Conference PaperPDF Available

FreeBOT: A Freeform Modular Self-reconfigurable Robot with Arbitrary Connection Point - Design and Implementation

Authors:
  • The Chinese University of Hong Kong, Shenzhen
  • The Chinese University of Hong Kong, Shenzhen

Abstract and Figures

This paper proposes a novel modular self-reconfigurable robot (MSRR) "FreeBOT", which can be connected freely at any point on other robots. FreeBOT is mainly composed of two parts: a spherical ferromagnetic shell and an internal magnet. The connection between the modules is genderless and instant, since the internal magnet can freely attract other FreeBOT spherical ferromagnetic shells, and not need to be precisely aligned with the specified connector. This connection method has fewer physical constraints, so the FreeBOT system can be extended to more configurations to meet more functional requirements. FreeBOT can accomplish multiple tasks although it only has two motors: module independent movement, connector management and system reconfiguration. FreeBOT can move independently on the plane, and even climb on ferromagnetic walls; a group of FreeBOT can traverse complex terrain. Many experiments have been conducted to test its function, which shows that the FreeBOT system has great potential to realize a freeform robotic system.
Content may be subject to copyright.
FreeBOT: A Freeform Modular Self-reconfigurable Robot with
Arbitrary Connection Point - Design and Implementation
Guanqi Liang1,2, Haobo Luo1,2, Ming Li1,2, Huihuan Qian1,2, and Tin Lun Lam1,2,
Abstract This paper proposes a novel modular self-
reconfigurable robot (MSRR) “FreeBOT”, which can be con-
nected freely at any point on other robots. FreeBOT is mainly
composed of two parts: a spherical ferromagnetic shell and
an internal magnet. The connection between the modules is
genderless and instant, since the internal magnet can freely
attract other FreeBOT spherical ferromagnetic shells, and not
need to be precisely aligned with the specified connector.
This connection method has fewer physical constraints, so
the FreeBOT system can be extended to more configurations
to meet more functional requirements. FreeBOT can accom-
plish multiple tasks although it only has two motors: module
independent movement, connector management and system
reconfiguration. FreeBOT can move independently on the plane,
and even climb on ferromagnetic walls; a group of FreeBOT
can traverse complex terrain. Many experiments have been
conducted to test its function, which shows that the FreeBOT
system has great potential to realize a freeform robotic system.
I. INTRO DUC TIO N
Modular self-reconfigurable robots (MSRR) have become
a hot research topic in recent years [1]–[9]. MSRR system
consists of many repeated modules, which can rearrange
themselves into different configurations according to task
requirements. The previous MSRR modules are difficult to
realize a freeform robotic systems because they have lots
of physical constrains such as: the module connectors are
gender-opposite and discrete; the modules need to plan
trajectories to align the connectors while self-assembly; the
connection between modules is time-consuming and has a
low success rate.
Through the docking mechanism, the MSRR can real-
ize the connection/separation and system reconfiguration
between modules. Therefore, the docking mechanism is
one of the most basic components of MSRR system and
many creative docking mechanisms have been designed. For
example, the hooks that are activated by DC motors [1], [5],
[10]–[12], permanent magnets [2], electromagnets [13], or
electro-permanent magnets [14]. In [15], the author proposed
the concept of “the area of acceptance” for MSRR, which
is defined as “the range of possible starting conditions for
which mating will be successful”; a connector with a larger
area of acceptance has a higher success rate when connect-
ing. The hooks activated by DC motor allow the modules
to be strongly connected, but has a small acceptance area
*This paper is partially supported by funding 2019-INT008 from the
Shenzhen Institute of Artificial Intelligence and Robotics for Society.
1The Chinese University of Hong Kong, Shenzhen.
2The Shenzhen Institute of Artificial Intelligence and Robotics for
Society.
Corresponding author is Tin Lun Lam tllam@cuhk.edu.cn
Fig. 1. A freeform MSRR system - FreeBOT
and needs accurate alignment, which requires the module
units to plan the trajectory when connecting [10]. It is not an
efficient connection mechanism for the MSRR system. Con-
nections between magnets or electromagnets can increase the
acceptance area [2], [14] since an accurate alignment is not
required. When two MSRR modules with paired magnets
(or paired electromagnets) approach, they are automatically
combined together under a magnetic field. However, the
previous connections between magnets or electromagnets
must be gender-opposed; it will increase some path planning
constraints for the connection between modules [3], [4], [16].
Since the configuration of the previous MSRR system is
restricted by the location and gender of the connector, it has
become an growing consensus to equip the MSRR module
with multiple connectors. If one module can be connected
to multiple modules at the same time, the configuration of
the MSRR system will be enriched to meet more functional
requirements. However, the module with multiple connectors
not only increases the weight, volume and manufacturing
cost of the robot, but also brings complex physical constraints
for path planning in the algorithm level. Therefore, it is
still challenging to design an effective and freeform MSRR
module.
This paper proposes a novel MSRR called FreeBOT
(Freeform Robot), which can be connected together freely in
an effective way with fewer physical constraints (as shown
in Fig. 1). FreeBOT has the same basic functions as the most
advanced MSRR: modules can move independently, modules
can be connected/separated without manual assistance, and
system configurations can be rearranged. In addition, The
connection between the modules is genderless and instant,
since the internal magnet can freely attract other FreeBOT
Fig. 2. Assembly exploded diagram of FreeBOT
spherical ferromagnetic shells, and not need to be precisely
aligned with the specified connector. When it comes to
motion performance, a FreeBOT can travel along planar
surfaces, and even climb ferromagnetic slopes or walls; a
group of FreeBOTs can be rearranged into different config-
urations to travel through more complicated terrain. Since
this connection has fewer physical constraints, the FreeBOT
system can be extended to more configurations to meet more
functional requirements, which has great potential to realize
a freeform robotic system.
