Highlights

Study and simulation of an under-actuated smart surface for material flow handling

d
Edoardo Bianchi, Olier Jonas Jorg, Gualtiero Fantoni, Francisco Javier Brosed Dueso, José A.
Yagüe-Fabra

we
• Novel under-actuated surface for material flow handling.

• Simulation environment of the under-actuated surface.

ie
• Simulation of Sorting, Slowing and Stopping tasks.

• Validation of the proposed simulation environment.

r ev
er
pe
ot
tn
rin
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 Study and simulation of an under-actuated smart surface for
material flow handling

d
Edoardo Bianchia,∗, Olier Jonas Jorgb , Gualtiero Fantonib , Francisco Javier Brosed Duesoa ,

we
José A. Yagüe-Fabraa
a University of Zaragoza, Department of Design and Manufacturing Engineering, C/ Maria de Luna
3, Zaragoza, 50018, Spain,
b University of Pisa, Department of Civil and Industrial Engineering, Largo Lucio Lazzarino, Pisa, 56122, Italy,

ie
Abstract

ev
Smart surfaces are becoming more and more popular in the field of intralogistics, as they
combine great flexibility with easy reprogrammability. Pursuing this trend, the following
article proposes a modular surface to perform handling tasks, such as sorting, stopping or
slowing down material flows. The surface used is under-actuated, thus, it exploits the speed,

r
already possessed by the object, or the gravity to perform, with a simplified hardware, the
above mentioned tasks. In practice, these handling actions are completed using an array
er
of rotors, of which only the direction of the rotation axis is controlled. Moreover, the axis
can only assume certain discrete orientations in the plane, further simplifying the design.
Thus, what is created is a controllable and under-actuated friction field to manipulate the
material flow. In the article, the analytic model of the surface is described and a software
pe
simulation environment is introduced to demonstrate its functioning. In addition, examples
of sorting, slowing down and stopping operations and a validation of the simulation itself
are presented.
Keywords: Smart surface, friction force field, under-actuation, feeding, simulation, material
flow handling.
ot

1. Introduction
tn

The material transport inside factories and warehouses is a very studied topic. How-
ever, companies are always searching for new cheaper solutions while keeping an eye on
efficiency and flexibility at the same time [1, 2, 3, 4]. Starting with the most basic conveyors,
many alternative systems [5, 6, 7, 8, 9, 10, 11, 12] have been developed in order to move, sort,
rin

orient and handle the material flow. More recently the focus of the investigation shifted to-
wards modular devices and surfaces, which allow high handling capability, adaptability and
reconfigurability [3, 4, 13]. This happens because the market increasingly demands a flexible
industry and this is reflected in all production levels, even the internal transportation. These
ep

devices are often called smart surfaces because they are controllable and reprogrammable
with computers. Thanks to these characteristics, these new systems permit to achieve their
goals without structural changes [4, 13]. In addition, the new devices are capable to iden-
tify a part with sensors and to act in diverse ways, according with their programming. As
a result, supervision by an operator is no longer required for the decision process and the
Pr

system becomes autonomous, promising to reduce errors and costs [14].

∗ Corresponding author
Email address: ebianchi@unizar.es (Edoardo Bianchi)

Preprint submitted to Mechanism and Machine Theory October 17, 2022

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 The most relevant solutions among the smart surfaces, according to the current state
of the art and classified by the operating principle, are: Micro Electro-Mechanical systems,
Vibrating surfaces, Ciliary motion, Variable morphology surfaces, Pneumatic surfaces, Sur-

d
faces with rotors and Mobile platforms. Micro Electro-Mechanical Systems (MEMS) [6, 15,
16] are an array of microscopic cantilevers or tilting planes, actuated electrically, that gener-

we
ate forces to transport the object in contact. The second category of the list are Vibrating sur-
faces. These consist of vibrating plates on which a sequence of supply frequencies is applied
to generate a two dimensional force field for the handling tasks [5, 17]. In [12, 18, 19, 20],
Ciliary motion is proposed to move and manipulate objects, taking advantage of an array
of controllable cilia for contact conveyance. Another class are Variable morphology sur-

ie
faces [9, 21, 22, 23]. For these devices, gravity together with inclined planes and vertical
actuators at different altitudes are used to create a preferential path for the movement and
rotation of an object. Additionally, Pneumatic surfaces [7, 24, 25, 26, 27, 28, 29, 30] were

ev
introduced to handle materials without direct contact. The working principle for the latter
is to have modules with nozzles to move parts by directing the air flow below them. On
the other hand the Surfaces with rotors take advantage of contact forces made by moving
wheels to manipulate the material flow [3, 11, 13, 14, 31, 32, 33, 34, 35]. Lastly, the Mobile

r
platforms [8, 36, 37, 38, 39] consist of mobile pallets, not connected to fixed axes and free to
move on a plane, transporting objects on top. Despite of the categorization, not all of these
are practically used for macroscopic intralogistics purposes (object size and displacement >
er
5 4 3 2
cm). In fact, as summarized in Fig. 1, some of them, like MEMS, together with some mobile
platforms [8, 37, 38, 39] and Ciliary devices [12, 18, 19] are more suitable for microscopic
transport (object size and displacement < mm). On the contrary, the pneumatic systems,
pe
the vibrating surfaces, the surfaces with rotors, the variable morphology surfaces, some mo-
bile platforms [36] and even tilted brushes (Cilia) [20] can be used to move macroscopic and
mesoscopic (object size and displacement > mm, < cm) objects.

MEMS Vibrating surfaces Ciliary motion Mobile platforms

ot
tn

Micro Meso/Macro Micro/Meso Micro/Meso/Macro

Pneumatic surfaces Variable morphology Surface with rotors

rin
ep

Air

Meso/Macro Meso/Macro Meso/Macro

Figure 1: Smart surfaces classification and field of application.

Pr

Referring to the previous classification, the system proposed in this paper can be in-
cluded in the surfaces with rotors category. This class usually takes advantage of motorized
and orientable wheels, which can be one [14, 34] or more per module [3, 13, 35], in order
to define the surface. In general, all the concepts of this category use at least one motor

2
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 per wheel to move, sort and manipulate the material flow, creating a totally actuated sys-
tem with significant handling skills, but also increasing the complexity of the control sys-
tem and the costs. On the other hand, according to the current state of the art, a simpler

d
under-actuated device without motors is missing, for the same intra-logistic tasks. For that
reason, seeking cost reduction and simplification of both control and design, without losing

we
the flexibility of this class of systems, in this paper the authors propose an under-actuated
surface composed of modules. Unlike existing concepts, each module of contains an idle
rotor instead of a motorized wheel. In addition, the axis of rotation of the mentioned ro-
tor can be oriented in a limited number of directions in the plane of the surface, to create a
direction-able friction force for object handling. In order to compensate the under-actuation,

ie
the material is moved by exploiting gravity or an initial velocity of the object provided by
another system (e.g. a previous conveyor belt). Summarizing, compared to the existing sys-
tems [3, 11, 13, 14, 31, 32, 33, 34, 35] the two main characteristics that distinguish this active

ev
surface are its under-actuation and the limited directions of its rotor axis. Against potential
assumptions of performance losses, the authors proved through simulations, that the same
goals [3, 4, 35] of sorting and handling can be achieved with a minimal design architecture,
consisting of little components, while decreasing costs and saving energy.

r
The focus of this article lies on the study of the surface, the description of the working
principle and its simulation with the purpose of providing an initial proof of concept for
possible applications and future developments. The simulations presented in this article are
er
carried out with a customized code developed by the authors using the software Matlab. In
addition, a validation of the simulation results was obtained with the well-known program
for multi-body dynamic simulations Hexagon D&E Adams MSC. The authors decided to de-
pe
velop their own simulation environment because of its significant lower computing time.
Thus their software is eligible to control the actual physical system in real time in the future.
The remaining paper is organized as follows: Section 2 describes the concept of the sur-
face. Section 3 explains the analytic model used to describe the working principle. Then,
Section 4 describes the simulations of the system for different tasks. Finally, Section 5 re-
ports the results, Section 6 consists in the validation of the latter and Section 7 describes the
ot

conclusions.

2. Concept Description
tn

The operating principle of the surface is based on an array of modules with a rolling
element with an orientable axis of rotation. These axes are used to generate directionable
friction forces on the body in contact with the surface. The origin of the friction forces lies in
the very working principle of a rotor. In fact, these are constructed and used (e.g. conveyor
rin

rollers) to facilitate motion in one direction (perpendicular to their axis of rotation). This
leads to a difference in the friction forces exchanged with an overlying object. In practice
there will be a smaller component, perpendicular to the axis, caused mainly by rotational
friction and a larger one, parallel to the axis, due to linear friction (Fperp and Fpar respectively
ep

in Fig. 2 (c)). Therefore, the main friction component will be predominantly along the axis,
so by orienting the latter, the main force will also be directed.
Pr

3
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 6 5 4 3 2

Surface y

d 90
Object

we
transported x
Module
y
2 1 0
x

ie
(a)
Rotor axis

ev
y y D N
90

45

x x

r
Contact point Fperp
Rotor

y
0
y
er Fpar

x x Rotor
pe
-45

Rotor axis Rotor axis

C

N (b) (c)

Figure 2: Illustration of: the modular active surface (a), the four possible orientations of the rotors (b) and the
ot

three contact forces between the object and the rotor (c).

