Content uploaded by Diego R. Llanos
Author content
All content in this area was uploaded by Diego R. Llanos on Mar 02, 2021
Content may be subject to copyright.
Towards a rational use of loading and unloading areas in
urban environments
Daniel Barbaa, Sergio Garcia-Villanuevaa, Hector Del-Campo Pardoa, Juan A. Marchb, and
Diego R. Llanosa,b
aDpto. Inform´atica, Universidad de Valladolid, Valladolid, Spain
bRDNest, Valladolid, Spain
ABSTRACT
Despite the efforts of the authorities, that promote the use of alternative transportation systems, the traffic still
increases in European cities, leading not only to traffic jams but also to pollution episodes. Delivery vehicles are
part of both problems, because of their intensive use, the advent of e-commerce, the limited number and sizes
of loading and unloading areas in many ancient European cities, and the difficulties associated to keep track of
the correct use of these spaces.
In this work we propose an holistic solution to the management of delivery vehicles in urban environments.
Our solution, called RYDER, is based on the use of BLE (Bluetooth Low Energy) devices that should be provided
by the local authority to delivery vehicles, as part of their authorization to use the loading and unloading
areas. With the help of low-cost, low-power antennas with Bluetooth and 4G capabilities installed next to
each loading/unloading area, the authorities are able to know in real time (a) the use of these areas by delivery
vehicles, (b) the paths of the vehicles while they travel across the city, (c) the time spent in each area by each one
of them, and (d) with the help of a mobile/tablet App, the local Police can check in seconds the permissions of
each vehicle using these public spaces. Moreover, the use of a GIS-based platform allows the Traffic Department
to track online each particular vehicle, based on the loading/unloading spaces being used, and to infer the most
representative paths they follow, an information that may guide the decision about where these spaces are really
necessary and whether each particular vehicle follows their associated usage rules.
The deployment of RYDER low-cost antennas can also serve for other purposes, such as to track the routes
followed by public loan bicycles, or by other fleets of public vehicles. With the help of low-cost sensors, antennas
can also return an estimation of pollution values, such as levels of ozone, particulate matter, carbon monoxide,
sulfur dioxide, and nitrous oxide, among others. This information may in turn drive the installation of certified
pollution detectors.
Keywords: delivery vehicles, loading/unloading areas, low-cost sensoring, sensor deployment, Bluetooth Low
Energy, beacons, pollution, smart cities, transportation, vehicles
1. INTRODUCTION
Pollution episodes and traffic jams in many European city centers are considered normal nowadays. Although
the use of bicycles is common in many Central Europe cities and towns, and their popularity is increasing in
other regions, the gases generated by internal combustion vehicles forces the authorities to keep a close look to
pollution levels. In turn, this leads to the installation of complex and expensive pollution measurement stations,
devices that also need a noticeable space and that do not fit well in the city centers of many ancient European
cities.
There are four main groups of internal combustion vehicles in our streets: private-owned cars, public transport
vehicles, such as buses and taxis, service vehicles, such as those dedicated to waste collections, and delivery
vehicles. Private users may be encouraged to use public transport systems, that are gradually switching to non-
polluting engines; waste collection trucks usually belongs to local administrations or subcontracting companies,
For both technological and business requests, please contact Dr. Diego R. Llanos, RDNest, PCUVa Paseo Bel´en 15,
47011 Valladolid, Spain. e-mail: diego@rdnest.com, phone: +34 983 185 642.
so it is also possible to gradually replace them with trucks that incorporate cleaner engines. However, it is very
little what the local authorities can do with respect to delivery vehicles, apart from obligate them to undergo
periodic inspections. In fact, local authorities do not usually know much about them. Very little is know about
their routes inside the city. Classic systems used to measure traffic, such as metal detectors used to measure
the amount of traffic, can not distinguish between private and delivery vehicles. The use of surveillance cameras
may help to discriminate between them, but it is still difficult to track a particular vehicle during its delivery
itinerary. The advent of e-commerce and the limited number and sizes of loading and unloading areas in many
ancient European cities, together with the difficulties associated to keep track of the correct use of these spaces,
makes the daily movement of delivery vehicles a source of both pollution and traffic jams.
Having more information on the routes followed by delivery vehicles can help the local authorities in several
ways. First of all, the number and size of loading and unloading areas may be properly adjusted to their effective
use. Nowadays, these areas are designed based on guess, but there are not figures that support these decisions.
