Dept. of Mech. Engg., K.I.T.

's College of Engineering, Kolhapur

A
MINI PROJECT REPORT
ON
“3D Legged Robot”
SUBMITTED TO THE
SHIVAJI UNIVERSITY KOLHAPUR
IN PARTIAL FULLFILLMENT OF THE REQUIREMENT FOR THE
DEGREE OF
BACHELOR OF ENGINEERING (MECHANICAL ENGINEERING)

SUBMITTED BY
Mr.Prajwal P Yadav
Mr. Aniket D Kasote
Mr. Digvijay Hajare
Mr. Omkar Ajitkar
Mr. Yash Sapkal
UNDER THE GUIDANCE OF
PROF. Jahida Khan

DEPARTMENT OF MECHANICAL ENGINEERING

K.I.T's COLLEGE OF ENGINEERING,
KOLHAPUR
YEAR 2018-2019

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Abstract

Soft robots have received an increasing attention due to their advantages of high flexibility and
safety for human operators but the fabrication is a challenge. Recently, 3D printing has been
used as a key technology to fabricate soft robots because of high quality and printing multiple
materials at the same time. Functional soft materials are particularly well suited for soft robotics
due to a wide range of stimulants and sensitive demonstration of large deformations, high motion
complexities and varied multi-functionalities. This review comprises a detailed survey of 3D
printing in soft robotics. The development of key 3D printing technologies and new materials
along with composites for soft robotic applications is investigated. A brief summary of 3D-
printed soft devices suitable for medical to industrial applications is also included. The growing
research on both 3D printing and soft robotics needs a summary of the major reported studies
and the authors believe that this review article serves the purpose.

1. Introduction

3D printing is an additive manufacturing (AM) process defined as ‘the process of joining

materials to make objects from 3D model data, usually layer upon layer, as opposed to
subtractive manufacturing methodologies, such as traditional machining’ 3D printing
can deliver parts of very sophisticated and complex geometries with no need of post-
processing, built from custom-made materials and composites with near-zero material
waste, while being applicable to a diversity of materials, including smart materials such
as shape memory polymers and other stimulus-responsive materials. Therefore, 3D
printing is a technology that offers increased ‘design freedom’ and allows designers and
engineers to create unique products that can be manufactured at low volumes in a cost-
effective way. One of the main example of the design freedom offered is that
conventional assemblies can be restructured in a single complex structure that could
not be manufactured with the current manufacturing processes. Another driver of the 3D
printing technology is that it is environmentally and ecologically favourable. 3D printing
technologies and methods are growing frequently in terms of application and market
share, spreading into various manufacturing divisions, such as robotics, motorized,
health and aerospace and are expected that this substantial growth will continue over
the next few years.

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

In the last few years, there has been a significant trend towards the use of 3D printing
technology to fabricate soft robots for various applications. Soft robots is a very young
research area and mostly inspired by nature mechanisms which are optimized since
centuries for a particular task. Mechanical robots and machines are made of hard
materials that limit their ability to elastically deform and adapt their shape to external
constraints and obstacles. Although they have the capability to be extremely powerful
and precise, these rigid robots tend to be highly specialized and rarely exhibit the rich
multi-functionality of nature. The soft robots are the next generation of robots which are
elastically soft and capable of safely cooperating with humans or steering through
constrained environments. Just as a mouse or octopus can squeeze through a small
hole, a soft robot must be elastically deformable and capable of steering through
narrowed spaces without inducing damaging internal pressures and stress
concentrations

1.1. Relation between 3D printing & soft robots

3D printing is an extremely useful technique to intelligently place them inside soft

structures. However, one fundamental concern in using 3D printing technology for
developing soft robots is that 3D printable soft materials have a large tendency to
deform under the normally used forces during the building process due to their own
weight so a support material becomes a necessity. There is room for the development
of such a material that can become soft after being printed as a rigid structure. There
are commercially available 3D printers with the ability to develop complex structures
through visco elastic hydrogels to be printed in span with self-supporting structures

