ECEASST-NETSYS2021-camera-ready
Electronic Communications of the EASST
Volume 080 (2021)
Guest Editors: Andreas Blenk, Mathias Fischer, Stefan Fischer, Horst Hellbrück, Oliver Hohlfeld,
Andreas Kassler, Koojana Kuladinithi, Winfried Lamersdorf, Olaf Landsiedel, Andreas Timm-
Giel, Alexey Vinel
ECEASST Home Page: http://www.easst.org/eceasst/ ISSN 1863-2122
Conference on Networked Systems 2021
(NetSys 2021)
Impact of radio channel characteristics on the longitudinal behaviour of
truck platoons in critical car-following situations
Salil Sharma, Ehab Al-Khannaq, Raphael Riebl, Wouter Schakel,
Peter Knoppers, Alexander Verbraeck and Hans van Lint
7 Pages
ECEASST
2 / 8 Volume 080 (2021)
Impact of radio channel characteristics on the longitudinal behaviour
of truck platoons in critical car-following situations
Salil Sharma1, Ehab Al-Khannaq2, Raphael Riebl3, Wouter Schakel4,
Peter Knoppers5, Alexander Verbraeck6 and Hans van Lint7
1S.Sharma-4@tudelft.nl
Delft University of Technology, Delft, The Netherlands
2ehabsoa@gmail.com
Van Oord, Rotterdam, The Netherlands
3Raphael.Riebl@carissma.eu
Technische Hochschule Ingolstadt, Germany
4w.j.schakel@tudelft.nl
5P.Knoppers@tudelft.nl
6a.verbraeck@tudelft.nl
7J.W.C.vanLint@tudelft.nl
Delft University of Technology, Delft, The Netherlands
Abstract: Truck platooning is an application of cooperative adaptive cruise control
(CACC) which relies on vehicle-to-vehicle communications facilitated by vehicle ad-hoc
networks. Communication uncertainties can affect the performance of a CACC controller.
Previous research has not considered the full spectrum of possible car-following scenarios
needed to understand how the longitudinal behaviour of truck platoons would be affected
by changes in the communication network. In this paper, we investigate the impact of radio
channel parameters on the string stability and collision avoidance capabilities of a CACC
controller governing the longitudinal behaviour of truck platoons in a majority of critical
car-following situations. We develop and use a novel, sophisticated and open-source
VANET simulator OTS-Artery, which brings microscopic traffic simulation, network
simulation, and psychological concepts in a single environment, for our investigations. Our
results indicate that string stability and safety of truck platoons are mostly affected in car-
following situations where truck platoons accelerate from the standstill to the maximum
speed and decelerate from the maximum speed down to the standstill. The findings suggest
that string stability can be improved by increasing transmission power and lowering receiver
sensitivity. However, the safety of truck platoons seems to be sensitive to the choice of the
path loos model.
Keywords: Truck platoons, cooperative adaptive cruise control, V2V, VANET, radio
channel,
Impact of radio channel characteristics on truck platoons
NetSys 2021 3 / 8
1 Introduction
Truck platooning is a promising technology that is expected to generate fuel savings, emission
reduction, and safer operations. It is an application of cooperative adaptive cruise control
(CACC), where multiple trucks are organized into a group of close-following vehicles. The key
technology behind CACC is vehicle-to-vehicle (V2V) communication facilitated by vehicular
ad-hoc networks (VANETs). It enables participating vehicles to exchange relevant information
(e.g., position, speed, acceleration) with each other over a self-organized wireless
communication network [1]. Previous research has shown that uncertainties in V2V
communications, arising from highly dynamic conditions of the wireless channel, can
significantly impact the performance of a CACC controller [2,3]. However, these studies only
look at one of the following critical car-following situations: stop-and-go [2] and deceleration
to a slower speed [3]. Since trucks are heavy and long vehicles, it is vital to verify the
performance of a CACC controller governing their longitudinal behaviour (i.e., string stability
and collision-avoidance capabilities) in a majority of critical car-following situations that might
arise in real traffic conditions.
Rapidly changing vehicular channels rely on a physical layer to enable the exchange of
information. However, previous research has simplified the modelling of the physical layer
regarding radio channels [4]. Therefore, a detailed sensitivity analysis is required to understand
the impacts of the radio channel characteristics on the string stability and collision-avoidance
capabilities of trucks in a platoon. In this regard, simulation-based investigations of VANET
applications have gained momentum due to their cost-effectiveness and are critical to configure
the operations of VANET systems [4]. This paper presents OTS-Artery, a novel and open-source
VANET simulator, which brings microscopic traffic simulation (OpenTrafficSim (OTS) [5]),
network simulation (Artery [6]), and psychological concepts [7] in a single environment to
simulate next-generation traffic operations.
Consequently, the aim of this paper is to investigate the impact of radio channel
characteristics on the longitudinal behaviour of truck platoons in critical car-following situations
using a VANET simulator. The rest of the paper is structured as follows: The CACC controller
for truck platoons is presented in section 2. The OTS-Artery VANET simulator is outlined in
section 3. The experimental setup including communication and traffic scenarios is discussed in
section 4 and its results are explained in section 5.