Progettato da Controllato da Approvato da Data

The choices of the axis directions are limited to a fixed number. For example, the simplest
ebianchi
Fperpconfiguration is with two orientations (0, 90°), while more complex are with four (0º, −45º,
tn

45º, 90º) (Fig. 2 (b)) or more. In Fig. 2 (b) the four orientations are shown, with the rotors
Disegno1
displayed like cylinders (the projection is a rectangle) and the axes like dashed lines. For
6 5 4 3 2
applications like slowing or stopping the material flow two orientations could be enough,
while for sorting four orientations should be taken into account. Therefore, modules with
rin

four directions will be considered in the following pages. B

Rotor
For its intended applications, the module must be used together with others to create the
surface (Fig. 2 (a)), so the sum of the forces exerted by every unit will control the objects for
Rotor sorting
axis and feeding purposes. In Fig. 2 (a) the surface set-up is summarized, showing the
ep

grid of modules with the object on top, while the arrows inside each cell represent directions
of motion promoted by them.
As mentioned in the introduction section, the system is under-actuated, so the surface
can only slow down the object while it completes its task of sorting or feeding. In fact, if, for
Pr

example, a body is simply placed on the surface without external actuation it will not move.
To overcome this situation, the surface plane can be tilted to use the gravity effect as the
missing actuation or an initial velocity can be given to the part before crossing the surface.
Data Data
ato da Controllato da
Depending Approvato da
on the application and the task of the systemA one solution may be preferred
chi 09/05/2022
over the other, or a combination of both may be chosen. For instance, if the extension of

Edizione Foglio
Disegno1 4 1/1
2 has not been peer reviewed. Electronic
This preprint research paper 1 copy available at: https://ssrn.com/abstract=4255407
 the surface is big, because the objective is to orient a body, the tilted solution could be better,
while, if the goal is to sort parts in a conveying line, the initial velocity of the object, given by
the conveyors before the sorting area, is enough. In addition, like for other active surfaces

d
that use rotors for intralogistic purposes, the application is limited to parts with at least one
planar surface, to have simultaneous contact with three or more rotors. Another limit will be

we
the maximum load per module, but this is related to the resistance of the device which is not
studied in this article. However, the study conducted is done considering the dimensions
of the modules and their resistance comparable with the similar actuated existing systems.
To provide a reference, the base of a transported object has the minimum dimensions of
b × h = [20 cm×20 cm] and the weight bearable by a single module is 20 kg.

ie
6 5 4 3
3. Analytical
6 Model 5 4 3
As stated in Section 2, the working principle of the module is based on the contact forces

ev
exchanged between the rotor and the transported object. With this in mind, the analysis
D starts from the actions on a single module and then expands to the whole surface and to the
D equations describing the motion of an object. This section reports the hypothesis, the forces
involved and the model applied to determine these.

r
er yg
View A
Z
yY g View A
View A
Surface Z Y View A
Object X Velo
pe
Surface on
Veloc
xg

Object X
on th
C Z
xg

X G
C YZ mgSin( Friction force
zg

G G rotation dir
X G' Friction force in
Y mgSin(
ot
zg

G rotation direc
mgCos( G' Rotor g
mg
g Friction fo
mgCos( Rotor in the axis direc
mg
Friction for
tn

(a) (b) in the axis directio

Figure 3: Schematic illustration of the object on the surface: lateral view(a), upper view(b).

Fig. 3 introduces the notation used in the next pages. In particular, [ x g , y g , z g ] are the
rin

B
coordinates of the object’s center of mass, G, according to the reference frame, while G’ is the
B projection of G on the object base along the Z axis (Fig. 3 (a)).

Assumptions
ep

1. The first assumption states that there are only the three forces for each interaction rotor-
object: two components in the plane of the device and one normal to this (Fig. 2 (c)).
Considering that the contact area of the module below the object is small compared to
the object itself, the previous assertion is plausible and it is translated into the possi-
bility of neglecting the exchanged torques.
Pr

2. The second assumption is about the normal component N. In fact, considering that the
contact in most of the cases is on more than three rotors, the constraint is hyperstatic
A Progettato da
and the equilibrium equations of the body are not enough to determine the normal re- ebianchi
A actions. In particular, for determining the distribution of the forces Ni an optimizationProgettato da C

ebianchi
algorithm described in Equation (1) is used.

6 5 5 4 3
This preprint6research paper has not been peer
5 reviewed. Electronic copy available
4 at: https://ssrn.com/abstract=4255407
3
 (
Aeq · X = beq
min X f ( X ) such that (1)
lb ≤ X ≤ ub

d
Where, the first vector equation represents the equilibrium in the vertical direction and

we
around the two axes of rotation in the plane, while the second vector equation states
the boundary conditions. Instead, the function to be minimized f ( X ) is the Euclidean
norm of the Ni array. The single terms are:

   
1 1 ... 1 mg cos(γ)

ie
Aeq (3 × n) =  yG′ P1 yG′ P2 ... yG′ Pn  , beq (3 × 1) =  −mÿz g ,
− xG′ P1 − xG′ P2 ... − xG′ Pn m ẍz g − mgz g sin(γ)

ev
 
N1    
 ...  0 mg cos(γ)
  0 mg cos(γ)
X ( n × 1) = 
 Ni  ,
 l b ( n × 1) = 
... ,
 u b ( n × 1) =  , f (X) = ∥X∥
 ... 

r
 ... 
0 mg cos(γ)
Nn
er
Where n is the number of rotors under the object, m is the mass of the object, g is the
gravity acceleration and γ is the inclination of the surface compared to the ground
(Fig. 3 (a)). In addition, xG′ Pi and yG′ Pi are the x and y components of the vector that
connect G’ to the ith contact point.
pe
3. The third assumption asserts that the inertia of the rotors is negligible, as they are small
and made of lightweight material. Instead, the rotational friction given by the motion
of the rotor is included in the calculation of contact forces.
4. The fourth assumption states that, for an object in motion on the rotors, the friction
conditions at the contact points (rotors-object) are as follows: Dynamic friction (slid-
ot

ing) in the direction of the axis of the rotors and pure rolling in the perpendicular
direction. This is plausible, as the center of instantaneous rotation (CIR) of the rigid
body (object) is very unlikely to be located at the center of an underlying rotor. This
tn

would guarantee a different static friction and force modeling (Fpar , Fperp ). However,
with only a few discrete points as supports, it is unlikely that the CIR is located on
these. In practice, this assumption, introducing Coulomb friction model, results in the
boundary condition F = µN for the calculation of friction forces.
rin

Friction forces model

Considering the previous assumptions, regarding sliding and pure rolling, and taking
into account that the friction forces oppose motion, the directions of the two forces (Fpar ,
Fperp ) and their final formulation can be determined. These directions are related only to
ep

the object velocity − →

v p = [ ẋ p , ẏ p ] in the contact point with the rotor (thanks to the third
assumption), an example is shown in Fig. 4 (a).
Pr

6
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
ew A

Velocity of the object

on the contact point
y Vpar

d
we
Friction force in the
Vp
rotation direction
Fperp

ie
x
5 5 4 4 3 3 2 2
Vperp

ev
Rotor axis
Fpar
Friction force

r
in the axis direction
er (a)

P P
G' G'
pe
Fy2 Fy2
P vp P vp Fx2 FFx4
x2 Fx4
G' G' Fy4 Fy4
g vgg vg Fy1 G' Fy1 G'
y y y y Fx1 Fx3Fx1 Fx3
Fy3 Fy3
ot

Object base
Object base Object base
Object base
z xz x
Point=Rotor
Point=Rotor z xz x
Point=Rotor
Point=Rotor
tn

(b) (c)

Figure 4: Schematic representations of: velocities and forces exchanged in a contact point (a), the calculation
of the velocity in the contact point (b), the resulting forces from the contacts (c).

Progettato da Controllato da Approvato da Data Data

In particular, the velocity −→
rin

v p can be split into a perpendicular and a parallel component

20/05/2022
ebianchi
(Fig. 4 (a)) and the two forces Fpar and Fperp are in the opposite directions of these velocities.
In Fig. 4 (a) three more angles are introduced: ε, β and α (all the angles are defined positive
in the anti-clockwise direction from the x axis).
Edizione Foglio
• ε is the angle between the rotor axis and the Disegno2_gamma
ep

x axis. 1/1
• β3is the angle between the velocity of the object’s
2 contact point and the x axis. 1
• α is obtained by the difference between the previous two (α = ε − β), so it is the angle
Pr

between the velocity of the body’s contact point andProgettato

the rotorda axis.
Progettato
Controllato
da da Controllato
Approvatodada Data
Approvato da Data

ebianchi ebianchi
The velocity of the object’s contact point is determined using the formula of the kinematic
−
→ −→ −
→
of rigid body: −→
vp = − →v g + θ̇ g × G ′ P. Where −→
v g = [ ẋ g , ẏ g ] and θ̇ g are the linear and Disegno3
Disegno3
−→
5 velocity
angular 3 while G ′ P is the 2vector 2
5 of the object (in4compact4 notation: Vg = [ ẋ3g , ẏ g , θ̇ g ]),

7
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 that connects the G’ to the contact point P. The velocity identification is also reported in
Fig. 4 (b), where the piece is schematized with the shape of the base, while the rotors with
a grid of points. Now all the elements to describe the two forces in the plane of the surface

d
(Fpar , Fperp ) have been introduced and the result, according with Fig. 4 (a), is summarized in
Equations (2a) and (2b).

we
Fpari = µmax N ∗ sgn(cos(α)) (2a)
Fperpi = µmin N ∗ sgn(sin(α)) (2b)
Where the two friction coefficients are respectively:

ie
• µmax is the friction coefficient between the object and the rotor.

