Optimum Approach Velocity for Twin Merging of Autonomous POD Vehicles

Abstract

Autonomous vehicles (AV) have gained ground in recent years. However, they still use the principles of traditional vehicles in terms of design and operation. This work proposes an adaptive transportation system based on autonomous POD vehicles, and investigates a major aspect of its operation. The PODs used in the proposed system can be considered a variant version of existing autonomous PODs. However, their unique design and concept of operation enable them to operate more efficiently than existing PODs. The proposed system involves docking and undocking of these PODs based on passengers’ demands. However, during the merging process, undesired collisions could happen due to unforeseen conditions. If the approach speed is high enough, it could induce damage in to the vehicles. This work investigates some possible scenarios of the potential collisions that could happen between these PODs during the merging process. Based on these scenarios, the allowed safe approach speeds are determined. These speeds can help in designing the operation of the proposed transportation system. Some of the variables considered in this work include; type of body material, shell thickness, impact speed, stress, deformation, and absorbed energy. The safe design merging speeds have been determined under different conditions.

Full text

Optimum Approach Velocity for Twin Merging
of Autonomous POD Vehicles
Osama M. Al-Habahbeh and Romil S. Al-Adwan
Mechatronics Engineering Department, School of Engineering, the University of Jordan, Amman, Jordan
Email: o.habahbeh@ju.edu.jo, romil.aladwan@ju.edu.jo
Abstract—Autonomous vehicles (AV) have gained ground in
recent years. However, they still use the principles of
traditional vehicles in terms of design and operation. This
work proposes an adaptive transportation system based on
autonomous POD vehicles, and investigates a major aspect of
its operation. The PODs used in the proposed system can be
considered a variant version of existing autonomous PODs.
However, their unique design and concept of operation
enable them to operate more efficiently than existing PODs.
The proposed system involves docking and undocking of
these PODs based on passengers’ demands. However, during
the merging process, undesired collisions could happen due
to unforeseen conditions. If the approach speed is high
enough, it could induce damage to the vehicles. This work
investigates some possible scenarios of the potential collisions
that could happen between these PODs during the merging
process. Based on these scenarios, the allowed safe approach
speeds are determined. These speeds can help in designing the
operation of the proposed transportation system. Some of the
variables considered in this work include; type of body
material, shell thickness, impact speed, stress, deformation,
and absorbed energy. The safe design merging speeds have
been determined under different conditions.
Index Terms— Urban transportation, POD modular vehicle,
Autonomous vehicle, Vehicle merging, Approach velocity,
FEM analysis, Vehicle impact
I. INTRODUCTION
In recent years, what was considered science fiction in
the eighties has become a reality. The advent of
autonomous systems is gaining momentum and passenger
vehicles are no exception. Autonomous vehicles (AVs) are
now disrupting the transportation industry. Many of these
vehicles have the same size as regular passenger cars,
while some of them have only one seat. The latter category
is often referred to as POD. Researchers have worked to
advance the design of AVs, such as He et al. [1] who
proposed a novel emergency steering control strategy
based on hierarchical control architecture consisting of a
decision-making layer and motion control layer. A novel
motion controller for automated vehicles is presented by
Xia et al. [2] which offers smaller steady-state errors and
faster convergence speed.
The idea of merging vehicles was considered by
researchers and designers. Merging vehicles and
separating them enables upsizing and downsizing of the
vehicle so as to accommodate a given number of
Manuscript received April 28, 2022; revised October 22, 2022.
passengers. However, most probably, there will still be one
or more vacant seats; this work proposes a solution to this
issue which will be discussed later. Another benefit of
merging vehicles is saving energy; imagine two vehicles
driving comfortably on a level or down-sloped road. If
these vehicles are joined, then one of their engines can be
switched off. One more benefit of merging vehicles is that
all passengers will reach the destination at the same time.
Moreover, once the passengers are onboard, they can
interact face-to-face. Some patents [3-9] have shown that
PODs can merge and separate during their operation in
response to demand. Operating Strategies for a new
modular electric autonomous vehicle were presented by
Ulrich et al. [10]; the vehicle consists of a drive unit and
an interchangeable capsule. However, even though it is a
sound idea, the capsule replacement process requires extra
equipment, which makes it impractical. A platooning-
control strategy for a fleet of Autonomous and Connected
Trucks (ACTs) was developed by Gungor et al. [11]; this
is an example of connecting cars for the purpose of
reaching the destination simultaneously as well as
reducing energy consumption. An experimental study of
Denial of Service (DoS) attacks against Platoon of smart
vehicles was carried out by Malik et al. [12].
During the vehicles’ merging process, minor collisions
could happen. Vehicle crashworthiness have received
considerable attention in the literature. An extensive
literature survey pertaining to the topic of crash box was
conducted by Abdullah et al. [13]. Front rails were
designed by Li et al. [14] to improve the crash performance
of vehicle and reduce its structural mass using finite
element analysis. A collaborative optimization process
using optimal Latin hypercube design and response surface
methodology was proposed by Liu et al. [15] to improve
the vehicle crashworthiness in the frontal impacts.
Munyazikwiye et al. [16] investigated whether a simple
piecewise Lumped Parameter Model can serve as an
accurate crash modeling tool. Yu et al. [17] applied a series
of tailor rolled blank (TRB) structures to the front-end
components of pure electric vehicle (PEV) for the design
optimization of vehicle crashworthiness and lightweight.
A multi-objective design approach with accelerated
methodology was developed by Oztürk et al. [18] for a B-
pillar (side door pillar) in which the intrusion velocity was
decreased and the crash energy absorbed. The feasibility
840
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res
doi: 10.18178/ijmerr.11.11.840-849