This low cost and advanced technique can be used to achieve new heights by creating
those soft robotic products that were not possible earlier. This technique is known as
UltiCast; it can print extremely complex shapes that are very difficult to achieve through
typical 3D printing techniques such as fused deposition modelling (FDM) because each
hot filament layer deforms the subsequent filament layers. A soft actuator can be printed
inside a mould through the technique of UltiCast as it will eradicate the manual casting
process hence resulting in faster speed. The freedom to personalize the robotic
behaviour through controlling the robot geometry has allowed to print a custom-made
soft gripper. Low cost 3D printing process can have several applications in soft robotics.
It has another inspiring aspect that it can be extremely useful in the medical sector as it
can reduce the cost of operation with assisted movement. A soft robotic glove with soft
actuators inside it was recently developed through 3D printing and can be helpful in
moving human fingers. This soft robotic 3D-printed glove can be useful for those who
are suffering from limited hand function, local paralysis and arthritis or it can be used as
a rehabilitation tool. Shape morphing materials with photosensitivity, thermal activation
and responsive to water can also be printed now a days through 3D printing technology.
These materials are extremely useful for soft robotic applications as they can be
brought to any desired shape using heat, light or water. The scientists have even
fabricated a 3D-printed cat tongue as a development in soft robotics after the successful

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

experiments of a jamming-based gripper, a prosthetic hand and a 3D-printed soft robot

with four legs that can walk on non-uniform and rough surfaces such as pebbles and
sand. Such soft robots can take part in rescue operations or used for the applications of
collecting sensor data from dangerous environments. This could only become possible
due to the ability of 3D printing technology to build extremely complex structures
consisting of both soft and hard materials inside soft robot legs. The ability of integrating
soft and hard materials in a single structure can lead us to the realization of more
compliant next generation soft robots that will be safer for human contact than their
predecessors. Researchers have also managed to develop a soft robotic hand through
3D printing technology with the ability feel surfaces like the natural hand of human
beings. Unlike other soft robotic hands that grip and sense through motors, this newly
developed hand by Shepherd and his team uses its external fingertips to collect data
through actually feeling the responsiveness internally. Such advancements in the field
of soft robotics were not possible had it not been the technological progress of 3D
printing technology over the last two decades.

2. MOTIVATION:

Some people believe that control systems give commands to mechanisms. But
mechanisms have a mind of their own: they will obey physical laws. Control is not to
compensate for the limitations of poorly designed mechanisms. The best systems will
have mechanism and control designed to work together in harmony.
-Marc Raibert
Marc Raibert was acknowledged as one of the leaders in the field of legged robotics until he
moved into industry in 1993, and the above extract from a talk he gave made me think about
the mechanisms which people use when trying to make robots walk. It seems to me that almost
without exception they are not designed for walking at all, but rather to be as general purpose
as possible, and indeed that the whole purpose of the “controllers” is to get them to walk in the
first place, not to control walking, as they have no innate ability. Exceptions to this include
passive walkers which are designed not to require a controller (or even power) for walking
down gentle slopes, and spring actuated robots, all of which have springs and dampers built
into the physical robot. However there is a significant problem with this approach: although the
end result is generally a very competent walking robot, because all the springs and linkages
which make the mechanism so effective are permanently in place the better the mechanism is
at walking the worse it becomes at everything else — until in the limit it becomes as
uncontrollable as the walking automata of the turn of the century whose intellectual successor
it is, and it becomes impossible to get it to manoeuvre or climb over obstacles or achieve
anything else which might be desirable in the context of a robot designed to carry out a task.
However, the idea of the mechanism being inherently able to walk makes sense since it would
make the control job so much more straightforward, if only there were a way of retaining the