• µmin is the friction coefficient that represents the rolling resistance of the rotor.

ev
The sign function (sgn), models the direction according with the orientation of the object
velocity in the contact point. The perpendicular force would be ideally 0, however it is
generally small as inertia forces are neglected and only the rolling friction of the rotor is
considered. The parallel friction force can reach significantly higher values and is opposed

r
(in the axis direction) to the object velocity. Therefore, it is the dominant driver to manipulate
the object. er
Equilibrium equations
At this point the three force components for each contact point below the object can
pe
be calculated and these can be summed to obtain the total reactions on the part and its
equilibrium equations. To achieve this, firstly, the friction forces in the plane at the various
contact points are decomposed along the x and y directions, knowing the arrangements of
the rotors (ε i ):

Fxi = − Fpari cos(ε i ) − Fperpi sin(ε i ) (3a)

ot

Fyi = − Fpari sin(ε i ) + Fperpi cos(ε i ) (3b)

The signs introduced for these forces are in accordance with Fig. 4 (a) and Fig. 4 (c). The
tn

resulting equilibrium equations are summarized in Equation (4).

n


 ẍ g = m1 ((∑i=i 1 Fxi ) + mg sin(γ))





n
ÿ g = m1 (∑i=i 1 Fyi )
rin

(4)





θ̈ = 1 ( ni (( x − x ) F − (y − y ) F ))
J ∑ i =1

g Pi g yi Pi g xi

Where ẍ g , ÿ g and θ̈ g are respectively the two linear and the angular accelerations of the
ep

object, while x pi and y pi are the coordinates of the ith point of contact and J is the moment
of inertia of the object around the axis perpendicular to the surface plane. Considering that
also J can be expressed as a function of the mass, it is possible to notice that the inertia affects
the system only from a structural point of view. In fact, as it is visible from the Equations
Pr

(1), (2a), (2b) and (4), m can be simplified.

8
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 4. Simulations
At this point, the illustrated analytic model was implemented in Matlab, creating the

d
software counterpart of the under-actuated system. As a first step, this simulation environ-
ment permitted to verify in a general way that moving objects with the proposed theory
and concept is possible. Secondly, it allowed several tests to be carried out in order to prove

we
the usability of the system for some common intralogistics applications, such as sorting,
orienting, stopping and slowing down material flows.

Surface and object

ie
initial data

Loop

ev
𝑥𝑥𝑔𝑔̇ ≤ 0𝑚𝑚/𝑠𝑠, 𝛾𝛾 ≠ 0° ? Rotors in contact
yes no

𝑥𝑥𝑔𝑔̇ = 0.0001𝑚𝑚/𝑠𝑠 𝑥𝑥𝑔𝑔̇ ≤ 0𝑚𝑚/𝑠𝑠, 𝛾𝛾 = 0° ? Friction forces

no yes

r
𝑛𝑛𝑖𝑖
Object equilibrium
� 𝐹𝐹𝑥𝑥𝑖𝑖 + 𝑚𝑚𝑚𝑚 sin 𝛾𝛾 > 0? er
𝑖𝑖=1 Acceleration G
no yes
𝑛𝑛𝑖𝑖 Integration
� 𝐹𝐹𝑥𝑥𝑖𝑖 + 𝑚𝑚𝑚𝑚 sin 𝛾𝛾 = 0
𝑖𝑖=1 Velocity G
pe
Eq. 3,4 [𝑥𝑥,̈ 𝑦𝑦,̈ 𝜃𝜃]̈ = 0 Integration
Object moving Object stopped Position G

(a) (b)

Figure 5: Block diagram of the logic process behind the model to avoid the reverse motion (a), Iterative loop
ot

scheme for the simulation (b).

The simulation takes advantage of an iterative loop (shown in Fig. 5 (b)), that involves the
following steps: initialization, rotor identification, force computation, equilibrium computa-
tn

tion and two steps of integration. The initialization of the problem is made at the beginning
by providing data from the object and the surface (example: object mass, base shape, initial
position and velocity, friction coefficients, module pattern, inclination of the surface, rotor
orientations, ect.). After that, the iterative loop can start. Each step (e.g. nth step) the rotors
rin

below the object are identified, knowing the position of the object from the previous iteration
((n − 1)th ) and the disposition of the modules. Once the rotors below the object and their
orientation are determined, the forces in the contact points can be computed with the help of
the object velocity from the previous iteration ((n − 1)th ). With these forces, the equilibrium
ep

of the body and the accelerations are calculated, according to Section 4. Finally, in order
to obtain the velocity and the position, two integration steps of the acceleration vector are
implemented. The derived values are the input of the next iteration ((n + 1)th ) of the loop.
So far the model represents the operation of the surface adequately when the body ve-
locity is greater than zero in the x direction. However, when the system is in a particular
Pr

condition, such as at standstill, with rotors and surface inclined, an undesirable phenomena
such as reversal of motion (negative x) can occur. This happens because in the friction as-
sumptions, the static condition is not initially considered, so the contact forces are always of
constant value (Equations (2a) and (2b)).

9
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 However, reverse motion is obviously not possible in reality, then, in practice, when the
condition of ẋ g ≤ 0 m/s is achieved, the process needs to be adjusted. Therefore, if γ = 0º
too, the object is stopped, while, if not, an initial positive speed is given ẋ g = 0.0001 m/s

d
and it is verified if the gravity effect is stronger than the friction forces in the x direction. At
this point, if the gravity wins, the object is moving according to the process defined before

we
(Equations (4), (3a) and (3b)), on the contrary, if the gravity is not enough, the friction forces
in the x direction will be of the same magnitude of the gravity effect and the displacement
will be in the y direction (always according to Equations (4), (3a) and (3b)). For instance,
this procedure permits to obtain the motion of the object when it is placed without an initial
velocity on the inclined surface, with the rotors oriented at 45°.In fact, with a sequence of

ie
displacement in the y and x direction the movement is achieved. The logic instructions
above defined are summarized in the block diagram of Fig. 5 (a).
The simulations conducted to test the surface are divided into three main categories,

ev
according to the application of the system:
1. Sorting of material flows on a conveying line.
2. Slowing of material flows on a conveying line.
3. Stopping of material flows on a conveying line.

r
The first category of simulations concerns sorting. The objective of the set-up and the
er
sorting itself is to divert an object from the transport line. The layout considered for the
simulation involves an array of modules, of which the subset performing the task has the
rotors with the rotation axes inclined at ±45° (in Fig. 6 (a) inclined at +45°). The area out-
side the line, where the objects are directed, is simulated as an array of low-friction (µexsur f )
pe
support points, distributed as the sorting modules. In this zone the friction is totally oppo-
site to the velocity of the object, without considering any rotor inclination. The sorting is
assumed achieved when the body moving on the line is deflected in such a way that it is
only in contact with the elements outside the line.
The second simulated application is the slowing activity. The rotor arrangement consists
ot

of a few modules within a line to create a controllable friction on the object to modulate its
speed, without pushing it to the sides. The layout in this case can use rotors oriented with
their axes in the direction of the flow (ε i = 0º) (Fig. 6 (b)) or inclined at +45º in a row and
−45º in the other (Fig. 6 (c)).
tn

The last application concerns stopping the motion of an object. This task is the extreme
with respect to the previous slowing down. In fact, the modules are still placed with their
axes in the flow direction (Fig. 6 (b)) or oriented at ±45º (Fig. 6 (c)), but aim at stopping the
object.
rin

0.8 0.8 0.8

ep

0.6 0.6 0.6

y [m]

y [m]

y [m]

0.4 0.4 0.4

0.2 0.2 0.2

Pr

0 0.5 1 0 0.5 1 0 0.5 1

x [m] x [m] x [m]
(a) (b) (c)

Figure 6: Examples of simulation layouts for the: sorting (a), slowing or stopping with ε i = 0º (b) and slowing
or stopping with ε i = ±45º (c)