A second advantage of knowing the main routes followed by delivery vehicles is that it can help the authorities
to better schedule traffic cuts for repairing works, since they will have the historic information needed to disturb
traffic as little as possible. This is not a minor advantage: While private drivers may reach their destination to
certain areas of the city center by feet or using public transportation, many goods should be delivered on time
to supermarkets, pharmacies and shops. The commercial and social implications of disturbing delivery traffic is
a major concern to local authorities.
There exists many systems to keep track of fleets of vehicles based on GPS, such as Tookan,1Orbcomm,2or
Amac.3By installing a GPS receiver with GPRS/3G/4G capabilities, the fleet manager can keep track all the
company vehicles online, representing their movements on an interactive map, usually web-based, and accessing
to their movement statistics. While being an excellent solution to manage fleets, GPS-based systems are of little
use for the authorities to keep track of delivery vehicles, due to several reasons. First, the cost of the devices that
should be installed in each particular vehicle are in the range of several hundred euros. Second, these devices
need a communication link to transmit their particular position, and these communications impose additional
costs. Third, the GPS coverage in many European cities are far from optimal, due to the existence of many
narrow streets surrounded by buildings, making GPS-based solutions to greatly reduce their accuracy during
potentially long time intervals.
In this work we propose an holistic solution to tracing the behavior of delivery vehicles in urban environments.
Our solution, called RYDER, is composed of four main elements:
•Low-cost, low-energy Bluetooth4beacons, that are provided by the authorities as part of their authorization
granted to delivery vehicles to use the city loading and unloading areas, authorization that the drivers should
expose in their vehicles’ dashboard.
•Low-cost Bluetooth receivers, based on Raspberry Pi 3.5These receivers are installed in each loading and
unloading area to be monitorized. Additional receivers can be installed in strategic junctions of the city, to
better track the delivery vehicles in their daily route. These receivers transmit all the gathered information
to a cloud-based service.
•A cloud-based service receives the information collected by the antennas, processes it and delivers all types
of statistical information to the local authorities, both in text and web-based form, including GIS-based
positioning in real time.
•An app for smartphones and tablets, called Fisher, that allows the Police agents to check whether the
vehicles in a 30-meters range effectively owns the authorization cards and beacons they show in their
dashboard.
With our solution, the authorities are able to know both in real time and from an historical perspective a
vast amount of information regarding delivery vehicles, such as their use of loading and unloading areas, the
paths of the vehicles while they travel across the city, the time spent in each area by each one of them, and, since
the local authorities know the model and age of each vehicle being authorized, an estimation of the accumulated
weight supported by each street, an information that can be used to make preventive maintenance. Moreover,
with the help of our smartphone/tablet app, the local Police can check in seconds the permissions of each vehicle
using these public spaces.
One of the main advantages of this solution is that the technology involved is relatively inexpensive: BLE
emitters cost are falling, and many DIY platforms such as Raspberry Pi 3 comes with Bluetooth and WiFi
connectivity.
In this paper we will examine in detail the solution design space to the problem of vehicle detection, discussing
the main advantages and drawbacks for each alternative. We will then describe our proposed solution in detail,
enumerating its functionalities. We will also discuss the possibility of using these antennas to also measure
pollution, thanks to the use of ozone, particulate matter, carbon monoxide, sulfur dioxide, and nitrous oxide
solid-state sensors. Other sensors to measure light and noise pollution are also simple to add, making these
antennas the all-in-one solution for smart cities sensing.
The rest of the paper is organized as follows. Section 2discusses the technological alternatives to GPS that
may be applied to solve this problem. Section 3describes in detail the hardware and software involved in the
development of the RYDER project. Section 4examines the possibilities for local authorities that the deployment
of the RYDER project offers. Finally, Sect. 5concludes our work.
2. TRACKING VEHICLES: TECHNOLOGICAL ALTERNATIVES
As we stated in the previous section, GPS-based solutions are not applicable to the problem we aim to solve:
GPS receivers are expensive; they need additional hardware for communicating the information outside the
vehicle; someone should pay for the communication costs associated to each vehicle; and GPS coverage is far
from optimal in the center of many ancient European cities.