of designing a honeycomb-like crash-box, as a cellular
structure, was analyzed by Saenz-Dominguez et al. [19]. A
multi-material design for a vehicle body considering both
crashworthiness safety and social effects was developed by
Chen et al. [20]. A hybrid design approach was introduced
by Yusof et al. [21] to develop a conceptual design of oil
palm polymer composite automotive crash box. Jin et al.
[22] hypothesized that occupants will be better protected
by using rotational seat to alter the occupant’s orientation
in accordance with the direction of impact. A data-driven
artificial intelligence (AI) inverse problem solution for
traffic collision reconstruction was successfully developed
by Chen et al. [23].
Some studies have investigated impact problems for a
variety of applications. They can provide insights for a
better understanding of impact behavior. For example, the
impact responses and residual properties of thin-walled
carbon fiber-reinforced plastics tubes and aluminum (Al)
tubes subjected to multiple axial impacts were explored by
Liu et al. [24]. Öztürk et al. [25] evaluated the effects of
failure criteria of steel B-pillar on the accuracy of impact
simulation.
It is predicted that the deployment of AVs will decrease
traffic accidents, because human error, which is considered
the main reason of such accidents, will be eliminated.
Nonetheless, AVs will still be subjected to other factors
that may increase the possibility of getting involved in
accidents. Some of these factors are due to vehicle’s design,
such as mechanical failures and sensor malfunction, while
others are due to environmental factors, such as wind and
frequency interference. Therefore, the study of
crashworthiness of AV is equally important to their man-
operated counterparts. An example of these studies is the
work done by Zong et al. [26], who modeled impacts and
driving characteristics of multiple AVs and RVs vehicles.
A collision avoidance/mitigation system (CAMS) was
proposed by Lee and Kum [27] to rapidly evaluate risks
associated with all surrounding vehicles. Sequence of
events data extracted from California automated vehicle
(AV) collision reports were used by Song et al. [28] to
investigate patterns and how they may be used to develop
AV test scenarios. The characteristics and patterns of
crashes involving connected and autonomous vehicles
(CAV) were investigated by Xu et al. [29]. Adaptive Stress
Testing (AST) in conjunction with encoding domain
relevant information into the search procedure was
implemented by Corso et al. [30] to identify useful failure
scenarios of AV. The mechanism of AV-involved crashes
was explored by Chen et al. [31]; they analyzed the impact
of each feature on crash severity. Walker et al. [32] called
the meteorological community to action and proactive
engagement with the transportation community to enhance
the safety of AV.
During vehicles’ collision, the type of material of the
vehicle’s body plays a major role in the impact response.
An example of past work studying this issue is the testing
of composite materials with metallic reinforcements under
dynamic axial loading by Dlugosch et al. [33]. A material
model normally used for modelling fiber-reinforced
plastics was adopted by Müller et al. [34] to generate a
material database for three hardwood species.
The current situation of autonomous vehicles is that
most of them have multiple seats. However, single-seat
POD vehicles are gaining a growing popularity due to their
low energy consumption. On the other hand, the concept
of modular AV is attracting more attention due the
developments in communication technology and
electronic processors. These modular AVs can join each
other to form new bigger vehicles. They can also separate
from each other in order to satisfy passengers’ demand. In
order to conduct the docking operation safely, it should be
done at a reasonable speed. If the speed exceeds a certain
limit, the resulting impact will be harmful to the vehicle’s
body. The exact value of the speed depends on the type of
vehicle’s body material. A fleet of modular AVs represents
an adaptive transportation system. The modular AV
considered in this work is a single-seat POD, as shown in
Fig. 1. This work will focus on determining the optimum
docking speed for two PODs in Twin configuration, as
shown in Fig. 1. The possible reasons for impact during
docking include wind, misalignment, sensor error,
frequency interference etc. In order to perform the analysis,
Finite element method (FEM) is used to simulate the Twin
POD docking operation. The results of this work can help
to design safer and more efficient operation of modular
autonomous POD vehicles. Merging the two POD vehicles
enables the following benefits:
a) There will be no vacant seats, which is reflected in
smaller vehicle size. This will save both energy and
road space.
b) In level and down-sloped roads, one of the motors
can be switched-off so as to save energy.
c) The passengers will reach the destination at the
same time.
d) The passengers can interact face-to-face, unlike
virtual communication in other methods.
The proposed docking process can be done quickly,
with minimal energy, while no extra equipment is needed.
When the PODs merge normally, there is no collision, and
the two matching hitching ports (as shown in Fig. 1) will
join each other. However, without the determination of the
proper speed, either higher or lower speed will be used. If
lower speed is used, it will cause delays and inefficiencies
in the process. On the other hand, if higher speed is used,
it will cause damage in the structure of the vehicles. The
safe docking speed is determined based on stress failure
criterion of the POD’s material. Other considered variables
include shell thickness and energy absorbed. Different
values of merging velocities are simulated and safe values
are determined based on von misses stress failure criteria.
The operation of modular AVs was investigated by
many researchers. However, most related work focused on
platooning of vehicles, which means they are not
physically connected as in this work, but they drive in one
formulation. For examples; Liu et al. [35] optimized bus
platooning to ensure dynamic capacity. A macro network
scheduling model for electric modular vehicles for public
transportation was proposed by Liu et al. [36]; where
several modular cars form a formation. A bus consists of a
841
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