10
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 Fig. 6 introduces the symbols used in the following sections to display the layouts of the
simulations implemented in Matlab: the red elements correspond to the transport line, the
blue dots correspond to the area (array of support points) out of the line and the square

d
contour corresponds to the object base. Regarding the transport line, dots represent rotors
centers, while lines represent the rotors axes. The red lines before the active area (where is

we
placed the green square) are oriented perpendicularly to the flow line and their effect is to
reduce in a minimum way the motion of the box in the flow direction, while the inclined
lines (ε i ̸= 90º) simulate the sorting, the slowing or the stopping surface.
The fixed initial parameters, which are used in all the simulations, are shown in Table
1. In particular, the first four parameters are about the dimension and the inertia of the

ie
object, while the following three are the friction coefficients. These last values have been
chosen arbitrarily by making the following considerations: µmax must be a medium-high
friction value (µ = 0.5, similar to the kinetic friction coefficients between paperboard and

ev
the conveyor belt in [40, 41]), as it models the sliding of the object on the rotor in the direction
of the axis; µmin must be a low value (µ ≤ 0.1), as it models the rotational friction of the
rotors; while, µexsur f has to be a low value too (µ ≤ 0.005), as it models the area outside
the line where one can imagine having load-bearing spheres (according to the Omnitrack
catalog [42] µ = 0.005 ÷ 0.03) supporting the material. Actually, the coefficients described

r
would depend on the materials in contact, which have not yet been defined. However,
the exact values are not relevant for the purpose of proving the surface capabilities, the
er
important thing is that µmax > µmin , µexsur f is maintained. Finally, the last parameters are
the distances in x and y directions between two rotors centers and the object initial position
and acceleration. Exceptions to these starting conditions are indicated with the results for
pe
each particular case, together with the missing parameters like initial velocity of the object:
Vg0 = [ ẋ g0 , ẏ g0 , θ̇ g0 ] = [Vgx0 , Vgy0 , Vgθ0 ] and the inclination of the surface: γ.
Table 1: Initial fixed parameters for the Matlab simulations.

bxh object base dim. 0.25 m×0.25 m

ot

zg height of G 0.1 m
m object mass 10 kg
J moment of inertia (1/12)m(h2 + b2 )
tn

µmax max. friction coef. 0.5

µmin min. friction coef. 0.01
µexsur f friction coef. out of the line. 0.005
rin

dxm x dist. between 2 rotors 0.1 m

dym y dist. between 2 rotors 0.1 m
Pg0 [ x g , y g , θ g ] obj. initial pos. [b/2, 1 m,0 rad]
a g0 [ ẍ g , ÿ g , θ̈ g ] obj. initial acc. [0 m/s2 , 0 m/s2 , 0 rad/s2 ]
ep

5. Results
This section presents the simulation results for all three categories: Sorting, Slowing and
Pr

Stopping. For each category the authors illustrate their results for relevant parameter sets
and indicate a practical use of the program for the real set-up design.

Sorting. Fig. 7 (a) and Fig. 7 (b) show the sorting towards the two directions, y > 0 and y < 0,
respectively, with an initial object velocity Vgx0 = 1.5 m/s and without surface inclination.
Fig. 7 (c), on the other hand, shows the sorting at different tilting angles of the surface and

11
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 without an initial velocity. Each plot displays the trajectory of G (object’s center of mass)
during the sorting and the position and orientation of the object at the end of the simulation
time.

d
1 1 1

we
=0°, Vgx0 =1.5 m/s =0°, Vgx0 =1.5 m/s =0°, Vgx0 =0 m/s
=5°, Vgx0 =0 m/s
0.5 0.5 0.5 =10°, Vgx0 =0 m/s

y [m]
y [m]

y [m]

ie
0 0 0

-0.5 -0.5 -0.5

ev
-1 -1 -1
0 0.5 1 0 0.5 1 0 0.5 1
x [m]

r
x [m] x [m]
(a) (b) (c)
er
Figure 7: Illustration of the trajectories and the final position and orientation of the object during: the sorting
towards y > 0 (up) task (a), the sorting towards y < 0 (down) task (b), the sorting towards y < 0 (down) with
different γ (c).
pe
In Fig. 7 (a), the green square and its trajectory show that with rotors inclined at −45º,
sorting towards y > 0 can be achieved, as the object is shifted completely off the line.
The opposite sorting condition was tested too, the only difference for the initial data
was the direction of the rotors (+45°). The result is shown in Fig. 7 (b) and the trajectory
is mirrored to the first one. In fact, like expected in reality, the ending displacements of the
object are equal in modulus and with opposite sign for the y and θ displacements (Table 2).
ot

Fig. 7 (a), (b) show the capability of the surface to deflect the object trajectory for sorting
purposes. Furthermore, coherent results were achieved for both sorting directions, for a
reasonable initial velocity and within one second of simulation time. From now on, since
tn

the mirroring of the results has been demonstrated for a symmetric rotor arrangement, the
outputs are shown only for one sorting direction.
After this first presentation of the sorting capability for the under-actuated system, in-
stead of using an initial object velocity on a horizontal surface, the sorting task has been
rin

studied with an initially stationary object on an inclined surface with the identical rotor
arrangement (Fig. 7 (c)).
In Fig. 7 (c) the first case (green) proofs that without any external actuation the object is
not moving in the simulation, as in the real word. In the second case (black square Fig. 7 (c)),
with the surface inclination of γ = 5º, the body moves, but for the chosen simulation time
ep

(t = 1.5 s) it has not yet completed its sorting. This can be seen from the fact that the
black square contour still contains the centre of a sorting rotor (red). Finally, the third case
(magenta square), where the surface inclination is γ = 10º, reports a completed sorting
activity. Summarizing the black and magenta bodies in Fig. 7 (c) demonstrate that it is
Pr

possible to transport and sort parts by inclining the surface. In particular the more it is tilted
the faster the object moves and the quicker the sorting is achieved. In Fig. 7 (c), the values
of the position and orientation reached at the end of the simulation (tblack = 1.5 s) for the
γ = 5º case (black line) are [ x g , y g , θ g ]black = [0.369 m, −0.171 m, −36.05º] (Table 2), while
those of the magenta line for the same value of y g = −0.171 m are [ x g , y g , θ g ]mag = [0.402

12
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 m, −0.171 m, −28.62º] and the simulation time for which they are obtained is tmag = 0.917
s. Comparing these values, it can be seen that when the surface is more inclined (magenta
line), it requires more space in the x direction to perform the sorting x g,mag > x g,black , while

d
the time required is less tblack − tmag = 0.583 s. In addition different γ also produce different
final orientations, in this case smaller for larger γ: θ g,mag < θ g,black .

we
To summarize the results and the initial data about the previous sorting simulations Table
2 is reported.
Table 2: Specific initial data and results for the sorting simulations.

Vg0 [m/s,m/s,rad/s] γ [°] sim. time [s] final [ x g , y g , θ g ][m,m,º] figure

ie
[1.5, 0, 0] 0 1 [0.859, 0.532, 14.83] Fig.7(a)
[1.5, 0, 0] 0 1 [0.859, −0.532, −14.83] Fig.7(b)

ev
[0, 0, 0] 0 1.5 [0, 0, 0] Fig.7(c) green
[0, 0, 0] 5 1.5 [0.369, −0.171, −36.05] Fig.7(c) black
[0, 0, 0] 10 1.5 [1.057, −0.515, −12.16] Fig.7(c) magenta

r
A practical use of the simulation is to determine the minimum number of modules nec-
essary for sorting before building the real system. In general, as it is already presented, this
er
result is affected by various initial conditions. However, it can be interesting to know how
many columns of rotor units are necessary to successfully complete the different tasks for
different object sizes and velocity ranges. In order to derive this relationship, a set of simu-
lations can be performed. Table 3 reports the results of the discussed analysis for two object
pe
sizes: b × h = 0.25 m ×0.25 m, called ”S-box” and b × h = 0.45 m×0.45 m, called ”B-box”.
”B-box” and ”S-box” have the same mass (m = 10 kg) and height (z g = 0.1 m).
The counting of the number of columns for the sorting is done starting from the first
column with the inclined rotors to the last one touched by the object before being out of
the line. As it is visible from Table 3, the sorting with low initial velocities is not always
ot

possible: with Vgx0 less than 0.8 m/s for ”S-box” and less than 1.1 m/s for ”B-box” the
boxes come to a stop within the transport line. Despite this, the speed values considered in
the simulations are within the most common operating ranges of conveyors, however, any
tn

speed can be implemented in the software. In general, the missed sorting occurs because the
deflecting forces also tend to slow the object, which may stop before reaching the target. In
addition, the displacement to achieve to be out of the line is related to the dimension of the
object, so the bigger it is, the more absolute transversal displacement is required to achieve
the sorting. Summarizing, thanks to the simulation described, it is possible to realize such
rin

tables depending on the specific application, considering that the results are influenced by
the layout of the modules and the initial position. In particular, for Table 3 the ”S-box” layout
is the same of the sorting simulations of Fig. 7 (b), but with more columns of sorting rotors
in order to test a bigger range of velocities. Instead, for the ”B-box” there are two more rows
ep

of rotors in the sorting line, making a total of four instead of two lines. This layout looks like
the one in Fig. 8 (c), (d) but with the rotors oriented for sorting towards y < 0 (ε i = +45º),
instead of the arrangement for slowing or stopping (ε i = 0º or ε i = ±45º).
Pr

13
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 Table 3: Minimum number of rotor columns required to achieve the sorting with γ = 0°, but changing Vgx0 .