Designed as an alternative to GPS, Local Positioning System (LPS)6do not offer a global coverage. Instead,
they cover a relatively small area, delimited by the range of the receivers being used. These receivers are installed
in known positions, and, with the help of appropriate algorithms, they allow to locate an object or a person in
a local coordinates system.
During the last years, different LPS techniques have been proposed, such as Infrared, Ultrasounds, WIFI,
RFID, or Bluetooth. Recently, some new alternatives have appeared, such as the use of the Ultra-Wideband
(UWB) technology. Each one of these alternatives offers a solution with unique characteristics in terms of range
covered, precision, sensibility to obstacles, and cost, making them suitable for specific use cases. A general
solution, that allows positioning in real time, with good precision, fast response to movements, good range of
operation and inexpensive has not appeared in the market yet.
There exist different works describing the state of the art in the field of LPS and Indoor Positioning, such
as.7,8These works center the discussion in four main alternatives: RFID, Ultra-Wideband, Bluetooth, and
Ultrasound, analyzing how they work and they advantages and drawbacks of they deployment in real scenarios.
The literature in this field is abundant, and some of these technologies have been used in the field of LPS for
years. In this section we will examine the subset of solutions that may give an answer to the particular restrictions
of our problem, that can be summarized as:
•Real-time positioning, with a precision in the range of a few meters.
•Deployment in uncontrolled environments, with movement of people and objects.
•Positioning with minimal participation of the person or object being tracked.
•Low-cost solutions, to allow their economical viability.
2.1 RFID: Radio Frequency Identification
RFID technology9,10 has a long history that can be traced to World War II, when the Allies started using
radar systems. To let the radar system distinguish between friends and enemies, Allied pilots rolled their planes
when returned to base, thus changing the radio signal reflected back. Nowadays, passive RFID technologies are
based on the use of small tags, composed by a chip and an antenna. Their most well-known use is to read the
information stored in them. RFID systems come in two flavors:
•Passive RFID: Uses simpler tags, that are activated thanks to the energy transferred by the RFID receiver
using radio frequency. For this reason, they operational range is severely limited (about three meters).
•Active RFID: Tags have their own batteries, increasing their operational range (up to 100 meters).
Although RFID tags are very cheap (particularly passive RFID tags), this technology present several disad-
vantages. First of all, RFID readers are expensive. Second. RFID reads have a strong directional component,
making necessary to install more readers for an acceptable coverage. For these reasons, the positioning systems
based on this technology are better suited to detect their transit under a RFID reader than to use them for
real-time positioning.
With respect to our problem, RFID systems have the advantage of the low cost associated to their tags.
However, the need of expensive, directional receivers makes this solution unfeasible for the problem of local
positioning of vehicles in the streets.
2.2 UWB: Ultra-Wideband
Ultra-Wideband (UWB) technology11,12 is relatively new. It consists in sending wide-band pulses. A UWB-
based positioning system consists on the use of a set of fixed stations and a number of UWB emitters. The
system works as follows:
1. The emitter sends a first message.
2. The receiver gets this message and sends an answer.
3. The emitter receives the answer and sends a second message.
4. The receiver gets the second message and calculates the distance by using the time difference between both
messages received.
UWB is a promising technology for LPS, showing several advantages, such as a relatively low energy con-
sumption, relative insensitivity to obstacles, high data rate transfer thanks to the use of a wide band, and an error
margin in the order of centimeters. However, this technology has also some drawbacks. The first one is its cost,
still high to allow large-scale deployment. A second problem is that UWB signals interfere with GPS systems, as
well as with systems that works in the frequency ranges of 250-750MHz, 3.2-4.7GHz, and 5.9-10.2GHz. A third
drawback is that it needs the use of specialized receivers: Low-cost computers and smartphones are not directly
capable to interact with them.
For cases when the interferences with GPS systems were acceptable, and if deployment costs continue to fall,
and if smartphones were able to act as receivers, this solution would allow an economically feasible solution with
an excellent precision. However, this is not still the case for the problem described in this paper.
2.3 Ultrasounds
Ultrasound-based solutions13,14 can be used to local positioning and tracking of persons and goods. This technol-
ogy is sometimes complemented with the use of RFID technologies for synchronization tasks. Its main advantage
is that is a low-cost technology.
There are three different ultrasound-based LPS. The first one is known as bat. The people or goods that are
wanted to be localized carry an emitter that is activated by receiving an external signal. A network of receivers
get the answer, thus allowing to determine the position of the emitter. The second alternative, called cricket,
uses emitters in fixed locations that send ultrasounds to receptors carried by the people and goods to be tracked.
