See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/347970953

Topology Optimization of Compliant Mechanism for Laparoscopic Surgery

Instruments

Conference Paper · October 2020

DOI: 10.1145/3438872.3439123

CITATIONS READS
0 201

3 authors:

Prabhat Kumar Rupesh Eknath Ghyar

Meril Life Sciences Pvt. Ltd. Indian Institute of Technology Bombay
10 PUBLICATIONS 12 CITATIONS 25 PUBLICATIONS 90 CITATIONS

SEE PROFILE SEE PROFILE

B. Ravi
Indian Institute of Technology Bombay
145 PUBLICATIONS 2,520 CITATIONS

SEE PROFILE

All content following this page was uploaded by Prabhat Kumar on 03 February 2021.

The user has requested enhancement of the downloaded file.

Topology Optimization of Compliant Mechanism for Laparoscopic
Surgery Instruments
Prabhat Kumar Rupesh Ghyar Bhallamudi Ravi
Mechanical Engg. Department Mechanical Engg. Department Mechanical Engg. Department
Indian Institute of Technology Indian Institute of Technology Indian Institute of Technology
Mumbai, Maharashtra, India Mumbai, Maharashtra, India Mumbai, Maharashtra, India
prabhat.ks21@gmail.com rupesh.ghyar@betic.org prof.b.ravi@gmail.com

ABSTRACT KEYWORDS
Compliant mechanisms are flexible single-piece structures that Compliant mechanism, Topology optimization, Finite Element
produce the desired movement by going through elastic Analysis
deformation, which is different from the traditional mechanisms
that are joined by rigid bodies. Some of the advantages of compliant
mechanisms include reduced friction, wear, noise and the 1 Introduction
possibility to generate unconventional actuation as well as ease of
Nature creates designs that are generally strong but not necessarily
manufacture and assembly. A critical step in the design of
stiff; they are compliant and often have embedded actuation and
compliant mechanisms is topology optimization for material
sensing capabilities. On the contrary, human-engineered devices
distribution. It helps designers to extend and fine-tune a design that
are conventionally designed to be robust as well as stiff. Engineers
already has near-optimum material distribution. It involves
create complex assemblies to perform electromechanical functions
determining the shape and location of holes of a structure as well
by assuming that the individual components are infinitely rigid. But
and the connectivity of the domain.
they have to face problems due to wear, backlash, and noise to meet
This study presents a compliant mechanism design for precision, cost, and reliability requirements [1]. Various practical
laparoscopic surgery instrument in which a feasible topology benefits can be obtained by taking the advantage of the elasticity of
(configuration) is evolved to fulfil a prescribed input-output force- the materials, leading to monolithic compliant mechanisms with
displacement relationship. Two different methods are used for this embedded sensing and actuation. Compliant mechanisms are
purpose. In the first method, an objective function is formulated to versatile single-piece structures that produce the desired movement
capture the need for compliance by going through the desired by going through elastic deformation, which is different from the
deformation (kinematic requirement) and stiffness to resist external traditional mechanisms which involve rigid body motions of joints.
loads (structural requirement) once the mechanism adopts the Compliance in design results in fewer, no-assembly (Fig. 1),
desired configuration. In the second method, the objective function monolithic mechanical devices being joined together and is
is to minimize the volume fraction given the design constraints as especially suitable for applications with a limited range of
limitation to translation of nodes. The optimization process is movements
implemented in MATLAB and Altair OptiStruct software. The
resulting topology is converted into 3-D CAD model followed by
Finite Element Analysis to determine the stress and deformations.
Some observations and recommendations are noted based on these
studies.

CCS CONCEPTS
• General and reference ~ Document types ~ General conference
proceedings