This paper is organized as follows. Section II describes the
mechanical design. The motion of FreeBOT and experiment
results are introduced in Section III and IV respectively. Sec-
tion V compares FreeBOT with the state-of-the-art MSRRs.
Finally, conclusions and future work are given in Section VI.
II. MEC HANICAL DESIGN
A. FreeBOT design
Fig. 2 is the assembly exploded diagram of the FreeBOT.
FreeBOT is a spherical robot equipped with internal magnets,
which is mainly composed of two parts: an ferromagnetic
spherical shell and an internal driving mechanism. The inter-
nal driving mechanism is a vehicle equipped with two rubber
wheels, which are driven by two DC motors through gear
boxes. A strong permanent magnet is installed at the bottom
of the internal vehicle, and the ferromagnetic spherical shell
is made of iron, so the internal vehicle will always adhere
to the spherical shell due to the great attraction caused
between them. It is a non-touch connection between the
magnet and the inner surface of the spherical shell, since
the internal vehicle only touch the inner surface through the
rubber wheel, the magnet and the spherical iron shell are
not in physical contact, so the magnet is easy to move in the
spherical shell. Two casters are placed on the front and back
of the internal vehicle through a strip-shaped aluminum alloy
plate to ensure that the internal vehicle remains balanced
in the spherical shell. The gravity of FreeBOT can be
changed through changing the position of the internal vehicle
in the spherical shell by controlling two DC motors, so
that FreeBOT rolling on the plane can be realized. Due to
Fig. 3. Magnetic field excited by the internal magnet
Fig. 4. Magnetic attraction versus distance between two FreeBOTs
the powerful internal magnet, a single FreeBOT can move
on the slot, even on the vertical ferromagnetic surface. In
addition, the position of the internal magnet in the spherical
shell depends on the position of the internal vehicle, this
makes a freeform connection become possible, which will
be discussed in the Section III.
B. Connector
A novel connection method is adopted in FreeBOT system:
a static permanent magnet is equipped inside the FreeBOT,
which makes the magnetic field not only can penetrate the
shell but also transmit to the outside; the ferromagnetic
spherical shell of FreeBOT is made of iron, so the shell is
magnetized when it approaches a magnetic field. Since the
size of the internal magnet is small, but the magnetic field
strength is large (the size of the magnet is 20mm×10mm×
10mm, and the magnetic remanence is 14700 gauss), so it
can excite a small but strong external magnetic field. Fig.
3 shows the magnetic field distribution when one FreeBOT
attracts another FreeBOT (ANSYS is adopted to analyze the
magnetic field of a FreeBOT). We can see that the internal
magnet excites a magnetic field, which penetrates through the
outer shell and transmits to the outside, and the two spherical
shells are magnetized by this internal magnet. Fig. 4 shows
the variation of magnetic force with the distance between
two FreeBOTs. When d= 0, the magnetic force reaches
its maximum value, which is 22.6N. With the increase of
distance, the magnetic force decreases exponentially.
Fig. 5 shows the attraction of the internal magnet to other
spherical shells when it moves, and the magnetic attraction
is divided into two directions, parallel and perpendicular. For
the parallel direction, the magnetic attractive force decreases
as θincreases. We can see that Akis always positive, which
Fig. 5. Magnetic attraction versus lifting angle
Fig. 6. Magnetic attraction versus lifting angle when two magnets are
close
means that as long as θ > 0, there will always be an force
point from the internal magnet to other FreeBOT. For the
perpendicular direction, as θincreases, Afirst decreases
and then increases, and the maximum value is only 1.4N.
During changing the position of the internal magnet, Akis
much larger than A. Therefore, we can choose a suitable
θto realize the reconfiguration of a FreeBOT on another
FreeBOT surface.
The above analysis shows that which small area of the
spherical iron shell will attract other FreeBOT depends on
the position of the internal magnet. However, when the two
internal magnets are close together, the case is different. Fig.
6 shows the forces of two internal magnets that are closed,
and we decompose the magnetic attraction into parallel and
perpendicular directions similarly. For the parallel direction,
when 0< θ < 8,Akis negative; when θ > 8,Ak
becomes positive. Akwill increase with the increase of
θ, and finally stabilize at 22.6N(it is also the maximum
attractive force in Fig. 4). This means that the two internal
magnets repel each other when they are very close and the
attraction of the internal magnet to other FreeBOT will return
to normal once the two internal magnets are far away. For the
perpendicular direction, Ais always negative, which means
that they are mutually exclusive. When θis small, there will
be a large repulsive force in the perpendicular direction,
but when θis large, the repulsive force will disappear. In
general, since the two internal magnets are of the same
gender, the two FreeBOT will repel each other when their
internal magnets face each other.
In summary, FreeBOT’s internal magnet can be connected
to the entire spherical shell of other FreeBOT. The location
of the internal magnet is the only blind spot for connection. It
is encouraging that this is a genderless connector since the
(a) Side view when rolling (b) Top view when turning
Fig. 7. Motion of FreeBOT
two FreeBOTs are essentially connected by their spherical
iron shell, and the connection of two FreeBOTs can be at
almost any point on their spherical iron shell. Therefore, as
long as two FreeBOTs touch each other, we can control the
internal magnet to realize their connection. Based on this
design, FreeBOT can achieve some interesting motions.