Finally, the dolphin system is a symmetric solution, where all devices are equal, and some of them are fixed in
known positions. The exchange of ultrasounds allow the system to estimate the position of the mobile devices.
The main disadvantages of this technology are that signals can not overcome obstacles such as walls; re-
fractions severely affects the measurements; and the identification of the particular object being tracked is
problematic. These limitations make ultrasounds a non-acceptable solution for our problem.
2.4 Infrared
Infrared-based technologies have been extensively used for indoor positioning.15,16 These systems need direct vi-
sion between an infrared emitter and its corresponding detector. The positioning is calculated using triangulation
algorithms, so both the angle of incidence and the position of the fixed elements should be known.
As with other technologies, this system can be indistinctly used by emitters and receptors. In any case, the
need of direct vision from them limits its use to situations where this visual line can be maintained to carry
out the triangulation process. For the identification of the particular object being tracked, each object should
send an identifier. The relatively low bandwidth is related to how frequent is the data being sent, and it needs
adequate synchronization mechanisms.
Infrared devices can reach a very high precision, in the range of centimeters. However, their use in LPS poses
challenges that are difficult to overcome: The need of visual contact between devices, the deployment difficulties
associated to this fact, and the limited bandwidth for identification purposes.
2.5 Computer Vision
Computer vision systems can be used to develop a LPS system.17 The system is composed by a set of cameras
that captures images of the area to be covered. Images are sent to a remote server for processing, identifying the
position of the person or element to be tracked. To do so, the server should detect some identification marks.
The advantage of computer vision systems is that they are completely non-intrusive: They only need a set of
fixed cameras in known locations. However, it present some disadvantages, such as the need of computing power
to perform tracking at real time, the need of training to detect and identify certain objects, and the high cost of
the hardware and software involved.
With respect to our problem, the use of computer vision systems are widely use to recognize licence plates
in roads around the world. To use this technology for our purposes is somewhat problematic, because in narrow
streets we do not always have a clear view of the delivery vehicle in order to read their license plate. Second, such
a read would allow us to know that the vehicle was in that position at a certain time, but, unless we combine
the lecture of the license plate with sophisticated vision algorithms, it is hard to tell how long the vehicle was
parked in a loading/unloading area, where the licence plate is not longer visible. Third, the processing of the
captured images can be done either locally or remotely. Local processing requires expensive hardware. Remote
processing, on the other hand, requires high bandwidth to send the images to the central server, increasing the
communication costs. These reasons make computer vision solutions not optimal to track the behavior of vehicles
in urban environments.
2.6 Bluetooth Low Energy
Bluetooth Low Energy (BLE) devices are a type of WPAN (Wireless Personal Area Network) device, although
their operational range could reach 30 or 40 meters, at the expense of a reduction in the bidirectional capabilities
of information transference. For LPS systems this does not represent a problem. These devices operate in the
band of 2.4GHz and their energy consumption is particularly low, allowing long autonomies (in the order of
years) with a single coin-cell battery.
LPS based on BLE18,19 consist of a set of BLE emitters, called beacons, and a set of BLE receivers. These
systems allow two different usage strategies. The most common one is to deploy a set of beacons in fixed
places. These beacons sends data packets at fixed intervals between 0.1 to 5 seconds, usually using the iBeacon 20
protocol. A receiving device is in charge of reading these packets. Based on the perceived intensity of the signal
received, the latter can calculate the distance with respect to each beacon, allowing the calculation of its position.
This approach has several advantages:
•The computing needs are transferred to the receiving devices, who just need to take care of their own
position with respect to the beacons that send the incoming packets.
•The deployment complexity is greatly reduced.
•The receiving devices can be either low-cost, general-purpose computers, such as the Raspberry Pi platform,
or smartphones running an Android or iOS application.
However, this approach shows also some drawbacks:
•The precision is in the order of 3..5 meters, so this technology is not useful when the precision required is
greater.
•Its use require the elements to be located to carry a computer or a smartphone.
•The duty of calculate and communicate the localization is transferred to the user, a situation that can be
a problem in certain circumstances.
•This solution needs a computer or smartphone for each element to be localized. In scenarios where there
are more elements to be positioned than places to be covered, this is an issue in economic terms.