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

multifunctionality which the more basic robots have. The answer in the end is clear — if you
don’t want the springs all the time, why not simulate them through the motors so that when
you switch off the simulation you are left with your general purpose robot again? There was a
clear precedent for this which encouraged optimism MOTIVATION 3 in the approach — the
vertebrate spinal cord. Take the cat, for example. Clearly this is an extremely versatile animal
capable of an enormous range of movements — yet it can be induced to walk very easily by
sending a simple signal down the spinal cord which excites the Central Pattern Generators
(CPGs) associated with walking.. These do not exactly simulate the springs and links mentioned
above, but they have a similar effect — they actively alter the dynamics of the legs through the
muscles to create a new dynamical system where walking is a stable attractor. The significance
of this point becomes clear when you realise that, both with the spinal cat1 and with a robot
with simulated springs and linkages, what you still have is a dynamical system which can be
controlled: in the cat’s case this is done by various higher centres in the brain, notably the
cerebellum, but in the robot’s case it can be done by a more conventional controller. This
should be much simpler than most walking controllers, as it no longer has to “compensate for
the limitations of a poorly designed mechanism”, but rather controls an (active) walking
mechanism and so only has to concern itself with maintaining the stability of the walking
behaviour on rough terrain and perhaps during gait changes, as well as higher level concerns
such as direction and speed of movement. The latter are very simple to control in vertebrates
— for instance, the higher the excitation of the CPGs in the cat, the faster the cat will go,
changing gaits automatically as it speeds up [Grillner, 1985], and in the lamprey, where Grillner
and his colleagues have mapped the entire structure of the CPGs [Grillner et al., 1991], it is
found that exciting the CPGs on one side of the body more than those on the other side (which
again is very easily done) causes the lamprey to move smoothly away from the excited side.
Strangely, Raibert’s robots [Raibert, 1986, 1988] do not follow his own advice — neither the
mechanism nor any individual part of the controller walks on its own, so the controller has to
do the whole job in one go and make the robot walk as well as control the walking all at once.
This has resulted in all of Raibert’s controllers being carefully handcrafted, a timeconsuming
process, though, despite that, the controllers that he and his successors in the MIT Leg Lab have
made have been the closest yet made to the general purpose controllers that we are looking
for. In fact very few people seem to have designed walking mechanisms at all since automata
were replaced by controlled robots in the 1950s. All those that arguably have done so have two
1 a cat whose spinal cord has been severed just below the brain 4 CHAPTER 1. INTRODUCTION
things in common (except the passive walkers which have no controller) — firstly they are
controlled by neural networks (this is almost inevitable as they are the only well researched
computational system which can be trained and have their own continuous dynamics), but
secondly and more interestingly, although they alter the dynamics of the robots with neural
networks to make them walk, they do not appreciate they have done this but rather describe

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

their neural networks as controllers in their own right and do not go one step further to then
design a (higher-level) controller for the walking robot. In trying to do it all in one go the results
tend to be fairly poor, and certainly considerably less impressive than Raibert’s algorithmic
approach.

3. LITERATURE REVIEW:

We propose an end-to-end solution for creating robotic creatures, implemented as an interactive