Vgx0 [m/s] nº columns S-box nº columns B-box

d
0.8 3 /
0.9 3 /

we
1.0 4 /
1.1 4 7
1.2 4 7
1.3 4 7

ie
1.4 4 7
1.5 4 8

ev
1.6 5 8
1.7 5 8
1.8 5 8
1.9 6 9

r
2.0 6 9
2.1 6 er 9
2.2 7 10
2.3 7 10
pe
2.4 7 11
2.5 8 11
2.6 8 12
2.7 8 12
ot

Slowing. The second set of results is derived from the slowing simulations. The purpose of
the simulations is to simplify and speed up the determination of the number of modules
required for the slowing task, which as with sorting is linked to the initial parameters.
tn

0.8 0.8 0.8 0.8

0.6 0.6 0.6 0.6

y [m]

y [m]

y [m]

y [m]

0.4 0.4 0.4 0.4

rin

0.2 0.2 0.2 0.2

0 0.5 1 0 0.5 1 0 0.5 1 0 0.5 1

x [m] x [m] x [m] x [m]

(a) (b) (c) (d)

ep

Figure 8: Layouts of the simulations for the slowing and the stopping tasks with: S-box ε i = 0° (a), S-box
ε i = ±45° (b), B-box ε i = 0° (c), B-box ε i = ±45° (d).

In order to show the results of the slowing task, in this case the evolution of the object
Pr

speed over time (Vg ) is reported. Differently from the sorting, here the object is not pushed
away from the line, so the trajectory does not represent the course of the activity as simply
as speed does. Considering the layout of the system the only velocity component different
from zero is in the x direction, which is the one plotted. Four different set-ups shown in Fig. 8
were simulated to demonstrate the concept’s operation. The graphs in Fig. 9 (a), (c) show

14
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 the results of the slowing simulations for the ”S-box” and ”B-box” cases, respectively. The
initial data for this analysis are the same of Table 1, with additionally γ = 0° and Vg0 = [1.5
m/s, 0 m/s, 0 rad/s]. As visible from Fig. 9 (a), (c), both the layouts are capable to reduce the

d
speed of the object, fulfilling the slowing task. However, the speed drop is different with the
orientation of the rotors. In fact, when the rotors are with ε i = 0°, all the Fpar forces in the x

we
direction cause a stronger slowing, while with ε i = ±45° layout, where the Fpar are inclined
like the rotor axes, the slowing is lower. For this initial parameters and rotor dispositions, the
ratios between the speed before and after the slowing rotors (r = Vgxout /Vgx0 ) are: r0S−box =
0.23, r0B−box = 0.18 and r± S−box
45
B−box
= 0.53, r± 45 = 0.52. Additionally, Fig. 9 shows the effect
of Fperp , which is slightly reducing the velocity when the rotors are with ε i = 90°. In fact,

ie
the plots in Fig. 9 (a), (c) show that the lines are a bit inclined before and after the slowing
area, which is included between the ”Start slowing” and the ”End slowing” dotted lines. It
represents the contribution of rotational friction to the deceleration.

ev
1.6 1.5

= 0° = 0°
1.4
= 45° = 45°

r
Start slowing Start stopping
1.2
End slowing = 0° 1 End stopping = 0°
End slowing = 45° End stopping = 45°
Vgx [m/s]

Vgx [m/s]

1 er
0.8
0.5
0.6
pe
0.4

0.2 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
t [s] t [s]
(a) (b)
ot

1.6 1.5

= 0° = 0°
1.4
= 45° = 45°
Start slowing Start stopping
tn

1.2
End slowing = 0° 1 End stopping = 0°
End slowing = 45° End stopping = 45°
Vgx [m/s]

Vgx [m/s]

0.8
0.5
rin

0.6

0.4

0.2 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
ep

t [s] t [s]
(c) (d)

Figure 9: Velocity trends for ”S-box” during the slowing (a) and the stopping (b) and the ”B-box” during the
slowing (c) and the stopping (d) with: ε i = 0° (red), ε i = ±45° (blue).
Pr

As for sorting, a parameters study was conducted within the simulation environment to
derive the required number of rotor columns to achieve the slow down task of the objects.
The results are displayed in Table 4. Also in this case, different initial speed and the two dif-
ferent object sizes were tested (”S-box”, ”B-box”). The reference layouts for the simulations

15
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 are the ones in Fig. 8. In Table 4 the counting of the columns number begins with the first
column with inclination: ε i = 0° or ε i = ±45°, that the object encounters starting from the
left. For example, considering Fig. 8 (a) and assuming that slowing is verified,the result will

d
be: 2 columns (with ε i = 0°). The target condition was set to at least halving the speed of
the object while remaining a positive value after the slowing process (0.5Vgx0 ≥ Vgxout > 0).

we
The two configurations (ε i = 0° and ε i = ±45°) were possible for the rotor layout, so the
configuration with the smallest number of columns was chosen case by case to satisfy the
slowing condition.
Table 4: Minimum number of rotor columns and orientations of their axes (ε i ) required to achieve the slowing
(0.5Vgx0 ≥ Vgxout > 0) with γ = 0°, but changing Vgx0 .

ie
Vgx0 [m/s] nº columns S-box nº columns B-box

ev
0.9 1 ⇒ ε i = ±45° 1 ⇒ ε i = ±45°
1.0 1 ⇒ ε i = ±45° 1 ⇒ ϵi = ±45°
1.1 1 ⇒ ε i = 0° 1 ⇒ ε i = 0°

r
1.2 1 ⇒ ε i = 0° 1 ⇒ ε i = 0°
1.3 2 ⇒ ε i = ±45° 2 ⇒ ε i = ±45°
1.4
1.5
er
2 ⇒ ε i = ±45°
2 −→ ε i = 0°
2 ⇒ ε i = ±45°
2 −→ ε i = 0°
1.6 2 ⇒ ε i = 0° 2 ⇒ ε i = 0°
pe
1.7 3 ⇒ ε i = ±45° 3 ⇒ ε i = ±45°
1.8 3 ⇒ ε i = 0° 3 ⇒ ε i = 0°
1.9 3 ⇒ ε i = 0° 3 ⇒ ε i = 0°
2.0 3 ⇒ ε i = 0° 3 ⇒ ε i = 0°
2.1 4 ⇒ ε i = 0° 4 ⇒ ε i = 0°
ot

2.2 4 ⇒ ε i = 0° 4 ⇒ ε i = 0°
2.3 4 ⇒ ε i = 0° 4 ⇒ ε i = 0°
tn

2.4 5 ⇒ ε i = 0° 5 ⇒ ε i = 0°
2.5 5 ⇒ ε i = 0° 5 ⇒ ε i = 0°
2.6 6 ⇒ ε i = 0° 5 ⇒ ε i = 0°
2.7 6 ⇒ ε i = 0° 6 ⇒ ε i = 0°
rin

Stopping. The last results are related to the simulation of the stopping condition. In this case,
the initial parameters and layouts implemented in the software are the same than before
(Fig. 8) and only the number of columns of the working rotors changes. In fact, starting
ep

from the slowing modules from Table 4 and adding to the right an extra column of rotors
with ε i = 0°, the motion of the object can be stopped. The addition of a column proved to be
an effective method for simulated speed values, it is not certain that for different values one
column is sufficient. Therefore, considering this, Table 4 can be used also for the stopping
layout planning. Like for the slowing, an example (Fig. 9 (b), (c)) is presented and only the
Pr

Vgx0 of the object is reported. However in this case the deceleration produced by the rotors
of the slowing, plus the extra column, is capable to stop the object even if the initial velocity
simulated is the same of the slowing (Vg0 = [1.5 m/s, 0 m/s, 0 rad/s]). In general, according
to Table 4 and Fig. 9, the size of the object for the slowing and the stopping tasks has less
influence than in the sorting (Table 3). This can be seen from the fact that, compared to Table

16
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 3, in Table 4, the numbers of columns to slow (then also to stop) the ”S-box” and the ”B-box”
are only different in one case (Vgx0 = 2.6 m/s).
In conclusion, the simulations enable to determine the layout of the surface, including

d
the module arrangement and the surface inclination for specific applications, while taking
the initial conditions of the object into account and which tasks have to be performed.

we
6. Validation
Introduction to the validation. In order to validate the previous results, the sorting system was
also modeled in a commercial software for multi-body dynamic simulations: Hexagon D&E

ie
Adams MSC. The multi-boy dynamic model, called hereinafter Adams for brevity, has two
main objectives:
1. Providing further demonstration of the functioning of the under-actuated system in-

ev
troduced in this paper;
2. Comparing the results with the ones from Matlab & Simulink to have a first validation
of the analytical model and the self-developed simulation environment.
Simulations with the Adams software require longer computational time compared to the

r
Matlab simulations (timeAdams ≈ 60 ÷ 300 s, timeMatlab ≈ 0.1 ÷ 2 s, with a HP ProDesk 400
G7). For this reason Adams was only used for the validation part and not as a standard
er
simulation environment. In fact, using a software as a planning and control tool within the
physical system in real time requires very short calculation times. Fig. 10 (a) shows one
surface configuration, that has been previously used in the Matlab environment for sorting.
The identical surface configuration has been replicated in Adams, shown in Fig. 10 (b).
pe
1
Contour of the piece base
Direction of the rotor axis
Rotor center
0.5 Area out of the line
ot
y [m]

0
tn

-0.5

-1
rin

0 0.5 1
x [m]
(a) (b)