power module vehicle and a van module vehicle, and the
van module vehicle needs to be towed by the power
module vehicle to move. Chinmay et al. [37] explored the
need to integrate the Network model and platooning
system of Connected and Autonomous Vehicles (CAV)
for highway environments.
Another aspect that has received considerable attention
is planning the operation of modular AVs; Tian et al. [38]
determined the optimal planning of transit services timing
with modular vehicles. Lamotte et al. [39] found that
allocating road capacity to bookable autonomous vehicles
can reduce congestion costs. Ji et al. [40] proposed
a strategy for flexible Modular Autonomous Vehicle
(MAV) scheduling on transit routes to meet the time-
varying passenger demand. Chen et al. [41] investigated
the joint design of dispatch headway and vehicle capacity
for one to one shuttle systems with oversaturated traffic to
achieve the optimal tradeoff between general vehicle
dispatching cost and customer waiting cost. A variable-
capacity operation approach with modular transits for
shared-use corridors was proposed by Shi et al. [42], in
which both dispatch headway and vehicle capacity are
decision variables. It is noted that the above researchers
have focused on scheduling and timing of operation rather
than physical docking of vehicles.
Adaptive transportation systems rely on a fleet of
modular AVs that are responsive to passengers’ demand.
A direct transit network dispatch model for public
transport EVs was developed by Liu et al. [43]. It enables
scalable transportation capacity in order to meet
changeable travel demands. They investigated two modes
of operation; Intelligent connected mode and Traction
mode. Tholen et al. [44] optimized the capacity of on-
demand modules of passenger and parcel compartments
onboard shared autonomous vehicle (SAV) used in urban
transportation. Fielbaum [45] studied a feeder system that
operates on-demand in a local zone. He showed that the
autonomous vehicle technology (AVT) encourages larger
fleets of smaller vehicles that follow more direct routes.
When compared with the traditional technology (TT), the
total savings induced by the AVT reach up to one third of
TT’s costs. A modular transit network system (MTNS)
concept is proposed by Pei et al. [46] to overcome the
mismatch between fixed vehicle capacity and spatially
varying travel demand in traditional public transportation
systems. In this work, the proposed PODs transportation
system operates based on travel demand, similar to the
above previous works. However, this work looks more
closely at the joining process and performs a dynamic
analysis to better plan the PODs’ docking operation in
terms of design, material and velocity.
II. METHODOLOGY AND SIMULATION MODEL
In order to simulate the docking process of the two POD
vehicles, two FEM models of these POD vehicles must be
built and simulated; the two autonomous PODs to be
joined as a twin are shown in Fig. 1. The width of the POD
is 1 m, the height is 1.5 m, and the length is 3 m. Each POD
can comfortably accommodate a single passenger along
with his belongings or shopping. The merging process will
be conducted by joining the hitching ports of both vehicles,
as shown in Fig. 1. This vehicle is part of the proposed
smart transportation system where the passenger orders a
ride using a mobile app. A central control room finds the
vehicle that is nearest to the customer and commands it to
drive to the customer’s location. The vehicle does exactly
that, picks up the customer, drives the customer to his
destination then drops him off at the planned location.
After that the vehicle will wait for new instructions or
probably the instructions could have been received while
the passenger was still onboard.
Figure 1. Front view of the two Autonomous PODs.
If two passengers order a vehicle, a corresponding
number of PODs will drive to the passengers’ location and
merge together to form a twin vehicle, as shown in Fig. 2.
The merging process will be done autonomously; when the
two PODs want to merge in Twin configuration; one of
them will be parking while the other one approaches it
laterally. The important parameters in this operation
include the approach velocity during merger, and the angle
of alignment. If the approach velocity is small, the merging
process will be slow, and after accumulating the lost times
over the whole fleet and over a long period of operation,
the loss of time (and earnings) will be high. On the other
hand, if the approach velocity is high, the merging impact
could cause damage in the PODs. Therefore, there is a
need to determine an optimum value of approach velocity,
where it is fast enough to maximize profit, while being
slow enough to maintain safety. The back view of the two
Autonomous PODs is shown in Fig. 2.
Figure 2. Back view of the two Autonomous PODs.
The geometric models of the POD vehicles shown in Fig.
1 and Fig. 2 will be used to simulate the merging process
using FEM. If the merging process is performed at the right
speed, no problem will occur. However, if the speed is too
high, an impact will take place during merging. The body
of the vehicle can be manufactured from different
materials including metal alloys and composites. Various
thicknesses of the POD’s outer body will be investigated.
Another possible variable is the impact angle. Different
842
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

impact velocities will be investigated in order to find the
optimum approach velocity. In order to assess the possible
damage due to impact, a failure criterion must be adopted.
It will depend on the type of material used. The simulation
process is performed using ANSYS™ software; where the
procedure starts with developing geometric models of the
two collided vehicles, as shown in Fig. 1 and Fig. 2. After
that a FEM mesh of the two collided vehicles is generated.
The collision loads are then applied and stresses and
deformations are determined. Applying the principle of
work and energy on the two PODs, we can write the energy
balance during collision as shown in (1), [47].
󰇛 󰇜    󰇛 󰇜 (1)
where T1 and T2 are the kinetic energies of the first and
second PODs, respectively. The subscripts i and f denote
the initial state before collision and the final state after
collision. U denotes the work done during collision, which
equals the force in eqn. (1) times the deformation.
Applying the impulse momentum theorem, we can write
the corresponding equation as:
󰇛 󰇜