III. MOTI ON OF FR EEBOT
A. Module Independent Motion
FreeBOT is essentially a spherical robot, so the general
movement method of spherical robots is also applicable to
FreeBOT. Fig. 7(a) shows a side view of FreeBOT during
rolling. When the two driving wheels rotate in the same
direction, the internal vehicle moves along the inner surface
of the spherical shell, and then the gravity center of FreeBOT
is raised and torque is provided to make FreeBOT roll
forward. The driving torque τis given by
τ=rmsinθ =Iα, (1)
where ris the distance from the sphere center to the gravity
center of the internal body, mis the mass of the internal
driving mechanism, θis the rolling angle of the sphere, I
is the inertial moment of the spherical shell and αis the
angular acceleration of the FreeBOT.
Assuming the spherical shell material is homogeneous, the
inertial moment of the spherical shell is
I=2
5mball
R5
1R5
2
R3
1R3
2
+mball
4(R1+R2)2,(2)
where mball is the mass of the spherical shell, R1is the
external radius of the shell, and R2is the internal radius of
the shell.
The angular acceleration αis denoted as
α=rmsinθ
I=Kr m
mball
sinθ, (3)
where
K=2
5·R5
1R5
2
R3
1R3
2
+(R1+R2)2
41
.
The coefficient Kdepends on the geometry, so for a Free-
BOT with given structure and mass, the rolling velocity only
depends on the rolling angle of the internal vehicle.
Fig. 8. Connection and separation between FreeBOTs
Fig. 7(b) shows the top view of FreeBOT during turning.
When two motors rotate in different directions, the friction
from the iron shell will produce a torque around the central
axis of the internal vehicle. The internal vehicle will rotate
around the axis to change the orientation.
Bicchi [17] introduces a kinematic model for general
spherical robot. It is found that this model is also suitable to
represent the properties of the FreeBOT moving on the flat
ground. According to [17], the kinematics in our notations
and coordinate system can be formulated as:
˙x
˙y
˙
φ
˙
β
˙
ψ
˙
θ
=
cos θ
sin θ
sin (ψθ)
Rcos β
cos (ψθ)
R
tan βsin (ψθ)
R
0
u1+
0
0
0
0
0
1
u2,(4)
where (x, y, φ, β, ψ, θ)Tdenote the configuration of the
robot, parameterized by the xy location of the sphere center,
the ZYX Euler angles, φ, β, ψ, and the steering angle of
the internal driving mechanism with respect to the FreeBOT
body, θ,u1is forward speed, and u2is the steering speed.
B. Connection and separation
Fig. 8 from left to right shows the connection between
two FreeBOTs. In Fig. 8(a), FreeBOT A and FreeBOT B are
not touching each other. Next, FreeBOT A independently
roll to FreeBOT B, as shown in Fig. 8(b). At this time, the
internal vehicle of FreeBOT A is at the bottom, so FreeBOT
A and FreeBOT B touch each other but are not connected.
Following Fig. 8(c) and Fig. 8(d), FreeBOT A’s internal
vehicle can move toward the contact point between the
two FreeBOTs, and FreeBOT A’s internal magnet generates
strong magnetic attraction to the FreeBOT B’s iron shell
to achieve the connection between modules. As mentioned
above, FreeBOT A’s internal magnet will excite a magnetic
field in one area, so there is still connecting force between
the modules even if the internal vehicle is not precisely
adjusted to the contact point. FreeBOT’s connectors are fault-
tolerant, just like some advanced MSRR systems. FreeBOT
A and FreeBOT B can establish a connection from all
directions through the spherical shells, without a complex
path planning to align the connector precisely, which to some
extent surpasses the existing MSRR system.
Similarly, Fig. 8 shows the separation between FreeBOT A
and FreeBOT B from right to left. In Figure 8(d), FreeBOT
Fig. 9. Force analysis when connecting/separating
A’s internal magnet is attracting FreeBOT B. Following
Fig. 8(b) and Fig. 8(c), the internal vehicle of FreeBOT A
gradually moves to the bottom of the spherical shell and
the magnetic attraction between FreeBOT A and FreeBOT B
gradually weakens. In Fig. 8(b), FreeBOT A and FreeBOT B
are touch each other but not connected. Finally, FreeBOT A
adjusts the internal vehicle to leave independently as shown
in Fig. 8(a). Obviously, the separation between FreeBOTs is
the reverse process of the connection.
In conclusion, we only control the position of the internal
magnet to realize the connector management of FreeBOT
system, without the need for a designated actuator or mech-
anism to provide this function. However, the connection
and separation between FreeBOTs is not feasible on all
grounds. If the ground is too smooth, there is not enough
friction provided to keep the FreeBOT spherical shell still,
so the internal vehicle cannot freely adjust its position to
connect or separate other FreeBOTs. Next, we analyze the
working conditions required of FreeBOT based on Fig. 9. If
FreeBOT’s shell can maintain the force balance and moment
balance, we have
f1+A(θ) + N2=G
Ak(θ) + f2=N1
f1(Rrcos θ) + N1rsin θ+f2(Rrsin θ) = N2rcos θ
.
(5)
When the spherical shell is about to slide, f1and f2can
be represented as
f1=µ1·N1
f2=µ2·N2.(6)
Therefore, the required friction coefficient µ2can be rep-
resented as the Eq. (7), and the required friction coefficient
is different for different lifting angles θ.