Figure 10: Model of under-actuated active surface configured for sorting in: Matlab (a), Adams (b).
ep

As it is visible in Fig.10(b), the sorting set-up is designed in Adams using three different
types of bodies: a box (orange wire frame) for the object, cylinders (colourful cylinders) for
the rotors and spheres (pink spheres) for the area out of the line. The chosen inertial and
dimensional parameters are shown in Table 5. It should be noted that, for the validation,
Pr

the values of some inertial parameters are different from previous simulations because the
mass properties in Adams were introduced by assigning the material of the bodies. The
constraints used for the model are: a fixed joint for the spheres and rotational joints for the
cylinders, while the box has contact constraints with the other bodies. The parameters of the
contact and the joints are illustrated in Table 6. In particular, for the first five values of this

17
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 table, the selection was done according to the suggestion of [43, 44]: St and Pd are the default
values for the contact, while Dp = 8 kg/s instead of 10 kg/s, Stv and Ftv are smaller than the
default (respectively 100 mm/s and 1000 mm/s) to simulate better the stiction at low speeds.

d
The friction coefficients are the same values implemented in Matlab, while the joint friction
parameters are selected to simulate the rotational friction of the joint, like in the analytic

we
model. To conclude the description, in Adams, the inclination of the surface is simulated
changing the gravity vector direction, while the initial velocity is simulated with a force on
the object, applied for a millisecond at the beginning of the simulation. Lastly, the simulation
time considered was t = 1 s and the step size in the range [1 × 10−4 s ÷ 0.5 × 10−4 s].
Table 5: Inertial and geometrical parameters im- Table 6: Contact and friction parameters imple-

ie
plemented in the Adams for the simulation. mented in the Adams for the simulation.

Objects parameters Contact parameters

ev
m box mass 5.48 kg St stiffness cont. 105 MPa
Ixx box m. inertia x axis 5.70 × 10−2 kgm2 Dp damping cont. 8 kg/s
Iyy box m. inertia y axis 4.68 × 10−2 kgm2 Pd penetration depth cont. 0.1 mm
Izz box m. inertia z axis 5.70 × 10−2 kgm2 Stv stiction transition vel. 0.1 mm/s

r
b × h × zg object dim. 0.25 m×0.25 m×0.1 m Ftv friction transition vel. 1 mm/s
m rotor mass 8.61 × 10−3 kg
er µmax friction coef. 0.5
Ixx rot. m. inertia x axis 4.30 × 10−7 kgm2 µexsur f friction coef. out the line. 0.005
Iyy rot. m. inertia y axis 2.87 × 10−2 kgm2 Joint friction parameters

Izz rot. m. inertia z axis 2.87 × 10−7 kgm2 µmin min friction coef. 0.01
pe
r×l rot. dim. 0.01 m×0.01 m Tv transition vel. 0.001

γ inclin. of the surf. [5 ÷ 15]º Friction option chosen sliding only

The Matlab set-up (Fig. 10 (a)) corresponds to the previous sorting set-up in Section 5
Fig. 7, but in this case it uses the same geometric and inertial data as Adams, to simulate the
same sorting process. In this way, the results from the two software differ only because of
ot

their intrinsic analytic modeling.

After this brief introduction to the Adams model, the simulations performed to meet the
two objectives are now described. The configuration with a tilted surface and no initial
tn

object velocity has been simulated for different angles of inclination ranging from: γ = 5º to
γ = 14º with an increment of 0.25º. Additionally, the horizontal configuration (γ = 0º) has
been simulated with an initial velocity Vgx0 = 0.878 m/s.

Results of the validation. The results of the validation are distinguished according to the two
rin

objectives of the Adams model. Concerning the first objective, i.e. the proof of functioning of
the under-actuated system, Fig. 11 graphically reports the results of some Adams simulations
carried out. Considering initially only Fig. 11 (a), (b), in Fig. 11 (a) sorting is simulated given
an initial velocity of zero (Vgx0 = 0 m/s) and an inclination of γ = 10º, in Fig. 11 (b) the
ep

initial velocity is Vgx0 = 0.878 m/s and the inclination is zero (γ = 0º). In the figures
Fig. 11 (a), (b) it is possible to see the trajectory taken by the G of the object and its final
position and orientation given the simulation time of t = 1 s. The object at the end of the
simulation is for both images (Fig. 11 (a), (b)) outside the sorting zone (zone with cylinders)
Pr

so the activity is considered completed.

18
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 d
we
(a) (b) (c)

Figure 11: Object trajectories from Adams with: γ = 10º, Vgx0 = 0 m/s and t = 1 s of simulation (a), γ = 0º ,
Vgx0 = 0.878 m/s and t = 1 s of simulation (b), γ = 5º, Vgx0 = 0 m/s and t = 1.8 s of simulation (c).

ie
In addition to the case illustrated in Fig. 11 (a), among all the thirty-seven simulations
made by varying the inclination of the surface, the ones with γ ≥ 9.5º (Fig. 12 (a), (b))

ev
complete the task. For the rest, the target is not really missed, but, because of the limited
simulation time (t = 1 s), it is not achieved yet. This time was selected because it represents
a normal operating condition, for example: there is 1 m of space available in the line for
sorting and the conveyor has a speed of 1 m/s. However, as a general rule, when √ the surface

r
is inclined and with the rotors like in Fig. 11 the motion does not self-stop, if ( 2/2)µmin <
sin(γ). As proof of this, Fig. 11 (c) displays the trajectory of the object with γ = 5° but t = 1.8
er
s of simulation time, instead of t = 1 s and the sorting is clearly obtained. Therefore, the
sorting is achievable with this under-actuated system, but it takes time and space. Instead,
if the surface is not inclined and there is just the initial speed, the time could be not the
only cause of the missed sorting. In this case the object velocity is inevitably reduced by the
pe
friction until the part is stopped, because the gravity is not helping the motion. The same
result was highlighted before in the Matlab environment, which confirms that the simulation
environment can be used to predict these conditions.
Fig. 12 plots the final positions when the sorting is considered achieved for the two soft-
ware (Fig. 12 (a) Adams, Fig. 12 (b) Matlab).
ot

0.6 =9.5º =10.75º =12º =13.25º 0.6 =9.5º =10.75º =12º =13.25º
tn

=9.75º =11º =12.25º =13.5º =9.75º =11º =12.25º =13.5º

0.4 =10º =11.25º =12.5º =13.75º 0.4 =10º =11.25º =12.5º =13.75º
=10.25º =11.5º =12.75º =14º =10.25º =11.5º =12.75º =14º
0.2 =10.5º =11.75º =13º 0.2 =10.5º =11.75º =13º
y [m]

y [m]

0 0
rin

-0.2 -0.2

-0.4 -0.4

-0.6 -0.6
ep

0 0.5 1 1.5 0 0.5 1 1.5

x [m] x [m]
(a) (b)

Figure 12: Final position with γ > 8.5º in: Adams MSC (a), Matlab (b).
Pr

Having completed the verification of the first objective of the Adams model, it is possible
to move on to the second one. Summarizing briefly, objective number two is to exploit the
Adams model to show the proximity with the Matlab results, considering that the same initial
data are provided. Fig. 13 shows a graphical comparison between the two models referred

19
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 to the two trajectories of Fig. 11 (a), (b). In this case the trajectories are transposed into a
graph (Fig.13) with the corresponding result from Matlab.

d
Adams Adams
0.6 0.6

we
Matlab Matlab

0.4 0.4

0.2 0.2
y [m]

y [m]
0 0

ie
-0.2 -0.2

-0.4 -0.4

-0.6 -0.6

ev
0 0.5 1 1.5 0 0.5 1 1.5
x [m] x [m]
(a) (b)

r
Figure 13: Comparison of the Trajectories with: γ = 10º , Vgx0 = 0 m/s (a), γ = 0º, Vgx0 = 0.878 m/s (b).
er
As it is visible from Fig. 13 (a), (b) the two objects follow very close trajectories and end
up almost in the same position and with the same orientation. When the initial input is the
velocity, the difference appears slightly higher, but this may be due to the way the initial
pe
speed was provided in Adams. The absolute and the relative percentage errors obtained for
these two simulations are: | Eabs | = [0.0058 m, 0.0035 m, 1.839°], | E% | = [1.28, 1.62, 7.49] %
when the input is γ = 10°, Vgx0 = 0 m/s and | Eabs | = [0.016 m, 0.0014 m, 1.64°], | E% | =
[3.77, 6.55, 7.16] % when the input is γ = 0º, Vgx0 = 0.878 m/s. The errors are calculated as
shown in Equations (5a) and (5a).
ot

Eabs = [ex , ey , eθ ] = FINAL DISPLACEMENT Adams − FINAL DISPLACEMENT Matlab (5a)

Eabs
E% = [ex% , ey% , eθ% ] = × 100 (5b)
tn

FINAL DISPLACEMENT Adams

The two examples graphically visualize the similarities and the differences between the
models. However, to show and summarize all the results for the thirty-seven set-ups (γ =
5°÷14°), the final displacements of the Adams and Matlab simulations together with the er-
rin

rors between them are reported respectively in Fig. 14 and Fig. 15.
ep
Pr

20
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 0.6

x g [m]
0.4

d
0.2
5 6 7 8 9 10 11 12 13 14

we
[º]
0
y g [m]