  󰇛 󰇜 (2)
where m1 and m2 are the masses of POD1 and POD2,
respectively, while v1 and v2 are the speeds of POD1 and
POD2, respectively. Note that the impulse term represents
the force of the moving POD’s motor. Internal forces are
reciprocated between the two PODs during collision. As a
result of these forces, stress develops in the adjacent
structures of the vehicles’ bodies. As mentioned earlier,
different materials of the vehicles’ bodies are investigated.
The properties of these materials are shown in Table I.
III. RESULTS AND DISCUSSION
The FEM models of the two PODs are used to simulate
the merging scenario into twin configuration. After
merging, the two vehicles will form a new bigger vehicle
as shown in Fig. 3. The ideal case is when the merging is
done accurately in terms of approach angle; that means the
two merging PODs are perfectly parallel before and during
attachment. In this case the impact angle equals zero, as
shown in Fig. 3. Two other possible cases are investigated;
they include impact angles of 30° and -30°. The resulting
stresses due to the preceding cases are plotted in Fig. 4.
The results in Fig. 4 correspond to a POD’s body thickness
of 1 mm, impact velocity of 7 km/h and impact Angle (IA)
of 0°, 30° and -30°. The POD’s body material used in the
simulations is Steel 4340, with a Yield Strength (Y) of 470
MPa. In Fig. 4, it is noted that the most critical load is
realized at IA= -30°. Therefore, all subsequent simulations
will be done using this angle.
A. Impact Angle (IA)=0°
This case is shown in Fig. 3. As a result of the merging
process, impact load will be distributed over the
corresponding sides of the two PODs. Fig. 3 shows the
directional deformation resulting from this load. In Fig. 4,
it is shown that at an IA=0, the stress is generally minimum,
as compared to the other two cases. Therefore, no further
investigation of this case is required. On the other hand,
the behavior of the impact energy during the PODs
attachment is shown in Fig. 5. It is shown that the internal
energy increases during the impact then decreases
afterwards, while the kinetic energy decreases during the
impact then decrease afterwards. As for the hourglass and
contact energies, they both stay constant during the impact.
TABLE I. MATERIAL PROPERTIES
No. Material
Yield Strength (MPa)
1 Steel 4340 470
2 AL 1060-H12 61
3
Composite, Epoxy glass fiber
440
4
Plastic, ABS high impact
27.4
Figure 3. FEM model of the two merging PODs.
Figure 4. Stresses for different impact angles.
Figure 5. Impact energy during the PODs attachment.
200
400
600
800
1000
0 1 2 3
Stress (MPa)
Time (ms)
IA= 0° IA= 30° IA= -30°
843
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

B. Impact Angle (IA)=30°
This case is shown in Fig. 6 where the two PODs are in
the process of merging, but due to certain condition such
as wind, road level, navigational error, etc. at least one of
them have shifted position. In this case an angle of 30° is
assumed, where the first contact happens in the front part
of the body side. As shown in Fig. 4, this case is generally
more critical than the perfectly parallel PODs. However, it
is clearly less critical than the case when IA= -30°.
Therefore, the latter case will be pursued further.
Figure 6. The two PODs merging at impact angle = 30°.
C. Impact Angle (IA)=-30°
Figure 7. The two PODs merging at impact angle = -30°.
This case is shown in Fig. 7. As mentioned earlier, it is
considered the most critical case because it results in the
highest stress as shown in Fig. 4. For this case, acceleration
versus time is plotted in Fig. 8. The POD’s body thickness
used in the analysis is 1 mm. the impact velocity is 8 km/h
and the body material used is Steel 4340, with yield
strength (Y) of 470 MPa. In Fig. 8, the deceleration shows
a high value at the beginning of the impact, then it
decreases later on. The deformation during the PODs’
merging is shown in Fig. 9; the simulation conditions
include; body thickness of 1 mm and impact velocity of 8
km/h. The body material used in this case is Steel 4340,
with Yield strength (Y) of 470 MPa. It is noted that the
deformation is faster at the beginning of the impact, then it
slows down as the impact progresses.
Figure 8. Acceleration during Pods’ merging.
Figure 9. Deformation during Pods’ merging.
The exerted force during the collision is shown in Fig.
10. The plot reflects the following conditions; body
thickness of 1 mm and impact velocity of 8 km/h using
Steel 4340 as the body material. It is noted that the force is
maximum at the beginning of the impact. Fig. 11 shows
the energy variation during the collision. The conditions
include; body thickness of 1 mm and impact velocity of 8
km/h, using Steel 4340 as the body material. It is noted that
the impact energy is highest at the beginning of the impact.
The Stress- Strain curve of the impact is shown in Fig. 12.
The tested body thickness is 1 mm made of Steel 4340 and
the impact velocity is 8 km/h. The upper part of the curve
represents loading and the lower part represents unloading,
with less stress. It is noted that most of the strain bounces
back, which means that the deformation is mostly elastic.
Fig. 13 shows the variation of stress with impact velocity
for steel 4340 with thickness of 1 mm. It is noted that the
relationship is linear and proportional. The variation of
equivalent stress with panel thickness is shown in Fig. 14;
where the impact velocity is 8 km/h and the material is
steel 4340. As expected, the relationship is proportional.
Bearing in mind that the stress load is calculated based on
Equivalent von Mises failure criterion, which is defined as
[48]:
󰇟
   
󰇠 (3)
where σI and σII are the principal stresses. It is typical to
see behaviors such as that shown in Fig. 15, and that is due
to the changing values of the principal stresses in each
loading scenario.
-160
-150
-140
-130
-120
-110
-100
0 1 2 3 4
Acceleration (g)
Time (ms)
0
2
4
6
8
10
12
14
01234
Deformation (mm)
Time (ms)
844
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