µ2(θ) = Ak(θ)(R2+µ1R1+µ1R4)+(A(θ)G)R4
(A(θ)G)(R2+R3+µ1R1) + µ1Ak(θ)R3
,
(7)
(a) Case on the upper hemisphere (b) Case on the lower hemisphere
Fig. 10. Force analysis of two connected FreeBOTs
where
R1=Rrcos(θ)
R2=rsin(θ)
R3=Rrsin(θ)
R4=rcos(θ)
,
Gis the gravity of FreeBOT, A(θ)is the magnetic attraction
when the angle between the internal magnet and the contact
point is θ,Ris the radius of FreeBOT spherical shell,
ris the distance between the center of gravity and the
spherical center, and µ1is the friction coefficient between
two FreeBOTs. Comprehensively, the two FreeBOT given
above parameters can be successfully connected and sepa-
rated when the friction coefficient between the ground and
the FreeBOT is greater than µ2.
C. Reconfiguration
The combination of multiple FreeBOTs shows some excit-
ing performance. For MSRR system, we are concerned about
how to rearrange these modules to different configurations.
Different from the previous MSRR system which provides
reconfigurable function by specified motor and mechanical
design, a FreeBOT can crawl on the surface of other Free-
BOT to realize the reconfigurable function. According to
the results in Fig. 4, when the internal magnet is slightly
raised, the magnetic attraction component in both directions
is always positive, which means that will generate a moment
force for rolling. Fig. 10 shows two connected FreeBOTs,
one of which is connected to a ferromagnetic wall and
suspended in the midair. Fig. 10(a) shows two FreeBOTs
connected in the upper hemisphere, while Fig. 10(b) shows
the case in the lower hemisphere. Next, we analyze the
conditions for two FreeBOTs to be connected from multiple
angles and maintain static force balance. If the two FreeBOTs
in Fig. 10(a) can keep the static force balance, we have:
The combination of multiple FreeBOTs shows some ex-
citing performance. For MSRR system, we are concerned
about how to reconfigure these modules into different con-
figurations. Unlike the previous MSRR system that provides
reconfigurable functions through designated actuators and
mechanisms, FreeBOT can crawl on the surface of other
FreeBOTs and maintain static connections at multiple angles
to achieve reconfiguration. Fig. 10 shows two connected
FreeBOTs, one of which is connected to a ferromagnetic wall
Fig. 11. A simple reconfiguration example of FreeBOT system
and suspended in the air. Fig. 10(a) shows two FreeBOTs
connected in the upper hemisphere, while Fig. 10(b) shows
the case in the lower hemisphere. Next, we analyze the
conditions for two FreeBOTs to establish a connection and
maintain static force balance from multiple angles. If the two
FreeBOTs in Fig. 10(a) can maintain static balance, then we
have
N=Gsin θ+A
Gcos θ=f=µN .(8)
So the required friction coefficient under different connec-
tion angles can be expressed as
µ(θ) = Gcos θ
A+Gsin θ<G
A.(9)
Therefore, we can obtain the connection conditions in
the upper hemisphere, that is, the surface friction coefficient
required between FreeBOTs are
µ > G
A.(10)
Similarly, if the two FreeBOTs in Fig. 10(b) can maintain
static balance, then we have
N+Gsin θ=A
Gcos θ=f=µN .(11)
So the required friction coefficient under different connec-
tion angles can be expressed as
µ(θ) = Gcos θ
AGsin θ<
AGqA2G2
A2
A2G2.(12)
Similarly, we can obtain the connection conditions in the
lower hemisphere, that is, magnetic attraction Aand friction
coefficient µfulfill
(A>G
µ > AGqA2
G2
A2
A2G2
.(13)
Obviously, the conditions in the lower hemisphere is
more strict than that in the upper hemisphere. Therefore, as
long as the two FreeBOTs can maintain a static connection
in the lower hemisphere, connections from all angles are
available. (13) is a sufficient condition for two FreeBOTs to
be connected from multiple angles and maintain static force
balance, that is, FreeBOTs should have a rough shell and
strong internal magnets.
FreeBOT can adjust the internal vehicle to connect to
another FreeBOT in multi directions. Fig. 11 shows a simple
reconfiguration example of FreeBOT system. First, four
TABLE I
SPE CIFIC ATIO NS AN D PER FO RM AN CE S OF FR EE BOT
Specification & Performance Value
Maximum forward Speed 1.2 Body Length/s
Maximum Steering Speed 3.5 rad/s
Time to dock 0.5 seconds
Time to undock 0.5 seconds
Holding force in tension 22.6 N
Wheel Speed (No Load) 60RPM (7.4V)
Wheel Torque 7 kg·cm
Static Module Power Dissipation 0.45 W (7.4V)
Moving Module Power Dissipation 1.38 W (7.4V)
Magnetic remanence 14700 gauss
Magneti size 20×20×10 mm
Overall Dimensions 120×120×120 mm
Module Weight 307.9g
Fig. 12. Prototype of FreeBOT
FreeBOTs are connected to form a robust base. Next, a new
FreeBOT independently rolls to and connects to the base.
The newly added FreeBOT can change the position of the
internal magnet so that the new FreeBOT can be moved to
any point on the base surface to rearrange the system into
different configurations. Compared with the previous MSRR
system, the FreeBOT system has fewer physical constraints
in the rearrangement, which means more configurations
are available. In general, FreeBOT has greater potential in
developing applications for freeform MSRR systems in the
future.
IV. EXP E RI M EN TS AN D RESU LTS
Fig. 12 shows the FreeBOT prototype. Some specifications
and performance of FreeBOT are tested, and the detailed
information is shown in Table I. Numerous experiments
have also been conducted to evaluate the performance of
FreeBOT in different aspects, i.e., 1) module independent
motion, 2) connection and separation, 3) climbing stairs, and
4) 3D reconstruction. In this paper, FreeBOTs are remotely
controlled to show these.