•While the autonomy of a beacon is measured in years, the autonomy of the receiver is just a few hours,
making this solution unfeasible when the element to be positioned does not have an energy source.
The second approach is the opposite one: Beacons are carried by the people or elements that should be
positioned, and the receiving computers are installed at fixed locations. This solution offer a solution to the
problems described above:
•The user or element to be positioned does not need to make any effort to keep the system running: They
just need to carry the beacon.
•Beacons are relatively inexpensive (less than 20 euros at the time of writing) so it is a feasible solution
when the number of elements to localize is greater than the places being covered.
•Beacons run during years with a single coin-cell battery, allowing the tracking of elements during long
periods of time.
•Fixed receiving systems, that need an external source power, can be connected to the electricity grid or be
fed with the help of small solar panels.
Figure 1. XtremeLoc (XLoc) beacons and receiving antenna. Beacons have a diameter of around 45mm.
Figure 1shows some beacons and a Raspberry-Pi receiving antenna developed by RDNest, a spin-off of the
University of Valladolid, Spain. These elements are part of the XtremeLoc (XLoc) solution developed by the
company, intended to offer LPS localization services for situations where GPS is not a feasible solution. As
indicated in the figure, beacons can be easily carried by the element to be tracked, while the receiving antennas,
build with IP protection, can be installed in either indoor or outdoor facilities. RDNest has also more discreet
antennas to be seamlessly installed in offices. The RYDER project consist of using XtremeLoc antennas and
beacons, together with specialized software, to monitorize the traffic of delivery vehicles in urban environments.
2.7 Why the use of LBE is the best solution to our problem
The latter solution described, with mobile elements carrying beacons and fixed receiving antennas, is particularly
well suited for the problem of tracking vehicles in urban environments, for the following reasons:
•The associated costs (less than 20 euros for beacons and a few hundred euros for antennas) makes the
solution economically feasible at different deployment scales.
•The precision of this solution is adequate for this problem, consisting in detecting vehicles within the areas
to be monitorized;
•The range of 30..40 meters is adequate for our needs, covering many loading/unloading areas with a single
antenna (more antennas can be installed on demand).
•Although precise positioning can be calculated by setting up at least three receiving antennas, it may
not even be necessary, since a single receiver per loading/unloading area suffices to detect the presence of
vehicles within range and their distance to the antenna.
•Since beacons emit a packet at intervals in the range of seconds, it is straightforward to know how much
time has the delivery vehicle spent in the loading/unloading area.
•The autonomy of beacons, that last 2..4 years with a single coin-cell battery, is adequate for tracking
delivery vehicles. When the beacons run out of battery, it can be easily replaced without losing the
beacons’ configuration. Moreover, the beacons continuously emit their percentage of battery left.
•Communication costs are modest: each receiving antenna incorporates a SIM card and transmit all the
gathered information to a central server at regular intervals.
•Identification of the particular vehicle is straightforward, since each beacon emits an unique identifier.
•Additional antennas may be located in important streets and junctions, to better keep track of the move-
ments of vehicles in the city.
All these reasons make LBE technology the best choice for tracking vehicles in urban environments. Moreover,
the same technology and the same antennas deployment can be used to track other vehicles as well, such as
bicycles that are part of a public loan system, or public service vehicles.
3. HARDWARE AND SOFTWARE INVOLVED
The solution developed by RDNest consists of four main elements:
1. A set of general-purpose beacons, incorporating the iBeacon protocol, adequately configured for this prob-
lem (with unique identifiers and transmitting power and frequency).
2. A set of low-cost receiving antennas to be installed in the areas to be covered, that send the collected
information to a cloud-based service.
3. A cloud-based service that keeps track of the elements and show the results in an online map, offering also
historical and statistical information.
4. A smartphone/tablet app, called Fisher, to help the Police agents to check whether all beacons in range
effectively belongs to the delivery vehicles they are installed into.
3.1 Receiving antennas
Antennas are based on the Raspberry Pi 3 platform,5with some additional hardware. This platform is a
complete, inexpensive computer system capable of running Linux and with extended connectivity, including not
only Bluetooth but WiFi and Ethernet. 3D/4D connectivity is possible by adding a SIM800L GPRS module
using the Raspberry Pi GPIO pins. The system incorporates a MicroSD card to store the operating system and
the dedicated software.