tool that allows the user to design the structure and motion of a robot while receiving immediate
feedback on its expected real-world behavior. Design Interface Our design interface is structured
into two viewports . one for editing the structure and motion of the robot, one for displaying the
resulting real-world behavior as predicted by our optimization or through physics simulation.
The heart of the interface is formed by a set of easy-to-use editing tools. Structure Editing The
user starts by loading a description file that specifies an initial skeletal structure of the robot, as
defined through a typical hierarchy of bones connected by joints. Initial geometry is created from
this information and a virtual, uni-axial motor is placed at each joint position. The user can freely
edit the robot’s structure at all times by adding or removing motors, thus altering the morphology
of the design, or by adjusting the position or orientation of the motors. Motion Editing The
motion of a robotic creature is completely described by the trajectories of its joints. However,
authoring motions directly in this high-dimensional space is unintuitive, tedious, and very
unlikely to lead to stable movements. We therefore propose a set of higher-level motion
authoring and editing tools, designed to be mutually orthogonal and intuitive for the user.
Motions are largely characterized by their footfall pattern, indicating for each instant during a
gait cycle which of the legs are in contact with the ground and which ones are in flight. Our
interface displays these information as a time dependent graph, allowing for quick inspection and
direct editing by the user.. In particular, the user can change the duration of the stance and swing
phases for any leg and change the relative ordering of the footfalls. While not all patterns lead to
desirable motions, this is immediately clear upon inspecting the evolution of the support polygon
through time or by observing the motions generated by our framework. The immediate feedback
provided by our framework allows the user to interactively adjust the footfall pattern in order to
find satisfying solutions. Higher-level goals such as the walking direction, speed or turning rate
can be provided by the user so as to specify the behaviour of the robotic creatures that are being
designed. Further, the user can control the overall motion style by editing the movement of the
robot’s center of mass and of the feet trajectories. The motions generated by our optimization-
based framework are guided by this user input, while a set of feasibility constraints ensures that
they are always stable. Optimization Given the structure and motion goals for a robotic creature,
our system computes time-varying motor values for dynamically-stable motions using a
trajectory optimization approach. The user can preview the optimized motions using physics
based simulation and iteratively adapt the design to explore the solution space and converge on a
desired result. In order to enable this seamless forward design experience, to robustly support a
wide variety of morphological features in a unified manner, and to ensure that the resulting
robotic creatures function well using off-the-shelf components, the model that we propose
requires a departure from conventional approaches. Space-time optimization methods that
consider the full dynamics of the systems they compute motions for are most general. However,
the complexity of the underlying models brings about a significant computational overhead—

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

even for simple creatures, generating motions is a matter of several minutes , thus clearly
prohibiting the type of interactive design process we seek to enable. Furthermore, space-time
methods are notoriously characterized by challenging optimization landscapes that often lead to
undesirable local minima. Our model avoids the complexity of considering full system dynamics
for three main reasons. First, we achieve important gains in efficiency, with the process of
optimizing motions taking at most a few seconds. Second, in conjunction with the optimization
scheme we employ, our system consistently converges to high-quality solutions. Last, because
the internal torques and ground reaction forces computed through space-time optimization
cannot be directly reproduced by off-the-shelf servomotors, and as physical actuators present
considerable limitations in terms of speed, bandwidth and strength, we focus on generating
motions that are more conservative In this setting, the dynamic interplay between the
instantaneous center of pressure and the motion of the center of mass is captured sufficiently
well through an inverted pendulum approximation. The simplified dynamics model we use is
adopted from robotics, where it is commonly used for model predictive control formulations
typically decouple the generation of center of mass trajectories, motion of the feet and full-body
joint angles When the morphology and proportions of a robot are fixed, this strategy is typically
sufficient. However, various heuristics are required to ensure that constraints between different
modules are satisfied, e.g., stepping locations can be reached given the center of mass trajectory
and robot kinematics. Given that our design system allows users to generate robot designs with a
vast range of morphological features, such a decoupling is not feasible. Rather, as detailed in
Sec. 4, our trajectory optimization method provides an efficient, unified formulation for
concurrently computing trajectories for the center of mass and feet that are consistent with the
structure and range of motion of the robotic creatures. Finishing Once the design iterations have
converged, we automatically generate 3D geometry for all body parts, including con- nectors for
the motors, which are then sent to a 3D printer for manufacturing