A. Module independent motion
Essentially, as a spherical robot, FreeBOT can use the con-
trol law of the general spherical robot to realize independent
(a) (b) (c) (d) (e)
Fig. 13. A FreeBOT climbing on the ferromagnetic wall
(a) (b) (c)
(d) (e) (f)
Fig. 14. Connection and separation between FreeBOTs
motion on a plane. In addition, due to the strong internal
magnet, FreeBOT can climb up ferromagnetic slopes or even
walls. Fig. 13 shows a FreeBOT climbing a ferromagnetic
wall. After FreeBOT on the ground adjusts the internal
magnet to attract the wall, the FreeBOT can independently
move on the wall plane by controlling the internal vehicle.
B. Connection and separation
Fig. 14 shows the connection and separation between
FreeBOTs. Four FreeBOTs are connected together to form
a robust base. A new FreeBOT independently comes, and
adjusts the internal magnet to connect to the FreeBOT base.
Due to fault-tolerant and freeform connector, the time to dock
is only 0.5 seconds (the time to dock is actually the time for
the internal vehicle to move from the bottom to the FreeBOT
contact point in Fig. 8). After connection, the new FreeBOT
can move freely along the surface of the base. Similarly, the
time to undock is only 0.5 seconds (the time to undock is
actually the time for the internal vehicle in Fig. 8 to move
from the FreeBOT contact point to the bottom).
C. Climbing the stairs
Compared with conventional robotic systems, the ability
to complete tasks collaboratively is a unique aspect of
the MSRR system. Fig. 15 shows two FreeBOTs climbing
stairs cooperatively. One FreeBOT cannot climb the stairs
independently, but two FreeBOTs can cooperate with each
other to achieve this task. The first FreeBOT came to the
stairs independently and served as a ladder for the second
FreeBOT. Next, the second FreeBOT will connect to the first
FreeBOT and adjust the internal magnet to crawl along the
(a) (b)
(c) (d)
Fig. 15. Two FreeBOTs climbing the stairs cooperatively
(a) (b) (c) (d)
Fig. 16. Two FreeBOTs show a 3D reconstruction demonstration
surface of the first FreeBOT. Finally, the second FreeBOT
separates from the first FreeBOT and comes to the top of
the stairs. However, since there is only one FreeBOT as a
ladder, it will not be as robust as the base shown in Fig. 14.
Therefore, only two FreeBOTs cooperated to climb the stairs
are not 100% successful, and it can become more stable if
more FreeBOTs join.
D. 3D Reconfiguration
In [2], a SMORES module lift another module on a plastic
structure of passive docking ports to show a demonstra-
tion of self-reconfiguration in 3D. We conducted a similar
experiment between two FreeBOTs and a ferromagnetic
wall (as shown in Fig. 16). A FreeBOT is connected to
a ferromagnetic wall and suspended in the air. The tested
FreeBOT is connected to the suspended FreeBOT and can be
lifted by adjusting the internal magnet. Due to the freeform
connector, the tested FreeBOT can be lifted along many
paths without constrain, which has great potential for 3D
self-reconfiguration.
V. CO MPARIS ON BE TWE EN FRE EBOT A ND PRE VIO US
MSRR SYS TE M S
FreeBOT has the same basic functions as the most
advanced MSRR: module independent motion, connec-
tion/separation between modules without manual assistance
and system reconfiguration. However, the previous MSRR
module is equipped with multiple actuators for different
tasks, which increases the weight, volume and manufacturing
cost of the robot. FreeBOT has only two motors for these
tasks, but it can form an MSRR system with fewer physical
constraints. In addition, FreeBOT shows better performance
than previous MSRR systems in many aspects, the detailed
comparison between these MSRR systems is shown in Table
II. But it should be noted that the holding force in tension
of FreeBOT is small, which is the weakness of FreeBOT.
Although FreeBOT is not competitive in the comparison of
holding force in tension, it is sufficient for most tasks.
Multiple MSRR modules forming joints are the main self-
reconfiguration method of most MSRR systems. Fig. 17
shows the joint formed by FreeBOT and some previous
MSRR system [1], [2], [18]–[20]. In MSRR systems with
different architectures, joints composed of modules have
different characteristics. The joint composed of the previ-
ous MSRR system can only rotate around one axis, while
the FreeBOT system can form an unlimited revolute joint.
This MSRR self-reconfiguration method with less physical
constraints makes the FreeBOT system has great potential to
realize freeform robot systems and more applications.
TABLE II
PER FO RM AN CE S OF MSRR MO DU LE S
Specification
FreeBOT
SMORES
ATRON
M-TRAN III
SuperBot
M3
No. of DoF 2 4 123 3
No. of Actuators 256 5 9 6
Maximum Connecting No. 12 48 6 6 3
Ability to Move Independently YYNN N Y
No. of Mech. Parts 24 132 145 162 - -
Holding Force in Tension(N) 15.5 60 800 25 - -
Dock Cycle Time (s) 0.5 2.3 4 5 50 -
Weight (kg) 0.31 0.52 0.83 0.42 1.2 0.8
Data Derived from [2] [2], [18] [1], [2] [2], [19] [2], [20]
(a) ATRON [18] (b) M3[20] (c) MTRAN-III [1]
(d) SMORES [2] (e) SUPERBOT [19] (f) FreeBOT
Fig. 17. Joints consisting of some MSRR systems
VI. CO N CL U SI O NS A ND FUT URE WORK
This paper proposes a novel MSRR “FreeBOT”, which can
be connected freely with less physical constraints. FreeBOT
only has two motors for multiple tasks: module independent
movement, connector management and system reconfigura-
tion. Due to the fault-tolerant and freeform connector, the
connection between FreeBOTs is genderless and instant.
Numerous experiments have been conducted to test its per-
formance. The experimental results show that the FreeBOT
system has great potential to realize a freeform robotic
system.