The antennas developed by RDNest run a tailored version of GNU/Linux, compiled entirely from scratch,
that consumes just 100MB. An exclusive, proprietary software solution minimizes the number of accesses to the
MicroSD card, performing an initial read at bootup and not accessing again except in very special circumstances.
This behavior greatly augments the lifetime of MicroSD cards, that are known to start degrading after a certain
number of read and write cycles. This behavior also makes the antennas extremely robust to unexpected energy
interruptions, since no data is stored locally, except when a momentary connectivity problem arises.
Software updates are also handled automatically. At bootup, the antennas register themselves in the system.
As part of the response, the cloud-based server checks if there is any software update to deploy to them, and
transmit it to the new antenna. The system guarantees that the new update is not installed in the antenna until
it has fully downloaded and its integrity has been checked. Time-based synchronization is also carried out at
bootup. All communications are carried out with SSL encryption.
At regular intervals, the antenna scans the Bluetooth spectrum, chooses the incoming packets that belong to
XtremeLoc beacons, and sends this information to the cloud-based server in JSON21 messages. Information sent
includes the identifier fields of each beacon, the measured intensity of its signal, a timestamp, and additional
information such as its percentage of battery left, a very useful data when several thousand beacons have been
deployed and many of them can stop working unexpectedly.
3.2 Cloud-based service
The cloud-based service is responsible of receiving and storing the information sent by the antennas, handling
their software upgrades, and return the information to the end user. The most simple information returned is the
real-time situation of the different loading/unloading areas (see Fig. 2), including the number of vehicles detected
in each one and their current capacity (with respect of the total number of vehicles that can use each area). The
color of each loading/unloading area (green, yellow, red) changes not only because of the number of vehicles,
but also taking into account their length with respect to the total space available in the loading/unloading area:
When authorizing a new delivery vehicle, the local authorities gathers the model of the vehicle being authorized,
so all its data (length, maximum authorized weight) is known by the system.
Besides the occupancy information, the system is also able to represent (pale blue boxes in Fig. 2) the number
of delivery vehicles detected in important junctions of the city center, thanks to the use of additional antennas.
These antennas have several purposes: to estimate the traffic of the corresponding junction due to these vehicles,
and to capture additional information on the routes followed by each vehicle.
Figure 2. Screenshot of basic GIS-based information on loading/unloading area current occupancy.
Figure 3. Screenshot of the behavior of a particular delivery vehicle during its delivery activities.
The system is not only intended to show the occupancy of loading/unloading areas. It also serves to study
the behavior, both individually and aggregated, of all the delivery vehicles that works in the area monitorized.
Figure 3) represents the route followed by a particular delivery vehicle during one morning, based on the packets
received by the antennas and their corresponding timestamps. Note that, although we may infer the particular
route followed by the vehicle using timestamps and the underlying cartography, we only can guarantee that the
vehicle was indeed present in the areas covered by our antennas. This is why we do not draw an inferred route
across the city. Colors indicate the amount of time spent in each area. As long as the maximum time the areas
can be used in our city is 20 minutes, we can color each visit according to the behavior of the vehicle with respect
to the local regulations.
The cloud-based server does not only return real-time information. It can be configured to obtain all kinds of
reports about the fleet of the authorized delivery vehicles in the city, and to infer the amount of delivery traffic
supported by each street in any time interval according to different parameters, such as the number of vehicles,
their size, and their estimated weight. We will return on this topic in Sect. 4.
Figure 4. The Fischer app quickly returns information about the vehicles associated with the beacons within range.
3.3 Inferring and measuring pollution
To measure pollution at the city center is not an easy task. Pollution counters are big, complex and expensive
systems that can not be seamlessly installed in the narrow streets of many ancient European cities.
The RYDER project can be also used to both infer and measure the pollution in the city center. Pollution
may be inferred by combining all the information regarding the delivery vehicles being tracked by the system.
The system knows the model of each vehicle, their age, and the time each one of them spent in different points
of the city. This information, adequately combined, may give an idea of the pollution that can be expected to
be suffered at particular times in different parts of the city.
However, it is always better to measure than to infer. The receiving antennas can be equipped with different
low-cost sensors. Magnitudes subject to measure include ozone, carbon dioxide, sulfur dioxide, nitrous oxide,
and suspended particles, together with temperature and humidity. All these measurements are transmitted by
the antennas in JSON messages, together with the information regarding the Bluetooth beacons within range.
