. C., 2007, “Design and [45] Yang, Y., Moen, C. D., and Guest, J. K., “Three-dimensional Force Flow Paths
Freeform Fabrication of Deployable Structures with Lattice Skins,” Rapid and Reinforcement Design in Concrete via Stress-dependent Truss-continuum
Prototyping J., 13(4), pp. 213-225. Topology Optimization,” ASCE Journal of Engineering Mechanics.
[29] Hiller, J. D., and Lipson, H., 2009, “Multimaterial Topological Optimization of [46] Carlo, H. J., Patrick Spicer, J., and Rivera-Silva, A., 2012, “Simultaneous Con
Structures and Mechanisms,” 11th Annual Conference on Genetic and Evolu sideration of Scalable-Reconfigurable Manufacturing System Investment and
tionary Computation, New York, July 8-12, pp. 1521-1528. Operating costs,” ASME J. Manuf. Sci. Eng., 134(1), p. 011003.
[30] Hiller, J. D., and Lipson, H., 2009, “Design Automation for Multi-Material [47] Pang, L., and Kishawy, H. A., 2012, “Modified Primary Shear Zone
Printing,” 20th Annual International Solid Freeform Fabrication Symposium, Analysis for Identification of Material Mechanical Behavior During Machining
Austin, TX, Aug. 3-5, pp. 279-287. Process Using Genetic Algorithm,” ASME J. Manuf. Sci. Eng., 134(4),
[31] Bendspe, M. P., 1989, “Optimal Shape Design as a Material Distribution Prob p. 041003.
lem,” Struct. Optim., 1(4), pp. 193-202. [48] Deuser, B. K., Tang, L., Landers, R. G., Leu, M. C., and Hilmas, G. E., 2013,
[32] Bendspe, M. P., and Sigmund, O., 1999, “Material Interpolation Schemes in “Hybrid Extrusion Force-Velocity Control Using Freeze-Form Extrusion Fabri
Topology Optimization,” Arch. Appl. Mech., 69(9-10), pp. 635-654. cation for Functionally Graded Material Parts,” ASME J. Manuf. Sci. Eng.,
[33] Yin, L., and Ananthasuresh, G. K., 2001, “Topology Optimization of Compliant 135(4), p. 041015.
Mechanisms With Multiple Materials Using a Peak Function Material Interpo [49] Keshavarz Panahi, A., Mianajiy, H., Miandoabchi, E., and Hussaini Fareed, M.,
lation Scheme,” Struct. Multidiscip. Optim., 23(1), pp. 49-62. 2013, “Optimization of the Powder Injection Molding Process Parameters
[34] Saxena, A., 2005, “Topology Design of Large Displacement Compliant Mecha Using the Sequential Simplex Algorithm and Sensitivity Analysis,” ASME J.
nisms with Multiple Materials and Multiple Output Ports,” Struct. Multidiscip. Manuf. Sci. Eng., 135(1), p. 011006.
Optim., 30(6), pp. 477-490. [50] Bryan, A., Hu, S. J., and Koren, Y., 2013, “Assembly System Reconfiguration
[35] Saxena, A., 2002, “On Multiple-Material Optimal Compliant Topologies: Dis Planning,” ASME J. Manuf. Sci. Eng., 135(4), p. 041005.
crete Variable Parameterization Using Genetic Algorithm,” ASME Interna [51] Park, H. S., and Anh, T. V., 2012, “Development of Evolutionary Method for
tional Design Engineering Technical Conferences and Computers and Optimizing a Roll Forming Process of Aluminum Parts,” ASME J. Manuf. Sci.
Information in Engineering Conference, Montreal, QC, Canada, Sept. 29,-Oct. Eng., 134(2), p. 021012.
2, pp. 1-12. [52] Leng, J., Li, Z., Guest, J. K., and Schafer, B. W., 2013, “Shape Optimization of
[36] Bailey, S. A., Cham, J. G., Cutkosky, M. R., and Full, R. J., 1999, “Biomimetic Cold-Formed Steel Columns With Manufacturing Constraints and Limited
Robotic Mechanisms via Shape Deposition Manufacturing,” Robotics Number of Rollers,” Proceedings of the Structural Stability Research Council,
Research: The 9th International Symposium, Snowbird, UT, Oct. 9-12, pp. Annual Stability Conference, St. Louis, MO, Apr. 16-20, pp. 1-19.
403-410. [53] Hofmann, D., Huang, H., and Reinhart, G., 2013, “Automated Shape Optimiza
[37] Rajagopalan, S., Goldman, R., Shin, K.-H., Kumar, V., Cutkosky, M., and tion of Orienting Devices for Vibratory Bowl Feeders,” ASME J. Manuf. Sci.
Dutta, D., 2001, “Representation of Heterogeneous Objects During Design, Eng., 135(5), p. 051017.