In this paper, FreeBOT is remotely controlled to demon-
strate these experiments. Our group is researching on the
relative localization [21] and motion planning algorithm [22]
for FreeBOT system. In the future, we will equip FreeBOT
with these technologies to realize an autonomous FreeBOT
system. In addition, we will increase the number of FreeBOT
to fully demonstrate the enormous potential of FreeBOT in
realizing more MSRR applications.
REF ER E NC E S
[1] H. Kurokawa, K. Tomita, A. Kamimura, S. Kokaji, T. Hasuo, and
S. Murata, “Distributed self-reconfiguration of m-tran iii modular
robotic system,” The International Journal of Robotics Research,
vol. 27, no. 3-4, pp. 373–386, 2008.
[2] J. Davey, N. Kwok, and M. Yim, “Emulating self-reconfigurable
robots-design of the smores system,” in IEEE/RSJ International Con-
ference on Intelligent Robots and Systems, 2012, pp. 4464–4469.
[3] D. Saldana, B. Gabrich, G. Li, M. Yim, and V. Kumar, “Modquad:
The flying modular structure that self-assembles in midair,” in IEEE
International Conference on Robotics and Automation, 2018, pp. 691–
698.
[4] J. W. Romanishin, K. Gilpin, and D. Rus, “M-blocks: Momentum-
driven, magnetic modular robots,” in IEEE/RSJ International Confer-
ence on Intelligent Robots and Systems, 2013, pp. 4288–4295.
[5] A. Spr¨
owitz, S. Pouya, S. Bonardi, J. Van Den Kieboom, R. M¨
ockel,
A. Billard, P. Dillenbourg, and A. J. Ijspeert, “Roombots: reconfig-
urable robots for adaptive furniture,IEEE Computational Intelligence
Magazine, vol. 5, no. 3, pp. 20–32, 2010.
[6] J. Seo, J. Paik, and M. Yim, “Modular reconfigurable robotics,Annual
Review of Control, Robotics, and Autonomous Systems, vol. 2, pp. 63–
88, 2019.
[7] S. Chennareddy, A. Agrawal, and A. Karuppiah, “Modular self-
reconfigurable robotic systems: a survey on hardware architectures,
Journal of Robotics, 2017.
[8] C. Liu and M. Yim, “Reconfiguration motion planning for variable
topology truss.” in IEEE/RSJ International Conference on Intelligent
Robots and Systems, 2019, pp. 1941–1948.
[9] C. Liu, S. Yu, and M. Yim, “A fast configuration space algorithm
for variable topology truss modular robots,” in IEEE International
Conference on Robotics and Automation, 2020.
[10] H. Wei, Y. Chen, J. Tan, and T. Wang, “Sambot: A self-assembly
modular robot system,” IEEE/ASME Transactions on Mechatronics,
vol. 16, no. 4, pp. 745–757, 2010.
[11] A. Kamimura, H. Kurokawa, E. Yoshida, K. Tomita, S. Kokaji, and
S. Murata, “Distributed adaptive locomotion by a modular robotic sys-
tem, m-tran ii,” in IEEE/RSJ International Conference on Intelligent
Robots and Systems, vol. 3, 2004, pp. 2370–2377.
[12] J. Liedke and H. W¨
orn, “Cobold-a bonding mechanism for modular
self-reconfigurable mobile robots,” in IEEE International Conference
on Robotics and Biomimetics, 2011, pp. 2025–2030.
[13] K. Kotay, D. Rus, M. Vona, and C. McGray, “The self-reconfiguring
robotic molecule,” in IEEE International Conference on Robotics and
Automation, vol. 1, 1998, pp. 424–431.
[14] T. Tosun, J. Davey, C. Liu, and M. Yim, “Design and characterization
of the ep-face connector,” in IEEE/RSJ International Conference on
Intelligent Robots and Systems, 2016, pp. 45–51.
[15] N. Eckenstein and M. Yim, “Area of acceptance for 3d self-aligning
robotic connectors: Concepts, metrics, and designs,” in IEEE Interna-
tional Conference on Robotics and Automation, 2014, pp. 1227–1233.
[16] J. W. Romanishin, K. Gilpin, S. Claici, and D. Rus, “3d m-blocks:
Self-reconfiguring robots capable of locomotion via pivoting in three
dimensions,” in IEEE International Conference on Robotics and
Automation, 2015, pp. 1925–1932.
[17] A. Bicchi, A. Balluchi, D. Prattichizzo, and A. Gorelli, “Introducing
the” sphericle”: an experimental testbed for research and teaching in
nonholonomy,” in IEEE International Conference on Robotics and
Automation, vol. 3, 1997, pp. 2620–2625.
[18] M. W. Jorgensen, E. H. Ostergaard, and H. H. Lund, “Modular atron:
Modules for a self-reconfigurable robot,” in IEEE/RSJ International
Conference on Intelligent Robots and Systems, vol. 2, 2004, pp. 2068–
2073.
[19] B. Salemi, M. Moll, and W.-M. Shen, “Superbot: A deployable,
multi-functional, and modular self-reconfigurable robotic system,” in
IEEE/RSJ International Conference on Intelligent Robots and Systems,
2006, pp. 3636–3641.
[20] M. D. Kutzer, M. S. Moses, C. Y. Brown, M. Armand, D. H.
Scheidt, and G. S. Chirikjian, “Design of a new independently-mobile
reconfigurable modular robot,” in IEEE International Conference on
Robotics and Automation, 2010, pp. 2758–2764.
[21] M. Li, G. Liang, H. Luo, H. Qian, and T. L. Lam, “Robot-to-robot
relative pose estimation based on semidefinite relaxation optimization,
in IEEE/RSJ International Conference on Intelligent Robots and
Systems, 2020.