-0.2

ie
-0.4
5 6 7 8 9 10 11 12 13 14
[º]

ev
-20
[°]
g

-30 Adams
Matlab
-40

r
5 6 7 8 9 10 11 12 13 14
[º]
er
Figure 14: Final displacements obtained with the two software: Adams (red) and Matlab (blue) .
pe
2 10

|e x|[mm]
|e x%|

1 5

0 0
5 6 7 8 9 10 11 12 13 14
[º]
ot

10
10
|e y|[mm]
|e y%|

5
5
tn

0 0
5 6 7 8 9 10 11 12 13 14
[º]

20 5
Relative % err.
rin

Absolute err.
|e |[°]
| %

10
|e

0 0
5 6 7 8 9 10 11 12 13 14
ep

[º]

Figure 15: Absolute (green) and Relative percentage (magenta) errors between the final displacements result-
ing from the two software.
Pr

Considering the x and y results (Fig. 14 and Fig. 15), the two models (Adams, Matlab) are
not so different and the relative percentage errors are below ex% < 3 % and ey% < 10 %. The
trend of the percentage error does not appear increasing and the higher error values occur
for small γ values, while the absolute errors, on the contrary, grow almost regularly with
γ. This is due to the fact that small inclinations generate reduced displacements, especially

21
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 along y and the sorting is not completed in the fixed simulation time (t = 1 s). This implies
that in the calculation of the percentage error there are initially very small denominators
and then they become an order of magnitude larger. Also for the rotation, the phenomenon

d
of the larger percentage error for smaller rotations is noticed, however, higher eθ% values
are reached (eθ% ≤ 20%) compared to those of x and y. In general, larger errors of rotations

we
can be explained by comparing how the calculation of normal forces occurs between the two
simulations. In fact, Adams uses a penetration model to calculate the contact forces, while the
optimization algorithm (Equation (1)) is implemented in Matlab. The distribution of normal
forces (and therefore also of friction) mainly affects the calculation of the moment, since the
position with respect to the pole is important. While, for the x and y displacements, the

ie
over mentioned distribution does not influence, since the resultant force does not change.
In addition, an explanation for others slight deviations in trends is the use of different step
sizes in some Adams simulations. This was necessary because in a few cases the specific step

ev
size chosen was creating a numerical problem, preventing the completion of the simulation.
In conclusion, however, the errors are limited and the two models behave in a similar way,
thus giving value to the results provided by Matlab.
To quantify how different Matlab and Adams values are, the mean values and standard

r
deviations of the two types of error are listed in Table 7 for the different displacements.
Table 7: Mean and standard deviation of the relative percentage and absolute errors obtained from the differ-
er
ence between Adams and Matlab simulation.

ex% ey% eθ% ex [mm] ey [mm] eθ [º]

µ −1.2 −0.5 −8.4 −5.5 3.4 1.8
pe
σ 0.5 3.4 7.3 2.8 5.4 1.5

As results from Table 7, the errors of x and y displacements are very limited. This guar-
antees that the placing made by the real system can be foreseen in advance with an accuracy
appropriated to the applications, so when the sorting line is designed, the simulation can
ot

be used to define the layout. For the rotation the errors are bigger, but, despite this, for a
simple sorting operation the final orientation is not interesting, so the Matlab model can be
considered reliable. Furthermore, if in an hypothetical application the errors allowed are the
in the range of those shown in Table 7, the model provides usable results. Concerning that,
tn

in many cases it is simply important that the object is not rotated by 180º or 90º with respect
to the desired condition.

7. Conclusions
rin

In this paper the authors present a new under-actuated system for several intra-logistical
tasks. Their system is a modular surface composed of idle instead of actuated rotors, whose
axes can be fixed in defined, discrete positions within the surface plane. The surface can
ep

be used in a horizontal orientation by exploiting an object’s initial velocity, in a tilted orien-

tation by exploiting the gravitational force on the object, or in a combination of both. The
authors derived an analytic model and implemented a programmable simulation environ-
ment with the software Matlab for this modular surface. As result, the functioning of the
surface concept for the sorting, stopping and slowing activities was demonstrated, together
Pr

with the capabilities of the simulation environment. In particular, the Matlab code showed
its potential for predicting, with very short calculation times (≈ or < 1 s) the number of
modules required for the three handling tasks and how the transported object will behave
by simply changing the initial conditions. These results and examples highlight the use-
fulness of this environment for real system planning and design. In addition, a validation

22
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 of the concept and of the mentioned simulation environment was conducted with the soft-
ware Adams. As Adams is a highly sophisticated and well excepted commercial software
for dynamic simulations frequently used by engineers to simulate and predict the physical

d
interaction of different components adequately, it is a reasonable tool to be considered as a
first reference for the comparison.

we
In conclusion, the simplifications introduced in the surface, such as the under-actuation
and the discrete number of orientations for the rotors, are not limiting the handling capa-
bilities, but rather they are minimizing the number of constructive components requested
and thus, the costs. In fact, the same goals can be achieved with a reduced design and using
in a convenient manner the external actuation, for example gravity or previous conveyors,

ie
already in the line. Additionally, the validation with Adams showed also the accuracy of the
main simulation. In fact, the differences between the two models are limited and the errors
acceptable for many intra-logistics applications. This may open the way to other possible

ev
tasks for the surface integrated with the Matlab environment, such as position and trajec-
tory tracking. In these cases, each time the external environment requires a new position or
trajectory of the object, the software calculates the orientation to be given to the rotors. The
physical system must be integrated with sensors to adjust in real-time the rotors and achieve

r
the tracking objectives. The implementation of sensors and control strategies for trajectory
and position tracking greatly increases the adaptability and flexibility of the system.
No reference is made in the article to construction details in order to keep the validity of
er
the work presented as general as possible. At this point, in order to provide some practical
6 6 5 4 4 3 3 2 2 1 1
elements and above 5all to highlight the feasibility of the concept, some schematic solutions
are proposed in Fig. 16.
pe
D D D D

Y Y Z Z Y Y Z Z
Electromagnet
Electromagnet Contact
Contact with
with the
the piece
piece Contact
Contact with
with the
the piece
piece
X X X X X X X X
Y Y
Y Y
Z Z Z Z Idle sphere
Idle sphere

Idle
Idle wheel
wheel
ot

NN S S NN S S
C C C C
Mangetic
Mangetic rotation
rotation
axis
tn

axis Mechanical rotation

Mechanical rotation
Actuated
Actuated pinpin axis
axis
(a) (b)

Figure 16: Schematic concepts for the under-actuated module: idle wheel concept (a), spherical concept (b).
rin

B B B B
Fig. 16 (a) shows an idle wheel mounted on a vertical axis of rotation. The operating
principle is similar to the functioning of a stepper motor, which could be used for this pur-
pose. In contrast, the concept in Fig. 16 (b) concerns a spherical rotor whose axis of rotation
is locked using mechanically or electro-mechanically driven pins. The new tracking objec-
ep

tives, an accurate design and the control law for the modules, are ongoing research topics.

Data Data Data Data

A A A A
Progettato
da da Controllato
da da Approvato
Approvato
da da
Acknowledgements Progettato
ebianchi
ebianchi
Controllato
16/09/2022
16/09/2022
Pr

This research work was undertaken in the con-text of DIGIMAN4.0 project (”Digital 1 /11/ 1
Disegno_concept_conclusioni
Disegno_concept_conclusioni
Edizione FoglioFoglio
Edizione

Manufacturing Technologies for Zero-defect”, https://www.digiman4-0.mek.dtu.dk/). DIGI-

6 6 5 5 4 4 3 3 2 2 1 1

MAN4.0 is a European Training Net-work supported by Horizon 2020, the EU Framework

Pro-gramme for Research and Innovation (Project ID: 814225).
This research received no external funding.

23
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 References
[1] G. Fantoni, M. Santochi, G. Dini, K. Tracht, B. Scholz-Reiter, J. Fleischer, T. Kristoffer

d
Lien, G. Seliger, G. Reinhart, J. Franke, H. Nørgaard Hansen, A. Verl, Grasping devices
and methods in automated production processes, CIRP Annals 63 (2014) 679–701.

we
[2] A. Kamagaew, J. Stenzel, A. Nettsträter, M. ten Hompel, Concept of Cellular Trans-
port Systems in facility logistics, in: The 5th International Conference on Automation,
Robotics and Applications, 2011, pp. 40–45.

[3] C. I. Rizescu, A. I. Pleşea, D. Rizescu, Modular Transport and Sorting System with

ie
Omnidirectional Wheels, MATEC Web of Conferences 343 (2021) 08011.

[4] C. Uriarte, A. Asphandiar, H. Thamer, H. Thamer, A. Y. Benggolo, M. Freitag, Con-

trol strategies for small-scaled conveyor modules enabling highly flexible material flow

ev
systems, ScienceDirect, Procedia CIRP Conference on Intelligent Computation in Man-
ufacturing Engineering, 18-20 July 2018,Gulf of Naples, Italy (2018).

[5] K.-F. Böhringer, V. Bhatt, B. R. Donald, K. Goldberg, Algorithms for Sensorless Manip-

r
ulation Using a Vibrating Surface, Algorithmica, Springer (2000).