D. Other Body Panel Materials
So far, all the analyses have been conducted based on
one material, which is steel 4340. However, there are other
important materials which are increasingly being used for
car body panels. In addition, some types of materials are
investigated to check if they provide better performance
than existing ones. The investigated body panel materials
include AL 1060-H12 (Y=61 MPa), Composite, Epoxy
glass fiber (Y=440 MPa) and Plastic, ABS high impact
(Y=27.4 MPa). The results for these materials are
presented in figures 15 to 21, based on an IA of -30° and
body panel thickness of 1 mm; for AL 1060-H12, the stress
variation with impact velocity is plotted in Fig. 15. It is
noted that the curve is not linear as in the steel case. The
variation of deformation with velocity is plotted in Fig. 16,
where it exhibits a linear relationship. For composite,
epoxy glass fiber, the variation of stress with velocity is
shown in Fig. 17, which also exhibits a linear relationship.
The variation of the deformation with velocity is shown in
Fig. 18, which exhibits a linear relationship as well. Finally,
for Plastic, ABS high impact, Fig. 19 shows the variation
of stress with velocity and Fig. 20 shows the variation of
deformation with velocity, where both figures exhibit
linear relationships.
Figure 10. Force vs. Time (Steel 4340).
Figure 11.
Energy vs. Time (Steel 4340).
Figure 12.
Stress vs. Strain (Steel 4340).
Figure 13.
Stress vs. Impact Velocity (Steel 4340).
Figure 14.
Stress vs. Thickness of body panel (Steel 4340).
Figure 15.
Stress vs. Impact velocity (AL 1060-H12).
0
20
40
60
80
01234
Force (kN)
Time (ms)
0.07
0.08
0.09
0.1
0.11
0.12
0.13
01234
Energy (kJ)
Time (ms)
300
400
500
600
700
800
900
0.0025 0.0035 0.0045 0.0055 0.0065
Stress (MPa)
Strain (mm/mm)
0
100
200
300
400
500
600
700
0246810
Stress (MPa)
Velocity (km/hr.)
600
800
1000
1200
1400
0.5 1 1.5 2 2.5
Stress (MPa)
Thickness of body panel (mm)
0
50
100
150
200
02468
Stress (MPa)
Velocity (km/h)
Loading
Unloading
845
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

Figure 16. Deformation vs. Impact velocity (AL 1060-H12).
Figure 17. Stress vs. Impact velocity (Composite, epoxy glass fiber).
Figure 18. Deformation vs. Impact velocity (Composite, epoxy glass
fiber).
Figure 19. Stress vs. Impact velocity (Plastic, ABS high impact).
Figure 20. Deformation vs. Impact velocity (Plastic, ABS high impact).
The values of safe merging speeds for different body
panel materials are shown in Table II and Fig. 21. The
maximum safe merging speed can be obtained using
Composite, Epoxy glass fiber as the body panel material,
followed by Plastic, ABS high impact. As for the steel
4340, it comes in the third place. On the other hand, using
AL 1060-H12 as the body panel material can only supports
a merging speed of 2.5 km/h, which is the lowest in the list.
Of course, additional materials can be investigated, as well
as different alloys of the same metals, which may exhibit
different behavior. It should be noted that it is not
suggested here that composites or plastics are stronger than
steel. Because the speeds that are tested are considered low
and do not represent typical accidents. Beside the
equivalent stress failure criteria that was employed in this
work, other failure criteria for the body panel materials can
be investigated, such as Hashin failure criteria [49], which
states that Epoxy glass fiber composite can be damaged at
an impact energy of 12 J. Moreover, Gohel et al. [50]
reported that a 300 G acceleration can be considered a
failure criteria for ABS high impact plastic. Furthermore,
the effect of possible repetitive impacts on stiffness was
studied by Kim and Cho [51].
TABLE II. MATERIAL PROPERTIES
No.
Body Panel Material
Safe Merging Speed (km/h)
1 Steel 4340 6
2 AL 1060-H12 2.5
3
Composite, Epoxy glass fiber
32
4
Plastic, ABS high impact
12
For validation purposes, the current results are
compared with past results from the literature and good
agreement was found, as shown in Table III. On the other
hand, the effects of the different variables considered in
this work on the PODs joining process are summarized in
Table IV.
Figure 21. Safe PODs merging speeds for different body panel
materials.
0
0.2
0.4
0.6
0.8
0 2 4 6 8
Deformation (mm)
Velocity (km/h)
0
100
200
300
400
500
010 20 30 40
Stress (MPa)
Velocity (km/h)
0
1
2
3
4
5
010 20 30 40
Deformation (mm)
Velocity (km/h)
0
10
20
30
40
0 5 10 15 20
Stress (MPa)
Velocity (km/h)
846
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