[22] H. Luo, M. Li, G. Liang, H. Qian, and T. L. Lam, “An obstacles-
crossing strategy based on the fast self-reconfiguration for modular
sphere robots,” in IEEE/RSJ International Conference on Intelligent
Robots and Systems, 2020.
... The FreeBOT is used to implement the motions of the two-DOF SRC joint and manipulator [27], where a driving trolley moves forward and steers inside the spherical iron shell to drive the joint motions, and a magnet at trolley's bottom attracts the other body not to separate. This article also develops a physics simulation system for the proposed SRC joint and manipulator and builds a real SRC joint manipulator using FreeBOTs, such that the two-DOF SRC joint model, manipulator kinematics, and control method can be validated. ...
... Similarly, (27) can be derived and results in ...
... The drive of the SRC joint can also be novelly physically implemented. Unlike conventional robotic manipulator that use combinations of revolute and prismatic joints implemented by dc motors, the realization of the SRC joint could be using the FreeBOT for example [27], as shown in detail in Section IV, where a driving trolley moves inside the spherical iron body and uses a magnet at trolley's bottom to attract the other body not separated. In turn, performing chain-type motion is common for a modular self-reconfigurable robot (MSRR) robot (like FreeBOT) in self-reconfiguration tasks [32], the established SRC joint model and manipulator kinematics are also useful for describing the reconfiguration motions of the FreeBOT. ...
Article
Full-text available
Rolling contact joints are attracting increasing interest in applications to robotic fingers and manipulators, due to the potential of the absence of abrasion wear, the simplification of the controller, and the enlargement of reachable configurations. This article first proposes a novel two-degree-of-freedom (DOF) spherical rolling contact (SRC) joint, with the joint model elements being formulated, including rotation matrix, position vector, and free modes, as with those of classic joints. As an application, a new kind of serial manipulator formed by the SRC joints is presented, and its forward and inverse kinematics are modeled. The motions of the two-DOF SRC joint and manipulator are implemented using the FreeBOT, and the control method is proposed for the FreeBOT, such that the SRC joint and manipulator realize the motion control. The kinematics and control of the two-DOF SRC joint and manipulator are validated using physics simulations and on a real manipulator formed by FreeBOTs.
... However, most are limited to 2D plane movement [12,13,14]. To our knowledge, only four modular robots have demonstrated the ability of 3D continuous connection, namely, FireAnt3D [15], FreeBot [16], FreeSN [17] and SnailBot [18]. FireAnt3D requires large electric power consumption, has short service life and is time-consuming. ...
Chapter
Full-text available
In the paper, a novel modular self-assembling and self-reconfiguring robot named Ant3DBot was proposed, which has many configurations. Ant3DBot consists of four semicircular iron spheroid shells, telescopic legs, and internal magnets that can rotate around the center. Ant3DBot can expand its shells and legs through a single motor, a synchronous belt and compressed springs, which results in two different docking states. Ant3DBot which has the height of 12 cm can traverse obstacles with the height of 8 cm, and pass through a 25° slope in extending configuration. For many unstructured environments, the cooperation of multiple Ant3DBots can reach a target point with simple control. The simulations show the basic ability of a single module to overcome obstacles as well as the cooperative motion of multiple robots. The results demonstrate that the Ant3DBot system has excellent locomotion performance and versatility.
Chapter
We introduce Calmbots, insect-based interfaces comprising multiple functions (transportation, display, drawing, or haptics) for use in human living spaces by taking advantage of insects’ capabilities. We utilized Madagascar hissing cockroaches as robots because of advantages such as mobility, strength, hiding, and self-sustaining abilities. Madagascar hissing cockroaches, for instance, can be controlled to move on uneven cable-lines floors and push light-weight objects such as tablespoon. We controlled the cockroaches’ movement using electrical stimulation and developed a system for tracking and communicating with their backpacks using augmented reality markers and a radio-based station, the steps of controlling multiple cockroaches for reaching their goals and transporting objects, and customized optional parts. Our method demonstrated effective control over a group of three or five cockroaches, with, at least, 60% success accuracy in dedicated experimental environments involving over forty trials for each test. Calmbots could move on carpeted or cable-lines floor and did not become desensitized to stimulation under a certain break interval. KeywordsInterfaces of insectsControl of multiple insects
Conference Paper
Full-text available
In this paper, the 2D robot-to-robot relative pose (position and orientation) estimation problem based on ego-motion and noisy distance measurements is considered. We address this problem using an optimization-based method, which does not require complicated numerical analysis while yields no inferior relative localization (RL) results compared to existing approaches. In particular, we start from a state-of-the-art method named square distances weighted least square (SD-WLS), and reformulate it as a non-convex quadratically constrained quadratic programming (QCQP) problem. To handle its non-convex nature, a semidefinite programming (SDP) relaxation optimization-based method is proposed, and we prove that the relaxation is tight when measurements are free from noise or just corrupted by small noise. Further, to obtain the optimal solution of the relative pose estimation problem in the sense of maximum likelihood estimation (MLE), a theoretically optimal WLS method is developed to refine the estimate from the SDP optimization. Comprehensive simulations and well-designed experiments are presented for validating the tightness of the SDP relaxation, and the effectiveness of the proposed algorithm is highlighted by comparing it to the existing approaches. I. INTRODUCTION Relative localization (RL), which refers to determine relative configurations of mobile agents with respect to (w.r.t.) other agents or landmark, is critically important in multi-robot systems since it is the prerequisite for robot teaming and swarming [1]-[3]. Many applications, such as target tracking [4], localization [5] and mapping [6] are often faced with RL problem. In the case of static multi-robot systems, a variety of algorithms based on convex optimization [7], sum of squares relaxation [8], multidimensional scaling [9] and flying anchor nodes approach, were employed to address the RL problem, where these methods can precisely locate the position of sensor nodes, meantime, fast and robust. However, w.r.t. mobile multi-robot systems, except for position , the relative orientation also has to be determined, and this is emphasized by plenty of literature [10]-[12]. Among the vast investigations, however, the relative pose estimation problem is usually supposed to be solved, and only a few works describe how to locate the relative pose in multi-robot systems. Manually measuring the relative pose between robots is simple and straightforward, but tedious, with low accuracy and not applicable to large robot teams. Alternatively, utilizing external references to achieve RL, such as global positioning system (GPS), compass, maps of the environment
Conference Paper
Full-text available
This paper introduces an obstacle-crossing strategy , and the self-reconfiguration algorithm for a new class of modular robots called the rolling sphere, which can fit obstacles represented by cubes of different sizes due to the chain connection of multiple spheres. For the self-reconfiguration of the rolling spheres, a large gradient is obtained by classifying its action types and hierarchically minimizing the distance between the initial configuration and the final configuration. The most direct use of this large gradient is the fast crossing of various obstacles, by jointing multiple self-reconfigurations according to the OctoMap of the obstacles. It is verified in simulation that the self-reconfiguration takes full advantage of the parallel movement of multiple modules to reduce the total time steps, and the obstacle-crossing strategy can adapt to a variety of obstacles.