2 CACC Controller for Truck Platoons
We use a modified version of the CACC controller developed by Faber et al. [8] in this paper
by removing redundancy. An ego truck is a CACC-equipped truck. Let 𝑟, 𝑣 and𝑟standstill be its
current headway spacing from its predecessor, its current speed, and minimum spacing at
standstill (3 m), respectively. Let 𝑣* and𝑎* be speed and acceleration of the predecessor of the
ego truck, respectively. 𝑟CACC
safe is the safe following distance required for the ego truck. The target
time gap 𝑡CACC is set to 0.5 s.
𝑟CACC
safe = 𝑡CACC ⋅ 𝑣 + 𝑟standstill (1)
ECEASST
4 / 8 Volume 080 (2021)
The ego truck responds to the acceleration of the predecessor, deviations between its desired
and current speed, and the deviation between the current distance headway and the desired
headway (see Equation 2).
𝑎CACC = 𝑘5 ⋅ 𝑎* + 𝑘6 ⋅ 7𝑣* − 𝑣9 + 𝑘: ⋅ 7𝑟 − 𝑟CACC
safe 9 (2)
where 𝑎CACC is bounded by the ego truck’s minimum (-3 m/s2) and maximum acceleration
(2 m/s2) capabilities. 𝑘, 𝑘5, 𝑘:, and 𝑘6 are chosen as 0.3, 1.0, 0.1, and 0.58, respectively to
provide a smooth acceleration response by minimizing overshoot and oscillations.
3 OTS-Artery
OTS-Artery is a novel and open-source VANET simulator. OTS incorporates human factors and
social interactions in a microscopic simulation framework to model driving behavior. Whereas
ARTERY is able to simulate the ETSI ITS-G5 protocol stack used in European VANETs. OTS-
Artery couples the OTS with Artery via the Sim0MQ middleware that is based on the high-
performance asynchronous ZeroMQ library. To simulate truck platoons, Artery is extended with
a GtuProxyService that uses the plain single-hop broadcast mode of GeoNetworking. Each truck
in Artery is equipped with the GtuProxyService which facilitates the exchange of relevant
information between a leader-follower pair of the truck platoon. As shown in Fig. 1, the leader
transmits the payload (sender GTU id, receiver GTU id, signal strength, time stamp, speed, and
acceleration) to the Artery OTS core. The follower then calls the core upon message reception
and receives the payload if the follower id matches the receiver GTU id in the payload. A new
payload is transmitted every 100 ms. More information about OTS-Artery is available at:
https://github.com/salilrsharma/OTS-Artery.
4 Experimental Setup
4.1 Communication Scenarios
1. Ideal: The V2V communication is handled idealistically within the traffic simulator OTS.
2. Realistic: A default radio channel is created based on IEEE 802.11p with OTS-Artery.
The fixed and variable parameters of OMNET++ are presented in Table 1 where bold-
faced values refer to the default channel. Transmission power (TP), receiver sensitivity
(RS), and type of path loss model are varied to assess the impact of radio channel
characteristics on the string stability and collision avoidance capabilities of truck platoons.
Figure 1: OTS-Artery architecture showing information exchange between a leader-
follower pair
Impact of radio channel characteristics on truck platoons
NetSys 2021 5 / 8
Table 1: OMNET++ parameters for radio channel based on IEEE 802.11p
Fixed parameters
Channel number 180
Carrier Frequency 5.9 GHz
Receiver Energy Detection -85 dBm
Receiver SNIR Threshold 4 dB
Variable parameters for sensitivity analysis
Transmission power (TP) {50 mW, 200 mW, 2000 mW}
Receiver Sensitivity (RS) {-85 dBm, -96 dBm}
Path-loss model {VanetNakagamiFading [9], Two-ray interference }
4.2 Traffic Scenarios
The platoon topology consists of five identical trucks moving on a single-lane straight road
section devoid of any other traffic. We consider the following three critical car-following
situations.
1. Stop-and-go: The platoon has an initial speed of 80 km/h (i.e., maximum allowed speed).
The leader decides to decelerate at 10 s with 3 m/s2 for the next 2 s. Then, the leader
accelerates with 2 m/s2 from 12-14 s. From 14 s onwards, the leader moves with a constant
speed of 72.80 km/h.
2. Acceleration from zero to the maximum speed: The platoon has an initial speed of 0 km/h.
The leader accelerates with 1 m/s2 to reach the final maximum speed of 80 km/h at 22.22
s.
3. Decelerating from the maximum speed to the standstill: The platoon has an initial speed
of 80 km/h. The leader decelerates with 1 m/s2 to reach the standstill situation at 22.22 s.
5 Results
In the OTS scenario, the CACC controller results in a string stable behaviour where acceleration
errors do not amplify when propagated upstream (see Table 2). This is attributed to the fact that
communications are robust and do not include any delay. In the OTS-Artery scenario, we set up
a more realistic radio channel to handle V2V communications. For situations involving stop-
and-go, we observe a delayed yet string stable acceleration response from the follower trucks in
the platoon.