[6] K. F. Böhringer, Programmable force fields for distributed manipulation, and their im-
er
plementation using micro-fabricated actuator arrays, Ph.D. thesis, Cornell University,
1997.

[7] S. Konishi, H. Fujita, A conveyance system using air flow based on the concept of
pe
distributed micro motion systems, Journal of Microelectromechanical Systems 3 (1994)
54–58.

[8] G. Kritikou, N. Aspragathos, Activation Algorithms for the Micro-Manipulation &

Assembly of Hexagonal Microparts on a Programmable Platform, 2018.
ot

[9] H. Oyobe, Y. Hori, Object Conveyance System ” Magic Carpet” Consisting of 64 Linear
Actuators - Object Position Feedback Control with Object Position Estimation, 2001
IEEE/ASME international Conference on Advanced Intelligent Mechatronics Proceed-
tn

ings 8-12 July 2001 Como, Italy, 2001.

[10] T. Piatkowski, Analysis of translational positioning of unit loads by directionally-

oriented friction force fields, Mechanism and machine theory 46 (2011) 201–217.
rin

[11] S. Tao, Z. Youpan, Z. Hui, W. Pei, Z. Yongguo, L. Guangliang, Three-wheel Driven

Omnidirectional Reconfigurable Conveyor Belt Design, in: 2019 Chinese Automation
Congress (CAC), 2019, pp. 101–105.

[12] J. Whiting, R. Mayne, C. Melhuish, A. Adamatzky, A Cilia-inspired Closed-loop Sensor-

ep

actuator Array, Journal of Bionic Engineering 15 (2018) 526–532.

[13] J. S. Keek, S. L. Loh, S. H. Chong, Design and Control System Setup of an E-Pattern
OmniwheeledCellular Conveyor, MDPI, Machines (2021).
Pr

[14] T. Krühn, S. Falkenberg, L. Overmeyer, Decentralized control for small-scaled conveyor

modules with cellular automata, in: 2010 IEEE International Conference on Automa-
tion and Logistics, 2010, pp. 237–242.

24
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 [15] K. F. Böhringer, B. R. Donald, N. C. MacDonald, Programmable Force Fields for Dis-
tributed Manipulation, with Applications to MEMS Actuator Arrays and Vibratory
Parts Feeders, THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH (1999).

d
[16] P. Will, W. Liu, Parts manipulation on a mems intelligent motion surface, USC INFOR-

we
MATION SCIENCES INSTITUTE (1994).

[17] P. U. Frei, M. Wiesendanger, R. Büchi, L. Ruf, Simultaneous Planar Transport of Mul-

tiple Objects on Individual Trajectories Using Friction Forces, Kluwer Academic Pub-
lishers, 2000.

ie
[18] M. Ataka, B. Legrand, L. Buchaillot, D. Collard, H. Fujita, Design, Fabrication, and
Operation of Two-Dimensional Conveyance System With Ciliary Actuator Arrays,
IEEE/ASME Transactions on Mechatronics 14 (2009) 119–125.

ev
[19] K. P. Becker, Y. Chen, R. J. Wood, Mechanically Programmable Dip Molding of High
Aspect Ratio Soft Actuator Arrays, Advanced Functional Materials 30 (2020) 1908919.

[20] G. Fantoni, M. Santochi, Development and testing of a brush feeder, CIRP Annals 59

r
(2010) 17–20.

[21] Z. Chen, Z. Deng, J. S. Dhupia, M. Stommel, W. Xu, Motion Modeling and Trajectory
er
Tracking Control for a Soft Robotic Table, IEEE/ASME Transactions on Mechatronics
(2021) 1–11.

[22] I. A. Raptis, C. Hansen, M. A. Sinclair, Design, modeling, and constraint-compliant

pe
control of an autonomous morphing surface for omnidirectional object conveyance,
Robotica 40 (2022) 213–233.

[23] M. A. Robertson, M. Murakami, W. Felt, J. Paik, A Compact Modular Soft Surface

With Reconfigurable Shape and Stiffness, IEEE/ASME Transactions on Mechatronics
ot

24 (2019) 16–24.

[24] K. Boutoustous, G. J. Laurent, E. Dedu, L. Matignon, J. Bourgeois, N. Le-Fort-Piat, Dis-

tributed control architecture for smart surfaces, in: The 2010 IEEE/RSJ International
tn

Conference on Intelligent Robots and Systems October 18-22, 2010, Taipei, Taiwan,
2010.

[25] X. Chen, W. Zhong, C. Li, J. Fang, F. Liu, Development of a Contactless Air Con-
veyor System for Transporting and Positioning Planar Objects, MDPI,Micromachines
rin

9 (2018).

[26] P.-J. Ku, K. Winther, H. Stephanou, R. Safaric, Distributed control system for an active
surface device, in: Proceedings 2001 ICRA. IEEE International Conference on Robotics
ep

and Automation (Cat. No.01CH37164), volume 4, 2001, pp. 3417–3422 vol.4.

[27] Y. Mita, M. Arai, A. Tixier, H. Fujita, Bulk Micromachined Durable Air-Flow Microac-
tuator Array for Robust Conveyance Systems, in: E. Obermeier (Ed.), Transducers ’01
Eurosensors XV, Springer Berlin Heidelberg, Berlin, Heidelberg, 2001, pp. 718–721.
Pr

[28] S. Mobes, G. J. Laurent, C. Clevy, N. L. Fort-Piat, B. Piranda, J. Bourgeois, Toward a

2D Modular and Self-Reconfigurable Robot for Conveying Microparts, in: 2012 Second
Workshop on Design, Control and Software Implementation for Distributed MEMS,
2012, pp. 7–13.

25
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407
 [29] M. Turitto, S. Ratchev, X. Xue, M. Hughes, C. Bailey, Pneumatic contactless microfeeder
design refinement through cfdsimulation, in: Proceedings of 4M 2007, The third inter-
national conference on multi-material micro manufacture, 2007.

d
[30] R. Zeggari, R. Yahiaoui, J. Malapert, J.-F. Manceau, Design and fabrication of a new two-

we
dimensional pneumatic micro-conveyor, Sensors and Actuators A-physical - SENSOR
ACTUATOR A-PHYS 164 (2010) 125–130.

[31] M. Costanzo, D. H. Myers, Diverting conveyor with bidirectional assist roller, 2015. US
Patent 9,156,629.

ie
[32] S. H. Mayer, Development of a completely decentralized control system for modular
continuous conveyors, volume 73, KIT Scientific Publishing, 2009.

[33] L. Kleczewski, Weighing and sorting roller belt conveyor and associated method, 2020.

ev
US Patent App. 16/640,445.

[34] L. Overmeyer, K. Ventz, S. F. andTobias Krühn, Interfaced multidirectional small-scaled

modules for intralogistics operations, Logistics Research (2010).

r
[35] R. A. Sanghai, P. P. Saundalkar, J. A. Mallick, B. Shah, Sort X Consignment Sorter
using an Omnidirectional Wheel Array for the Logistics Industry, in: 2020 International
er
Conference on Convergence to Digital World - Quo Vadis (ICCDW), IEEE, 2020.

[36] A. Bhattacharjee, Y. Lu, A. T. Becker, M. Kim, Magnetically Controlled Modular Cubes

With Reconfigurable Self-Assembly and Disassembly, IEEE Transactions on Robotics
pe
(2021) 1–13.

[37] T. A. T. Dang, M. Bosch-Mauchand, N. Arora, C. Prelle, J. Daaboul, Electromagnetic

modular Smart Surface architecture and control in a microfactory context, Computers
in Industry 81 (2016) 152–170.
ot

[38] W. Huayu, Y. Shunsuke, T. Shuji, Fabrication of Multi-Axis Moving Coil Type Electro-
magnetic Micro-Actuator Using Parylene Beams for Pure In-Plane Motion, in: 2021 21st
International Conference on Solid-State Sensors, Actuators and Microsystems (Trans-
tn

ducers), 2021, pp. 667–670.

[39] S. D. Tisnés, L. Petit, C. Prelle, F. Lamarque, Modeling and Experimental Validation of

a Planar Microconveyor Based on a 2 × 2 Array of Digital Electromagnetic Actuators,
IEEE/ASME Transactions on Mechatronics 26 (2021) 1422–1432.
rin

[40] T. Piatkowski, M. Wolski, K. Dylag, Angular positioning of the objects by the system of
two oblique friction force fields, Mechanism and Machine Theory 140 (2019) 668–685.

[41] T. Piatkowski, M. Wolski, Model of positioning objects by the system of oblique friction
ep

force fields on horizontal and vertically offset planes, Mechanism and Machine Theory
156 (2021) 104155.

[42] Omnitrack, Omnitrack catalogue, Technical Report, 2022.

Pr

[43] V. C., Adams methodologycontact modeling, Benelux ADAMS User Meeting (2012).

[44] J. Giesbers, CONTACT MECHANICS IN MSC ADAMS A technical evaluation of the

contact models in multibody dynamics software MSC Adams, Bachelor thesis, Faculty
of Engineering TechnologyApplied Mechanics, 2012.

26
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4255407