Conference Paper
Full-text available
We introduce ModQuad, a novel flying modular robotic structure that is able to self-assemble in midair and cooperatively fly. The structure is composed by agile flying modules that can easily move in a three dimensional environment. The module is based on a quadrotor platform within a cuboid frame which allows it to attach to other modules by matching vertical faces. Using this mechanism, a ModQuad swarm is able to rapidly assemble flying structures in midair using the robot bodies as building units. In this paper, we focus on two important tasks for modular flying structures. First, we propose a decentralized modular attitude controller to allow a team of physically connected modules to fly cooperatively. Second, we develop a docking method that drives pairs of structures to be attached in midair. Our method precisely aligns, and corrects motion errors during the docking process. In our experiments, we tested and analyzed the performance of the cooperative flying method for multiple configurations. We also tested the docking method with successful results.
Article
Full-text available
Modular self-reconfigurable robots present wide and unique solutions for growing demands in the domains of space exploration, automation, consumer products, and so forth. The higher utilization factor and self-healing capabilities are most demanded traits in robotics for real world applications and modular robotics offer better solutions in these perspectives in relation to traditional robotics. The researchers in robotics domain identified various applications and prototyped numerous robotic models while addressing constraints such as homogeneity, reconfigurability, form factor, and power consumption. The diversified nature of various modular robotic solutions proposed for real world applications and utilization of different sensor and actuator interfacing techniques along with physical model optimizations presents implicit challenges to researchers while identifying and visualizing the merits/demerits of various approaches to a solution. This paper attempts to simplify the comparison of various hardware prototypes by providing a brief study on hardware architectures of modular robots capable of self-healing and reconfiguration along with design techniques adopted in modeling robots, interfacing technologies, and so forth over the past 25 years.
Article
This article reviews the current state of the art in the development of modular reconfigurable robot (MRR) systems and suggests promising future research directions. A wide variety of MRR systems have been presented to date, and these robots promise to be versatile, robust, and low cost compared with other conventional robot systems. MRR systems thus have the potential to outperform traditional systems with a fixed morphology when carrying out tasks that require a high level of flexibility. We begin by introducing the taxonomy of MRRs based on their hardware architecture. We then examine recent progress in the hardware and the software technologies for MRRs, along with remaining technical issues. We conclude with a discussion of open challenges and future research directions. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems Volume 2 is May 3, 2019. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
Conference Paper
Alignment of module connectors is a crucial component of self-reconfiguration in modular robotics. Accomplishing this process using passive mechanical geometry saves resources such as space and power for the modular robot. We present concepts for evaluation of these geometries as well as a new 3D geometry, the 3D X-Face. For comparing different connectors independent of the rest of the robot, figures of merit are presented which are based on the ability for connectors to mate in the presence of position and orientation errors (offsets). Figures of merit for many current connectors are presented. The 3D X-Face alignment behavior is simulated in Gazebo over several sets of initial conditions to estimate the full area of acceptance, and the connector is tested on a CKBot robotic platform. For the situation without rotation, results indicate a 27% improvement over current gendered connectors and a 467% over ungendered connectors. The 3D X-Face is further simulated over a full five-dimensional set and metrics are estimated on that set.
Article
This paper presents the mechanical design of a modular robot called the 3D M-Block, a 50mm cubic module capable of both independent and lattice-based locomotion. The first M-Blocks described in [1] could pivot about one axis of rotation only. In contrast, the 3D M-blocks can exert on demand both forward and backward torques about three orthogonal axes, for a total of six directions. The 3D M-Blocks transform these torques into pivoting motions which allow the new 3D M-Blocks to move more freely than their predecessors. Individual modules can employ pivoting motions to independently roll across a wide variety of surfaces as well as to join and move relative to other M-Blocks as part of a larger collective structure. The 3D M-Block maintains the same form factor and magnetic bonding system as the one-dimensional M-Blocks [1], but a new fabrication process supports more efficient and precise production. The 3D M-blocks provide a robust and capable modular self-reconfigurable robotic platform able to support swarm robot applications through individual module capabilities and self-reconfiguring robot applications using connected lattices of modules.