When we change the values of RS, TP, and type of path-loss model, the acceleration
responses observed in the stop-and-go situation are similar to that of the default radio channel.
When the leader of a platoon accelerates from the standstill to the maximum speed, the
acceleration of followers significantly fluctuates and might be fuel-inefficient. These
fluctuations are reduced if we decrease the RS and increase the TP. Over and undershoots are
observed; however, these do not amplify upstream. Large under and overshoots which stabilize
until 80 seconds are observed when the path-loss model is changed to a deterministic one.
Similarly, when the leader of a platoon decelerates to standstill, string stability is affected only
in the case of the two-ray interference path loss model. Consequently, safety also gets worsened
as the last two trucks of the platoon collide with each other in the same scenario.
ECEASST
6 / 8 Volume 080 (2021)
Table 2: Acceleration profiles of truck platoons for traffic and communication scenarios
Communication
scenarios
Traffic scenarios
Stop-and-go Accelerating to the
max speed
Decelerating to
standstill
Ideal
(OTS)
Realistic
(default radio
channel)
Realistic
(RS = -96 dBm)
Realistic
(TP = 50 mW)
Realistic
(TP = 2000 mW)
Realistic
(Path loss model =
Two-ray interference)
Impact of radio channel characteristics on truck platoons
NetSys 2021 7 / 8
6 Conclusions
In this paper, the impact of radio channel characteristics on the longitudinal behaviour of truck
platoons in critical car-following situations is studied. Our main conclusions are as follows.
1. An open-source simulator OTS-Artery is developed to evaluate VANET applications.
2. Short-term changes in the acceleration behaviour of the followers of a platoon (e.g., stop-
and-go) are not significantly affected by the radio channel characteristics.
3. Situations involving truck platoons to accelerate to the maximum speed or decelerate
down to the standstill require most of the attention while assessing the impact of radio
channel characteristics and designing a robust CACC controller.
4. Path loss models can significantly affect safety in a truck platoon configuration.
A promising research direction can be to test the impact of the radio channel in other critical
car-following situations such as emergency braking and gap-closing. Further, the impact of
competing V2X services can also be studied. Another possibility is to test the impact of radio
channels on the lane-changing behaviour of truck platoons. Afterward, large-scale VANET
simulations can be conducted to assess the traffic and safety impacts of truck platoons on
surrounding traffic in real traffic situations.
Acknowledgements: This work was supported by the Dutch Research Council (NWO), TKI
Dinalog, Commit2data, Port of Rotterdam, SmartPort, Portbase, TLN, Deltalinqs,
Rijkswaterstaat, and TNO under the project “ToGRIP-Grip on Freight Trips".
Bibliography
[1]. Zhao, C., Duan, X., Cai, L., & Cheng, P. (2020). Vehicle Platooning With Non-Ideal
Communication Networks. IEEE Transactions on Vehicular Technology, 70(1), 18-32.
[2]. Ramyar, S., & Homaifar, A. (2019, October). Impact of Unreliable Communication on
String Stability of Cooperative Adaptive Cruise Control. In 2019 IEEE International
Conference on Systems, Man and Cybernetics (SMC) (pp. 4263-4268). IEEE.
[3]. Lei, C., Van Eenennaam, E. M., Wolterink, W. K., Karagiannis, G., Heijenk, G., & Ploeg,
J. (2011, August). Impact of packet loss on CACC string stability performance. In 2011 11th
International Conference on ITS Telecommunications (pp. 381-386). IEEE.
[4]. Gozalvez, J., Sepulcre, M., & Bauza, R. (2012). Impact of the radio channel modelling on
the performance of VANET communication protocols. Telecommunication Systems, 50(3),
149-167.
[5]. Van Lint, H., Schakel, W., Tamminga, G., Knoppers, P., & Verbraeck, A. (2016). Getting
the Human Factor into Traffic Flow Models: New Open-Source Design to Simulate Next
Generation of Traffic Operations. Transportation Research Record, 2561(1), 25–33.
ECEASST
8 / 8 Volume 080 (2021)
[6]. Riebl, R., Günther, H. J., Facchi, C., & Wolf, L. (2015, June). Artery: Extending veins for
VANET applications. In 2015 International Conference on Models and Technologies for
Intelligent Transportation Systems (MT-ITS) (pp. 450-456). IEEE.
[7]. Fuller, R. (2000) “The task-capability interface model of the driving process”, Recherche –
Transports – Sécurité 66, pp. 47-57.
[8]. Faber, T., Sharma, S., Snelder, M., Klunder, G., Tavasszy, L., & Van Lint, H. (2020).
Evaluating traffic efficiency and safety by varying truck platoon characteristics in a critical
traffic situation. Transportation research record, 2674(10), 525-547.
[9]. Cheng, L., Henty, B. E., Stancil, D. D., Bai, F., & Mudalige, P. (2007). Mobile vehicle-to-
vehicle narrow-band channel measurement and characterization of the 5.9 GHz dedicated
short range communication (DSRC) frequency band. IEEE Journal on Selected Areas in
Communications, 25(8).