TABLE III. RESULTS VALIDATION
Material Past Experiments/Unsafe
Impact
This work/
Safe Impact
Steel Alloy
Speed of
18 km/h [52]
Speed
of 6 km/h
Aluminum Alloy
Load of 90 kN [53] Load of 58 kN
Composite Deformation of 27 mm
[54] Deformation of
4.5 mm
Plastic S
peed of 16 km/h
[55]
Speed
of 12 km/h
TABLE IV. VARIABLES EFFECTS ON THE JOINING PROCESS
Variable Effect
Body material The best material for this application is
Composite, followed by Plastic, then Steel, while
the worst is Aluminum.
Shell thickness A higher thickness is better for this application.
However, there must be a trade between cost and
performance.
Impact speed The speeds mentioned in Table II should not be
exceeded.
Stress All stresses should be maintained below the
levels mentioned in Table I.
Deformation Maximum deformation is noted in Plastic,
followed by Composite, then Aluminum, and
finally Steel.
Absorbed energy The more energy absorbed, the safer the merging
process. The best energy absorbent is Composite,
followed by Plastic, then Steel, and the worst is
Aluminum.
IV. CONCLUSIONS AND FUTURE WORK
The results indicate that it can be safe for two POD
vehicles to merge together without causing any damage,
even if the joining process was not perfect in terms of
alignment. Due to the body design of the POD, the results
showed that the most limiting misalignment angle during
merger is -30, followed by 30, whereas perfect parallel
docking results in the highest permitted joining speed.
Therefore, to be on the conservative side, the most limiting
angle of -30 is used for all subsequent simulations. Based
on this fact, ceilings for merging speeds for different body
panel materials have been suggested. These values can be
elaborated further to cater for the exact need of a specific
POD design and material. The use of epoxy glass fiber as
the body panel material yielded the maximum allowable
merging speed, followed by ABS high impact plastic. On
the other hand, traditional metal alloys such as steel 4340
allowed less merging speeds, especially for AL 1060-H12
which showed intolerance to merging accidental impact.
In general the procedure presented in this work has yielded
reasonable results and can help to develop flexible
autonomous transportation systems that can be adaptable
to passengers’ demand. As a recommendation for future
work, alternative failure criteria for the body panel
materials can be explored; for example, Epoxy glass fiber
composites can be damaged at a certain level of impact
energy. On the other hand, high values of acceleration can
be used as a failure criteria for ABS high impact plastics.
Furthermore, the effect of repetitive impacts on the vehicle
strength may need further assessment.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
AUTHORS CONTRIBUTIONS
Osama M. Al-Habahbeh conducted the research and
drafted the manuscript; Romil S. Al-Adwan analyzed the
data and did the formatting and editing. Both authors had
approved the final version.
REFERENCES
[1] X. He, Y. Liu, C. Lv, X. Ji, and Y. Liu, “Emergency steering control
of autonomous vehicle for collision avoidance and stabilization,”
Vehicle System Dynamics, vol. 57, no. 8, pp. 1163-1187, 2019.
[2] H. Xia, J. Chen, Z. Liu, and F. Lan, “Coordinated motion control for
automated vehicles considering steering and driving force
saturations,” Transactions of the Institute of Measurement and
Control, vol. 42, no. 1, pp. 157-166, 2020.
[3] Priev, “Vehicles for variably sized transportation,” US Patent
10,882,575, Jan 5, 2021.
[4] P. R. Williams, P. Kessler, and L. Herold, “Transportation system,”
US Patent 10359783B2, 2019.
[5] Bick et al., “Vehicle combination and method for forming and
operating a vehicle combination,” US Patent 9,740,213, Aug 22,
2017.
[6] Han et al., “Electronic docking vehicle,” US Patent 9,914,495, Mar
13, 2018.
[7] Evans, “Intelligent POD management and transport,” US Patent
10,150,524, Dec 11, 2018.
[8] Cervantes et al., “Road vehicle convoy control method, and road
vehicle convoy,” US Patent 10,538,240, Jan 21, 2020.
[9] Wright, “Self-powered actively steerable converter dollies for long
combination vehicles,” US Patent 10,518,831, Dec 31, 2019.
[10] C. Ulrich, H. E. Friedrich, J. Weimer, and S. A. Schmid, “New
operating strategies for an on-the-road modular, electric and
autonomous vehicle concept in urban transportation,” World
Electric Vehicle Journal, vol. 10, no. 91, 2019.
[11] O. E. Gungor, R. She, I. L. Al-Qadi, and Y. Ouyang, “One for all:
Decentralized optimization of lateral position of autonomous trucks
in a platoon to improve roadway infrastructure sustainability,”
Transportation Research Part C: Emerging Technologies, vol. 120,
Art. no. 102783, 2020.
[12] S. Malik, P. Bandi, and W. Sun, “An experimental study of denial
of service attack against platoon of smart vehicles,” in Proc. 4th
International Conference on Connected and Autonomous Driving,
MetroCAD, Detroit, Apr. 2021, Code 171216, pp. 23-30.
[13] N. A. Z. Abdullah, M. S. M. Sani, M. S. Salwani, and N. A. Husain,
“A review on crashworthiness studies of crash box structure,” Thin-
Walled Structures, vol. 153, Art. no. 106795, 2020.
[14] Q. Q. Li, E. Li, T. Chen, L. Wu, G. Q. Wang, and Z. C. He,
“Improve the frontal crashworthiness of vehicle through
the design of front rail,” Thin-Walled Structures, vol. 162, Art. no.
107588, 2021.
[15] X. Liu, R. Liang, Y. Hu, X. Tang, C. Bastien, and R. Zhang,
“Collaborative optimization of vehicle crashworthiness under
frontal impacts based on displacement oriented structure,” Int. J.
Automotive Technology, vol. 22, no. 5, pp. 1319-1335, 2021.
[16] B. B. Munyazikwiye, D. Vysochinskiy, M. Khadyko, and K. G.
Robbersmyr, “Prediction of vehicle crashworthiness parameters
using piecewise lumped parameters and finite element models,”
Designs, vol. 2, no. 4, Art. no. 43, pp. 1-16, 2018.
[17] L. Yu, X. Gu, L. Qian, P. Jiang, W. Wang, and M. Yu, “Application
of tailor rolled blanks in optimum design of pure
electric vehicle crashworthiness and lightweight,” Thin-Walled
Structures, vol. 161, Art. no. 107410, 2021.
[18] I. Oztürk, N. Kaya, and F. Oztürk, “Design of vehicle parts under
impact loading using a multi-objective design approach,” Materials
Testing, vol. 60, no. 5, pp. 501-509, 2018.
[19] I. Saenz-Dominguez, I. Tena, A. Esnaola, M. Sarrionandia, J. Torre,
and J. Aurrekoetxea, “Design and characterisation of cellular
composite structures for automotive crash-boxes manufactured by
out of die ultraviolet cured pultrusion,” Composites Part B:
Engineering, vol. 160, pp. 217-224, 2019.
[20] Y. Chen, X. Cheng, and K. Fu, “Multi-material design of a vehicle
body considering crashworthiness safety and social effects,” Int. J.
Crashworthiness, vol. 25, no. 5, pp. 517-526, 2020.
847
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