Permission to make digital or hard copies of all or part of this work for personal or Figure 1: Compliant mechanism based surgical instrument. [4]
classroom use is granted without fee provided that copies are not made or distributed for
profit or commercial advantage and that copies bear this notice and the full citation on
the first page. Copyrights for components of this work owned by others than ACM must Due to the absence of conventional joints in these flexible
be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to
post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
structures, they offer many benefits, such as manufacturing
Request permissions from Permissions@acm.org. flexibility, reduced assembly costs, reduced friction, wear, and
RICAI 2020, October 17–19, 2020, Shanghai, China noise [2]. These systems are especially suitable for built-in sensing
© 2020 Association for Computing Machinery.
ACM ISBN 978-1-4503-8830-6/20/10…$15.00 with active material actuators for haptic feedback and embedded
https://doi.org/10.1145/3438872.3439123 actuation [3]. Compliant mechanism-based surgical instruments
provide a range of possible advantages over conventional change here [5]; however, node positions and beam dimensions can
Minimally Invasive Surgeries (MIS) devices and existing robotic differ throughout the optimization. The system output is fine-tuned
systems. Compliant mechanisms [5] implemented in MIS in terms of geometric advantage (motion amplification) and/or
instruments allow complex movement completion with natural or other efficiency metrics.
intuitive hand movements and haptic feedback to be accurately
controlled. Due to the uniform nature of compliant mechanisms, Deepak et al. [7] presented a comparative study of five different
they have the advantage of no wear debris, pinch points, and no formulations reported in the literature. These are used to solve three
lubrication, all of which are serious in the body's sensitive internal benchmark examples with the same requirements for input and
environment [3]. Further, monolithic design simplifies the output, and the same algorithm for numerical optimization. They
processes of fabrication and sterilization. The built-in return spring implemented about 35 different prototypes of synthesis for
action is another advantage of the compliant instruments; a surgeon illustration using the different formulations. The resulted designs
needs only to apply force in one direction to operate the device in are analyzed and correlated with all other objective functions. This
both [6]. On the other hand, force feedback problem is complicated study provides a full description of the benchmark problems and
by the spring action. The surgeon may simultaneously experience tests. Therefore, by considering aspects of implementation, it is
both the cutting / grasping force of the procedure and the rigidity of argued here that MSE / SE formulation is the most straightforward
the instrument and will have trouble differentiating between them. and convenient for mechanism design even as the stiffest structure
design minimizes mean compliance.
An organized method of design of compliant mechanisms is
available in the literature. The design of compliant mechanisms In this paper, a compliant mechanism design is established in which
involves two main aspects as follows: a feasible topology (configuration) is developed to meet the
prescribed input-output force-displacement relationship. In this
(i) Topology synthesis – It refers to the connectivity of the study, two methods are employed to optimize the topology of the
material. It involves the generation of a functional design in the compliant mechanism for designing an instrument for laparoscopic
form of a possible topology starting from input/output surgeries. The resulting topology from both the methods is then
force/motion specifications. compared and converted into a 3-D CAD model. The Finite
Element Analysis is carried out in ANSYS software to find out the
(ii) Shape and size optimization – Size optimization refers to the stress and deformations. The comparison between two different
cross-sectional area of each segment, and shape optimization methods of the optimization process is presented.
deals with the orientations of the connecting sections and
locations of the junctions.
2 Input Design Domain Specifications
The basic step in designing a compliant mechanism is to create a In order to design any compliant mechanism, the very first step is
kinematically functional design that generates the desired output to find out the topology of the mechanism, and for that topology an
motion when subjected to prescribed input forces. This is termed input design domain is required which needs to be optimized to get
topological synthesis. In this stage, though the size and shape of the final one. As an initial starting point, a rectangular domain is
individual elements can be optimized to some extent, the local considered for input whose dimensions are shown in Fig 2. Since
constraints, including stress and buckling constraints, cannot be the laparoscopy instruments are designed to be of outer diameter
enforced. (OD) of 5 mm, the dimensions of other regions of the domain are
also decided in agreement to the OD.
Kota et al. illustrate a design method in which they demonstrate
how an algorithmically compliant gripper is created. In Stage I
topology synthesis, a conceptual design is automatically created,
starting with functional requirements of desired forces and
displacement. The design domain is defined, i.e., physical space
constraints are set for the mechanism. Then, a method of
optimization, such as linear sequential programming (SLP), or a
nonlinear procedure, such as a genetic algorithm (GA), is used to
change the parameterized structure until the objective function is
achieved. Based on the results of the optimization, a mechanism
topology is created, which indicates how many elements the Figure 2: Dimensions of the design domain
structure has and how they are interconnected. In Stage II, the exact
size, shape, and configuration of each beam element are optimized As per ISO 7151 standards for laparoscopy surgical instruments, a
based on material constraints (permissible stress, strain), force of approximately 10 N needs to be applied on the grasper
manufacturing constraints (minimum feature size, etc.), external surface, which is called grasper force for this research, denoted
loads, and desired mechanical advantage. The topology may not by 𝐹𝑔 . For simplicity, this force is assumed to be acting as point
 Topology Optimization of Compliant Mechanism for
WOODSTOCK’18, June, 2018, El Paso, Texas USA
Laparoscopic Surgery Instruments