4. PROBLEM DEFINITION

● Current state: 3D printer heads use plastic filament, which has variable diameter and is quite
expensive. If the diameter varies too much, the feed mechanism fails o Operation at
temperatures up to 280°C
● Desired state: 3D printer head that uses ordinary plastic injection molding pellets as its
feedstock. Operation at temperatures up to 380°C
● Constraints: Cost < $500, preferably $200 o Must be compatible with a majority of hobby
grade 3D printers
 There are many modeling and control problems but we will mainly focus on a
stabilization of the biped walking in the field of specified movement. In general, a
bipedal locomotion system consists of several members that are interconnected with
actuated joints. Its point of support changes discretely. The parameters of control are
the time of the beginning of the next step and coordinates of a reference point. So to
stabilize walking it is necessary to define «when and where to put a foot». Feedback is

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

based on the equations of the ideal mechanism - the turned pendulum. The analysis of
its movement allows to solve a problem "when and where to put a foot". In this case
this problem has solution set. For modeling a control system it is necessary to work
out the nonlinear differential equations.
 Most important It is difficult for humans to reach in compact places, or during search
and rescue operations in disaster areas where electricity is not available. So there is a
need to design a six leg moving support mechanism with low cost and simple
mechanism which can carry out the operations. Which can be easy to operate and
effective for carrying out the operations.

5. OBJECTIVES:

1. Project Proposal
2. Mechanical Design & Build of a Leg
3. Main Controller & Leg Board Prototype
4. Software Design
5. Main Controller & Leg Board Design and Production
6. Mechanical Design of a body - Full Assembly of a robot
7. Poster Presentation
8. Main Report

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

6. WORKING PRINCIPLE / METHODOLOGY:

Study of Kinematic
Mechanism

Concept Design of 3 leg

moving mechanism

3D modelling in Catia &

Solidworks respectively

Analysis of 3 legged model in

Ansys

Fabrication of 3 legged robot

with the help of 3d printing

Testing and inspection of

respective model

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

 Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

7. COST OF PROJECT:

Key Features
 3-legged multi-terrain space robot
 The robot takes directional and combat commands from you via a 2.4 GHz handset
 The rotating head covers the 360 degree range

Specifications
Material - Plastic, metal
Package dimensions - 37.5 x 30.5 x 23.5 cm

Cost -
₹ 3642.70/-

8. CONCLUSION
3DP offers unique advantages with respect to fabrication soft robots with a complex external
anatomy shape and internal porous structure. Coupling complicated porous 3D design with AM
techniques can create a range of soft robots with bone and muscles from various materials. For
soft robots, control over mechanical behaviour while retaining the designed structure is very
important. The outside part of the soft robot is denser than the inner part, mimicking such
structures is very difficult using 3DP, which is a challenge related to non-uniform shrinkage
during sintering. Apart from the issues related to all 3DP settings as well as selecting suitable
materials, usage of 3D printing in soft robotics is still a big challenge. In fact, there are many
obstacles along this long and difficult road. The gap between the concept and the practical use of
3D printing comprises the main factors: the necessity of fabricating soft robots based on these
design specifications. Despite all the advances in materials science, and system development
there are still major gaps in this field relative to the printing multiple materials and adhesion
between materials. This field is still new and not much commercial fabrication has been done. A
lot of 3DP techniques like SLS and ink jet printing soft robots have been successfully fabricated.
Most reports have been limited to using models as guide templates and for in vitro and in vivo
experiments, whereas implantations of 3D-printed soft robots in the human body are still rare.
Demand for 3D printing technologies such as SLS and 3DP will increase in the future due to
their capability to make custom soft robots that can be tailored for application-specific and
defect-specific needs. Integrating all key points mentioned as well as finding solutions to cope
with the challenges and issues are important in guiding the progress of these techniques towards

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

achieving the objective of advanced soft robots. Lastly, commercial success depends on new
innovation in soft lithography, 3D printing and other rapid prototyping technologies to mass
produce soft structures and robots that are inexpensive and satisfy market demand. Three-leg
moving support mechanism is developed using simple Klann's linkage mechanism. It is electro-
mechanical mechanism which can also run on batteries. Three leg walker can be operated on
uneven or rough surfaces. It can access compact places were humans cannot access. Strength of
three leg walker is high so it can also be used for carrying loads on uneven surfaces. Adding
night vision cameras and obstacles detecting sensors to the six-leg walker it can be used for
surveillance at night.