[21] N. S. B. Yusof, S. M. Sapuan, M. T. H. Sultan, and M. Jawaid,
“Conceptual design of oil palm fibre reinforced polymer hybrid
composite automotive crash box using integrated approach,” J.
Central South University, vol. 27, no. 1, pp. 64-75, 2020.
[22] X. Jin, H. Hou, M. Shen, H. Wu, and K. H. Yang, “Occupant
kinematics and biomechanics with rotatable seat in autonomous
vehicle collision: A preliminary concept and strategy,” in Proc.
Conf. Proc. International Research Council on the Biomechanics of
Injury, IRCOBI, Athens, Sep. 2018, Code 143981, pp. 106-113.
[23] Q. Chen, Y. Xie, Y. Ao, T. Li, G. Chen, S. Ren, C. Wang, and S. Li,
“A deep neural network inverse solution to recover pre-crash
impact data of car collisions,” Transportation Research Part C:
Emerging Technologies, vol. 126, Art. no. 103009, 2021.
[24] Q. Liu, H. Shen, Y. Wu, Z. Xia, J. Fang, and Q. Li, “Crash responses
under multiple impacts and residual properties of CFRP and
aluminum tubes,” Composite Structures, vol. 194, pp. 87-103, 2018.
[25] I. Öztürk, N. Kaya, and F. Öztürk, “Effects of material failure
criteria on design of vehicle parts under impact loading,” Int. J.
Crashworthiness, 2020.
[26] F. Zong, M. Wang, J. Tang, and M. Zeng, “Modeling AVs & RVs’
car-following behavior by considering impacts of multiple
surrounding vehicles and driving characteristics,” Physica A:
Statistical Mechanics and its Applications, vol. 5891, Art. no.
126625, 2022.
[27] K. Lee and D. Kum, “Collision avoidance/mitigation system:
Motion planning of autonomous vehicle via predictive occupancy
map,” IEEE Access, vol. 7, Art. no. 8698763, pp. 52846-52857,
2019.
[28] Y. Song, M. V. Chitturi, and D. A. Noyce, “Automated vehicle
crash sequences: Patterns and potential uses in safety testing,”
Accident Analysis and Prevention, vol. 153, Art. no. 106017, 2021.
[29] C. Xu, Z. Ding, C. Wang, and Z. Li, “Statistical analysis of the
patterns and characteristics of connected and autonomous vehicle
involved crashes,” J. Safety Research, vol. 71, pp. 41-47, 2019.
[30] A. Corso, P. Du, K. Driggs-Campbell, and M. J. Kochenderfer,
“Adaptive stress testing with reward augmentation for autonomous
vehicle validation,” IEEE Intelligent Transportation Systems
Conference ITSC, Auckland, Oct. 2019, Art. no. 8917242, pp. 163-
168.
[31] H. Chen, R. Zhou, Z. Liu, and X. Sun, “Exploring the mechanism
of crashes with AV using machine learning,” Mathematical
Problems in Engineering, Art. no. 5524356, 2021.
[32] C. L. Walker, B. Boyce, C. P. Albrecht, and A. Siems-Anderson,
“Will weather dampen self-driving vehicles?” Bulletin of the
American Meteorological Society, vol. 101, no. 11, pp. E1914-
E1923, 2020.
[33] M. Dlugosch, J. Fritsch, D. Lukaszewicz, and S. Hiermaier,
“Experimental investigation and evaluation of numerical modeling
approaches for hybrid-FRP-steel sections under impact loading for
the application in automotive crash-structures,” Composite
Structures, vol. 174, pp. 338-347, 2017.
[34] U. Müller, T. Jost, C. Kurzböck, A. Stadlmann, W. Wagner, S.
Kirschbichler, G. Baumann, M. Pramreiter, and F. Feist, “Crash
simulation of wood and composite wood for future automotive
engineering,” Wood Material Science and Engineering, vol. 15, no.
5, 2020.
[35] X. Liu, X. Qu, and X. Ma, “Improving flex-route transit services
with modular autonomous vehicles,” Transportation Research Part
E: Logistics and Transportation Review, vol. 149, 102331, 2021.
[36] X. Liu, X. Ma, and Z. Liu, “Scheduling optimization of electric self-
driving modular vehicles for public transportation,” Chinese
Journal of Highways, vol. 35, no. 3, 2022.
[37] M. Chinmay, F. Hua, W. Honggang, and Y. Qing, “Adaptive
vehicle platooning with joint network-traffic approach,” IEEE
Global Communications Conference, GLOBECOM-Proceedings,
Madrid, 7-11 Dec. 2021, Code 176805.
[38] Q. Tian, Y. H. Lin, D. Z. W. Wang, and Y. Liu, “Planning for
modular-vehicle transit service system: Model formulation and
solution methods,” Transportation Research Part C: Emerging
Technologies, vol. 138, 103627, 2022.
[39] R. Lamotte, A. Palma, and N. Geroliminis, “On the use of
reservation-based autonomous vehicles for demand management,”
Transportation Research Part B: Methodological, vol. 99, pp. 205-
227, 2017.
[40] Y. Ji, B. Liu, Y. Shen, and Y. Du, “Scheduling strategy for transit
routes with modular autonomous vehicles,” International Journal
of Transportation Science and Technology, vol. 10, no. 2, pp. 121-
135, 2021.
[41] Z. Chen, X. Li, and X. Zhou, “Operational design for shuttle
systems with modular vehicles under oversaturated traffic:
Continuous modeling method,” Transportation Research Part B:
Methodological, vol. 132, pp. 76-100, 2020.
[42] X. Shi, Z. Chen, M. Pei, and X. Li, “Variable-capacity operations