load on the upper part (shown in Fig. 3 (a)). The pulling force (𝐹𝑝 ) corresponding output displacement is 𝑼𝒐𝒖𝒕 , and the displacement
which needs to be acted on the lower part is calculated using the of the origin is 𝑼𝒊𝒏 [7]. If K is the efficient stiffness matrix of the
force balance and moment balance equations (shown in Fig. 3 (b)) discrete finite element model and the output spring, and u is the
based on the grasping force. global displacement vector, then the problem of optimization is
formulated as:
MSE U out U in
Max  
 SE 0.5uT Ku 0.5 FinU in (1)
This specific objective function often provides a measure of the
mechanism’s geometric advantage. Here, ρ denotes the design
variables in this and other formulations. This method is called
mutual strain energy/strain energy (MSE/SE) formulation
Figure 3: (a) Direction of forces acting on the design domain (b) because 𝑈𝑜𝑢𝑡 and Uin represent the MSE and SE, respectively, and
Free body diagram of the design domain 0.5𝐅𝐢𝐧 is just a factor. This optimization is carried out in MATLAB.

Force equation used for the design domain: In this approach, the specified design domain is first divided into
 Fx  0  Fy  0 several nodes. Each node is connected via a modular array of beam
 F1x  Fp  F1 y  Fg
elements to several other nodes. This works as an initial guess.
Some nodes are ‘fixed’ to signify the points at which the
Moment equation for the design domain: mechanism is anchored to the substrate. The cross-sectional area of
 MA  0 each beam element works as the design variable with the upper and
 Fg  20  Fp  3  0 lower limits defined. The goal is then to distribute the material in
Fg  20 such a way as to optimize the objective function. During the
 Fp    66.67 N
3 optimization process, all beam elements whose cross-sectional area
Considering the factor of safety (FoS) equal to 2.5, the required exceeds the lower bound are eliminated, leaving only a network of
input force becomes: beam elements whose surface has reached the upper boundary. That
determines the topology of the compliant mechanism.
𝐹𝑝 = 66.67 * 2.5 = 166.67 N
After calculation of the input force required, the geometry was The early guess is a sectional beam structure (Fig. 4 (a)) with a
optimized for its topology based on two different methods which uniform distribution of cross-sectional areas. The force F of 66.67
are explained in the next section. N is acted horizontally (x-axis), which is causing a deformation
along the vertical line. When the algorithm converges, the solution
will consist of beam members whose design variable has exceeded
3 Topology Optimization Using Strain Energy the upper bound. This eliminates members of beams whose size
Method variables surpass the lower bound limit. The optimized solution and
corresponding finite element model are shown in Fig. 4 (b), where
In the design of any compliant mechanism, the first step is to obtain the red lines denote the deformed shape, and the solid lines indicate
a feasible topology for the required objective. This step aims to the undeformed shape. The initial optimization resulted in
create a feasible topology (configuration) to meet the prescribed removing the materials from inside and the edges are very thin.
force-displacement relationship input-output. Even if multiple Since the outer diameter of the design is very small (5 mm), this
input forces and multiple-output displacements can be design will not be able to withstand the required forces.
recommended, for simplicity's sake, only a single input-output case
is considered [8]. Here the device input is a force acting at a single
point, and the output required is the displacement at a given point.
An objective function is formulated based on this design
specification, which captures the need for: (a) compliance to
undergo the desired deformation and (b) stiffness when the system
assumes the desired configuration to withstand external loads. We
then use a formal structural optimization technique to synthesize a
form that is an ideal topology, shape, and size of a mechanism that
is compliant enough to perform the intended purpose.

Problem Formulation: A force 𝑭𝒊𝒏 is applied in this formulation

Figure 4: (a) Initial guess (b) Optimized topology showing
at the input port, and spring is incorporated at the output port. The
deflection
Hence, another iteration (Fig. 5 (a)) is carried out in which the (Analysis > forces). A force of 66.67 N is applied in the x-axis in
constraint is given to the edge material that they will not be this case (Fig. 6). After this, the optimization phase is starting
removed. This region is called Non-Design domain because the where all the optimization parameters are defined.
optimization or removal of materials from this region will not be
performed. The optimized solutions are given in Fig. 5 (b) along
with the deformations.