9. FUTURE SCOPE

1. Adoption of Industrial Internet of Things (IIoT) Technology

Robots will increasingly deploy smart sensors at the edge of production to collect data previously inaccessible
to manufacturers. This trend is currently underway and will lead to new levels of productivity and efficiency.

2. Industrial Cybersecurity as a Priority

As robots become more connected to internal systems for data collection, the cybersecurity risks increase.
Manufacturers will be forced to address vulnerabilities in their processes and invest heavily in cybersecurity to
ensure safe, reliable production.

3. Big Data Analysis Becomes a Competitive Differentiator

Robots will become a key source of information on the factory floor. The collection of data, however, is just one
piece of the puzzle. Manufacturers will have to implement systems to organize and analyze all of this
information in order to act on it.

4. Open Automation Architectures Will Be Implemented

As robotic automation gains widespread adoption, the need for open automation architectures grows. Large
industry players will work with industry organizations to produce standards and open documentation that make
robotic integration easier while improving product compatibility.

5. Virtual Solutions Will Invade Physical Processes

Virtual solutions will become an integral part of industrial robotics. One current growing application is the virtual
representation of robotic systems for proof of concept and offline programming.

6. Collaborative Robots Will Continue to Grow in Popularity

Collaborative robots can work safely alongside humans and are often far cheaper than their industrial
counterparts. As collaborative robots become more capable in tough industrial settings, they will see greater
adoption by manufacturers with strict ROI requirements.

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Robotic automation has been a revolutionary technology in the manufacturing sector, but it’s still poised to
transform the industry over the next couple of years. The six trends above will be some of the most impactful
advances in robotic automation in the future.

10. REFERENCES

1. http://jacobsschool.ucsd.edu/news/news_releases/release.sfe?id=2211

2. Duda T, Raghavan LV. 3D metal printing technology. IFAC-PapersOnLine

2016;49:103–110. DOI:10.1016/ j.ifacol.2016.11.111

3. ] Ivanova O, Williams C, Campbell T. Rapid Prototyping Journal Additive manufacturing

(AM) and nanotechnology: promises and challenges. Rapid Prototyp J. 2013;19:353–364.
DOI:10.1108/RPJ-12- 2011-0127

4. ] Vaezi M, Seitz H, Yang S. A review on 3D microadditive manufacturing technologies.

Int J Adv Manuf Technol. 2013;67:1721–1754. DOI:10.1007/s00170- 012-4605-2

5. Abhishek Ranadive, Ivan Lewis, Pranay Yadav, Samhita Pashte, Nilesh Ghongade,
“Design and Construction of Hexapod (Six Legged) Robot”, “International Journal on
Recent and Innovation Trends in Computing and Communication”, ISSN: 2321-8169,
Vol. 4, Issue: 4.

6. ] R. Arjunraj, A. Arunkumar, R. Kalaiyarasan, B. Gokul, R. Elango, “Fabrication of Six-

Legged Kinematic Moving Mechanism”, “South Asian Journal of Engineering and
Technology”, Vol.2, No.23, 2016.

7. R.A. Brooks, “A Robot that walks; Emergent Behavior from a Carefully Envolved
Networks”, “MIT Artificial Intelligence Lab”, 1989.

8. Uluc Sarani, Martin Buebler and Daniel E, Koditschelc “A Simple and Highly Mobile
Hexapod Robot”, “International Journal of Robotics Research”, 2001.

9. https://www.zdnet.com/article/strider-a-three-legged-walking-robot/

10. https://en.wikibooks.org/wiki/Robotics/Types_of_Robots/Walkers

11. https://en.wikipedia.org/wiki/Legged_robot

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur

Roll No Name Sign

Dept. of Mech. Engg., K.I.T.'s College of Engineering, Kolhapur