Processing and Freeform-Fabrication,” Mater. Des., 22(3), pp. 185-197. [54] Svanberg, K., 1987, “The Method of Moving Asymptotes—A New
[38] Bejgerowski, W., Gerdes, J. W., Gupta, S. K., and Brack, H. A., 2011, “Design Method for Structural Optimization,” Int. J. Numer. Methods Eng., 24(2), pp.
and Fabrication of Miniature Compliant Hinges for Multi-Material Compliant 359-373.
Mechanisms,” Int. J. Adv. Manuf. Technol., 57(5-8), pp. 431- 452. [55] Guest, J. K., Prevost, J. H., and Belytschko, T., 2004, “Achieving Minimum
[39] Bejgerowski, W., Gerdes, J. W., Gupta, S. K., Brack, H. A., and Wilkerson, S., Length Scale in Topology Optimization Using Nodal Design Variables and Pro
2010, “Design and Fabrication of a Multi-Material Compliant Flapping Wing jection Functions,” Int. J. Numer. Methods Eng., 61(2), pp. 238-254.
Drive Mechanism for Miniature Air Vehicles,” ASME International Design En [56] Guest, J. K., 2009, “Topology Optimization With Multiple Phase Projection,”
gineering Technical Conferences and Computers and Information in Engineer Comput. Methods Appl. Mech. Eng., 199(1-4), pp. 123-135.
ing Conference, Montreal, QC, Canada, Aug. 15-18, pp. 69-80. [57] Guest, J. K., Asadpoure, A., and Ha, S.-H., 2011, “Eliminating Beta-
[40] Gouker, R. M., Gupta, S. K., Brack, H. A., and Holzschuh, T., 2006, Continuation From Heaviside Projection and Density Filter Algorithms,” Struct.
“Manufacturing of Multi-Material Compliant Mechanisms Using Multi- Multidiscip. Optim., 44(4), pp. 443-453.
Material Molding,” Int. J. Adv. Manuf. Technol., 30(11), pp. 1049-1075. [58] Buhl, T., Pedersen, C., and Sigmund, O., 2000, “Stiffness Design of Geometri
[41] Rajkowski, J. E., Gerratt, A. P., Schaler, E. W., and Bergbreiter, S., 2009, “A cally Nonlinear Structures Using Topology Optimization,” Struct. Multidiscip.
Multi-Material Milli-Robot Prototyping Process,” IEEE/RSJ International Con Optim., 19(2), pp. 93-104.
ference on Intelligent Robots and Systems, IEEE Computer Society, Mechani [59] Brans, T. E., and Tortorelli, D. A., 2001, “Topology Optimization of Non-
cal Engineering Department, Institute for Systems Research, University of Linear Elastic Structures and Compliant Mechanisms,” Comput. Methods Appl.
Maryland, College Park, MD, Oct. 11-15, pp. 2777-2782. Mech. Eng., 190(26-27), pp. 3443-3459.
[42] Vogtmann, D. E., Gupta, S. K., and Bergbreiter, S., 2011, “A Systematic [60] Sigmund, O., 2009, “Manufacturing Tolerant Topology Optimization,” Acta
Approach to Designing Multi-Material Miniature Compliant Mechanisms,” Mechanica Sinica/Lixue Xuebao, 25(2), pp. 227-239.
ASME International Design Engineering Technical Conferences and Computers [61] Jansen, M., Lombaert, G., Diehl, M., Lazarov, B. S., Sigmund, O., and Scheve-
and Information in Engineering Conference, Washington, DC, Aug. 28-31, pp. nels, M., 2013, “Robust Topology Optimization Accounting for Misplacement
211- 221 . of Material,” Struct. Multidiscip. Optim., 47(3), pp. 317-333.
[43] Gaynor, A. T., Guest, J. K., and Moen, C. D., 2012, “Reinforced Concrete Force [62] Schevenels, M., Lazarov, B. S., and Sigmund, O., 2011, “Robust Topology
Visualization and Design Using Bilinear Trass-Continuum Topology Opti Optimization Accounting for Spatially Varying Manufacturing Errors,” Com
mization,” J. Struct. Eng., 139(4), pp. 607-618. put. Methods Appl. Mech. Eng., 200(49-52), pp. 3613-3627.
[44] Amir, O., and Sigmund, O., 2013, “Reinforcement Layout Design for Concrete [63] Luo, J., Luo, Z., Chen, S., Tong, L., and Yu Wang, M., 2008, “A New Level
Structures Based on Continuum Damage and Trass Topology Optimization,” Set Method for Systematic Design of Hinge-Free Compliant Mechanisms,”
Struct. Multidiscip. Optim., 47(2), pp. 157-174. Comput. Methods Appl. Mech. Eng., 198(2), pp. 318-331.