Figure 6: Initial topology and loading conditions

Optimization process: Instead of actually changing the shape of

the part, it selects the mesh of the design volume and assigns some
element density value to every element. In the first iteration, all the
Figure 5: (a) Iteration 2 for topology (b) Optimized topology for elements in the design volume are assigned the value given by
iteration 2 showing deflection volume fraction constraint (i.e. if the volume fraction is set at an
upper bound of 0.3, it assigns all the elements in design volume an
4 Topology Optimization using Volume element density value of 0.3). Then the resulting volume is
Minimization calculated. As per this volume, it assigns a new element density
value to all the elements. It is increased for the elements having
displacement more than the average and decreased for the elements
Topology optimization aims to find the optimum material and voids
with displacements value lower than the average. This new model
distribution from a given design domain. Finite element method
is again analyzed for the displacement and used for next iteration.
(FEM) is used to solve this problem, and design domain is divided
The change in the displacement is monitored and used as a
into discrete elements (mesh). The resulting problem is then solved
converging criterion. The optimization runs for 53 iterations in this
using optimization methods to figure out which elements are
case and after that the topology is obtained (Fig. 7 (a)). The
material and which ones are not. This results in a so-called 0-1
convergence curve of the entire optimization process is shown in
problem. The elements either exist or not, which is an integer
figure 8. The problem formulation for the optimization process is
problem with two separate states of each part, a so-called ISE
briefly described below:
topology (Isotropic Solid or Empty elements) [9]. The
mathematical formulation of the method is reported in the article  Objective Function – Minimize Volume
by Bendsoe and Kikuchi [10]. In this work structural optimization  Response – Volume and Deflection
is carried out using Optistruct, a module of Hyperworks software  Design Constraint – Deflection < 0.5mm where forces are applied
(Altair Engineering Inc). Another module called HyperMesh is
used to discretize (mesh) a CAD model, set boundary conditions, According to the simulation results, the region in red (Fig. 7(b)) has
properties and options and to set up the problem to be solved the least material density followed by the colors green to blue in the
(optimization, static analysis, modal analysis etc.). These are saved order of increasing material density. That is, the material in red
in a file and processed using Optistruct; the results are evaluated color can be conveniently removed from the geometry without
using the HyperView postprocessor. affecting the strength of the part.

First the geometry is created and its material is assigned. In this

case, steel is selected as the part material and since it is a 2-D
optimization, ‘Shell’ element is assigned to the model. The
geometry is automatically meshed using 2D quad elements to
enable easier element quality control. The element size of 0.5 mm
is used for this analysis. Hypermesh uses modules called ‘Load Figure 7: (a) Convergence Curve (b) Optimized topology
Collectors’ to store the loads and ‘Load Step’ to store a loading
condition. Load collectors are linked to load step to define a loading 5 Results
condition. Two load collectors were created to accommodate all the
types of loading conditions this case. The first load collector has The topology resulting from different optimization methods is
the boundary condition constraint in which all the DoF are realized in concept design. Based on the results, the material is
restricted as it is a fixed support case. The second load collector distributed on the edges and is inter-connected to give deformation
includes the force which is added by using ‘Force’ function to the top end. From topology optimization, all dimensions for the
 Topology Optimization of Compliant Mechanism for
WOODSTOCK’18, June, 2018, El Paso, Texas USA
Laparoscopic Surgery Instruments

concept design were translated. (Fig. 8). The CAD model is created 6 Conclusions and Future Work
using SolidWorks. As an initial guess, the depth (out of plane
thickness) of the prototype was set to be 5 mm. This study presented a compliant mechanism design in which a
feasible topology (configuration) is established to meet prescribed
input-output force-displacement relationship. Results show that
this approach of using topology optimization with displacement
constraints can produce a successful design concept. Two different
methods were used to optimize the topology of the compliant
mechanism for designing an instrument for laparoscopic surgeries.
The topology resulting from both methods came out quite similar.
The given topology is converted into 3-D CAD model and then
Finite Element Analysis is carried out to determine the stress and
deformations.