with modular transits for shared-use corridors,” Transportation
Research Record, vol. 2674, no. 9, pp. 230-244, 2020.
[43] X. Liu, X. Ma, and Z. Liu, “Dispatch optimization of electric
autonomous modular vehicles for public transport,” Zhongguo
Gonglu Xuebao/China Journal of Highway and Transport, vol. 35,
no. 3, pp. 240-248, 2022.
[44] M. Tholen, B. A. Beirigo, J. Jovanova, and F. Schulte, “The share-
A-Ride problem with integrated routing and design decisions: The
case of mixed-purpose shared autonomous vehicles,” Lecture Notes
in Computer Science, vol. 13004, LNCS, pp. 347-361, 2021.
[45] A. Fielbaum, “Strategic public transport design using autonomous
vehicles and other new technologies,” Int. J. ITS Res., vol. 18, pp.
183-191, 2020.
[46] M. Pei, P. Lin, J. Du, X. Li, and Z. Chen, “Vehicle dispatching in
modular transit networks: A mixed-integer nonlinear programming
model,” Transportation Research Part E: Logistics and
Transportation Review, vol. 147, Art. no. 102240, 2021.
[47] R. C. Hibbeler, Engineering Mechanics- Dynamics, 14th ed., New
Jersey 07030, USA: Pearson Prentice Hall, 2016.
[48] F. A. Leckie and D. J. Dal Bello, Strength and Stiffness of
Engineering Systems, Mechanical Engineering Series, NY 10013,
USA: Springer Science, 2009.
[49] N. A. Nassir and M. R. Gharkan, “Impact response of composite
laminates based on epoxy and glass fibre,” in Proc. 3rd
International Conference on Materials Engineering & Science,
Material Today: Proceedings, vol. 42, part 5, pp. 1901-1907, 2021.
[50] G. Gohel, S. K. Bhudolia, S. B. S. Elisetty, K. F. Leong, and P.
Gerard, “Development and impact characterization of acrylic
thermoplastic composite bicycle helmet shell with improved safety
and performance,” Composite Part B, vol. 221, 2021.
[51] S. Kim and H. Cho, "Interpretation of stiffness change over the
trunk side of a vehicle upon a re-accident after repair its former rear
impact collision accident," International Journal of Mechanical
Engineering and Robotics Research, vol. 10, no. 7, pp. 406-413,
2021.
[52] A. T. Zehnder and A. J. Rosakis, “Dynamic fracture initiation and
propagation in 4340 steel under impact loading,” International
Journal of Fracture, vol. 43, no. 4, pp. 271-285, 1990.
[53] C. Chen, H. Zhang, X. Ren, and J. Wu, “Investigation of flat-
clinching process using various thicknesses aluminum alloy sheets,”
International Journal of Advanced Manufacturing Technology, vol.
114, no. 7-8, pp. 2075-2084, 2021.
[54] Y. Chen, X. Zeng, Y. Deng, and G. Wei, “Investigation on
manufacturing and low-velocity impact performance of all-
composite sandwich structure with S-type foldcore,” Composite
Structures, vol. 29015, Art. no. 115539. 2022.
[55] X. Wang, X. Tang, X. Hu, and P. Wang, “Dynamic tensile
mechanical behavior of polyurethane films incorporated with
functionalized graphene oxide,” Polymer, vol. 24619, Art.
no. 124755, 2022.
Copyright © 2022 by the authors. This is an open access article
distributed under the Creative Commons Attribution License (CC BY-
NC-ND 4.0), which permits use, distribution and reproduction in any
medium, provided that the article is properly cited, the use is non-
commercial and no modifications or adaptations are made.
Osama M. Al-Habahbeh is an Associate
Professor and Head of the Mechatronics
Engineering Department at the University of Jordan.
He earned his PhD in Mechanical Engineering
from Clarkson University at New York. He was a
Deputy Dean of students’ Affairs and Director of
Career Guidance and Alumni Office. He published
2 books and 40 Journal and conference papers. His
research interests include reliability, energy, and
autonomous systems. Dr. Al-Habahbeh has long
years of experience in aerospace, manufacturing,
tourism, and academia. He held technical and leading positions at different
848
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res

organizations in Jordan, Malaysia, Australia, Kuwait, and United States.
Currently, he provides consulting for reputable regional clients.
Romil S. Al-Adwan earned his PhD in
Mechanical Engineering from the University
of Northumbria at Newcastle, UK. He is
currently teaching at the department of
mechatronics engineering, at the University of
Jordan. He also taught several industrial and
mechanical engineering courses, at different
universities. Dr. Al-Adwan has more than 30
years of experience in various sectors
including defense, telecom, manufacturing,
and academia. He held technical and managerial positions at different
international organizations in Jordan,
Canada, Hong Kong, and United
States.
849
International Journal of Mechanical Engineering and Robotics Research Vol. 11, No. 11, November 2022
© 2022 Int. J. Mech. Eng. Rob. Res