The optimization of topology produces qualitative results in which

it provides a mechanism that functions kinematically. It cannot
present the performance features prescribed for a mechanism.
Therefore the next sensible step is to perform a size and shape
optimization once the topology is known. The shape and size
optimization needs to be carried out to get the final optimized
design. Only after that will the design be finalized, and the
(a) (b) prototyped will be fabricated and checked for the targeted design
Figure 8: (a) Initial guess (b) Optimized topology showing requirements.
deflection

The Finite Element Analysis (FEA) of the resulting CAD model is

REFERENCES
[1] B. P. Trease, Y. M. Moon, and S. Kota, 2005. “Design of large-displacement
performed in ANSYS Software. Static structural analysis is carried compliant joints,” J. Mech. Des. Trans. ASME, vol. 127, no. 4, 788–798. DOI:
out. The material which is used for laparoscopic instruments is https://doi.org/10.1115/1.1900149.
[2] M. D. Murphy, A. Midha, and L. L. Howell, 1996. “The Topological Synthesis
Stainless Steel. For the analysis, this material is applied, and the of Compliant Mechanisms,” Mech. Mach. Theory, vol. 31, no. 2, 185–199. DOI:
maximum displacement of the top point for the first iteration is https://doi.org/10.1016/0094-114X(95)00055-4.
found to be 0.5 mm, and the stress comes out to be 210 MPa (Fig. [3] S. D. B. Bhargav, S. Chakravarthy, and G. K. Ananthasuresh, 2012. “A
Compliant End-Effector to Passively Limit the Force in Tele-Operated Tissue-
9 (a)) when a pulling force of 166.67 N (using a FoS of 2.5) is Cutting,” J. Med. Devices, Trans. ASME, vol. 6, no. 4, 1–7. DOI:
applied. The displacement is quite low, which needs to be improved 10.1115/1.4007638.
[4] M. B. Hong and Y. H. Jo, 2012. “Design and evaluation of 2-DOF compliant
further while the stress is well within limits. Due to material forceps with force-sensing capability for minimally invasive robot surgery,”
constraints, another iteration of optimization was carried out. The IEEE Trans. Robot., vol. 28, no. 4, 932–941. DOI: 10.1109/TRO.2012.2194889.
maximum displacement of the top point for 2nd iteration is found to [5] S. Kota, K. J. Lu, Z. Kreiner, B. Trease, J. Arenas, and J. Geiger, 2005. “Design
and application of compliant mechanisms for surgical tools,” J. Biomech. Eng.,
be 0.15 mm, and the stress comes out to be 163 MPa when a pulling vol. 127, no. 6, 981–989. DOI: 10.1115/1.2056561.
force of 166.67 N is applied (Fig. 9 (b)). The displacement is lower [6] J. Arata, S. Warisawa, M. Hashizume, and M. Mitsuishi, 2007. “Development
of a dexterous minimally-invasive surgical system,” IFMBE Proc., vol. 14, no.
than the initial iteration, which is not the desired outcome and hence 1, 4153–4156. DOI: 10.1007/978-3-540-36841-0_1053.
needs to be improved further while the stress is well within limits. [7] S. R. Deepak, M. Dinesh, D. K. Sahu, and G. K. Ananthasuresh, 2009. “A
comparative study of the formulations and benchmark problems for the topology
optimization of compliant,” J. Mech. Robot., vol. 1, no. 1, 1–8. DOI:
10.1115/1.2959094.
[8] S. Kota, J. Joo, Z. Li, S. M. Rodgers, and J. Sniegowski, 2001. “Design of
compliant mechanisms: Applications to MEMS,” Analog Integr. Circuits Signal
Process., vol. 29, no. 1–2, 7–15. DOI: 10.1023/A:1011265810471.
[9] G. I. N. Rozvany, 2001. “Aims, scope, methods, history and unified terminology
of computer-aided topology optimization in structural mechanics,” Struct.
Multidiscip. Optim., vol. 21, no. 2, 90–108. DOI: 10.1007/s001580050174.
[10] Martin Philip Bendsoe and Noboru Kikuchi, 1988. “Generating optimal
topologies in structural design using a homogenization method,” Comput.
Methods Appl. Mech. Eng., vol. 71, 197–224.

(a) (b)
Figure 9: Deformation and stress distribution of optimized
topology for (a) 1st iteration (b) 2nd iteration

View publication stats