The measures obtained, although unofficial (because the measurement instruments are not certified for this
purpose) can give additional insights about the pollution of the city, both at real time and in historic terms, and
can in turn guide the installation of certified pollution counters where they are really necessary.
4. APPLICATIONS
The deployment of RYDER to monitorize vehicular activity in a city has a myriad of applications, including:
•To know the real use of all loading/unloading areas of the city.
•To detect at real time vehicles that have exceeded their authorized usage time for a particular load-
ing/unloading area, indicating this fact to Police agents surrounding the area.
•To offer to local authorities all kind of information about the delivery activities in the city, both at real
time and aggregated.
•To offer the owners of delivery vehicles an estimation, both at real time and historically, of the usage of
the loading/unloading areas, to better schedule their activity.
•With additional processing, the system can answer many interesting questions, including:
–The approximate path followed by each delivery vehicle, and by all the vehicles that belongs to the
same delivery company.
–The type of vehicles that use each loading/unloading area.
–The time intervals with bigger amount of delivery traffic for each city zone.
–Which loading and unloading areas requires more space, and which ones can be safely reduced or
removed.
–How much time spends the vehicles in each area on average.
–How much weight due to delivery vehicles suffers each area.
–With the help of other sources of information and Big Data techniques, a complete model of the traffic
of the city can be built, including the contribution of all types of vehicles.
•Optionally, the deployment of RYDER can also give the local authorities an estimation of the pollution in
different parts of the city, by measuring ozone, carbon dioxide, sulfur dioxide, nitrous oxide, and suspended
particles, together with temperature and humidity. This information, although unofficial, can guide the
local authorities to examine the need of setting up certified pollution measurement equipments at certain
places.
The use of beacons in each vehicle can also allow the authorities to detect improper use of the authorization
cards issued by the city. We have developed a smartphone/tablet app, called Fischer, that helps the local Police
to quickly check whether the beacon shown by a particular vehicle indeed belong to it. The Fischer app shows
the Police agents the characteristics of the vehicles associated to all the beacons within range (see Fig. 4).
5. CONCLUSIONS AND FUTURE WORK
The RYDER project is a complete sensoring solution for smart cities, that can be used for multiple purposes,
including the use of loading/unloading areas, the estimation of pollution, not only due to exhaust gases, but
also due to light or noise, and the tracking of all kind of vehicles and people that carries a Bluetooth emitter.
Additional use cases are easy to imagine, from tracking valuable goods to mascots, among a myriad of other
uses.
ACKNOWLEDGMENTS
This research has been partially supported by MICINN (Spain) and ERDF program of the European Union:
HomProg-HetSys project (TIN2014-58876-P), CAPAP-H6 Network (TIN2016-81840-RED), and COST Program
Action IC1305: Network for Sustainable Ultrascale Computing (NESUS). The authors would also like to thank
Roberto Riol for several inspiring conversations on this topic.
REFERENCES
[1] “Tookan Fleet Management.” https://tookanapp.com/complete-fleet-management- software-system/.
(Accessed: 10 August 2017).
[2] “Orbcomm, Fleet Management and Telematics Solution.” http://www2.orbcomm.com/
fleet-management/. (Accessed: 10 August 2017).
[3] “Amac Fleet Manager.” http://www.amacfm.com/. (Accessed: 10 August 2017).
[4] Heydon, R., [Bluetooth low energy: the developer’s handbook ], Prentice Hall (2013).
[5] Upton, E. and Halfacree, G., [Raspberry Pi user guide], John Wiley & Sons (2014).
[6] Kolodziej, K. W. and Hjelm, J., [Local positioning systems: LBS applications and services], CRC press
(2006).
[7] Mainetti, L., Patrono, L., and Sergi, I., “A survey on indoor positioning systems,” in [Software, Telecom-
munications and Computer Networks (SoftCOM), 2014 22nd International Conference on], 111–120, IEEE
(2014).
[8] Song, Z., Jiang, G., and Huang, C., “A survey on indoor positioning technologies,” in [Theoretical and
Mathematical Foundations of Computer Science], 198–206, Springer (2011).
[9] Bekkali, A., Sanson, H., and Matsumoto, M., “Rfid indoor positioning based on probabilistic rfid map and
kalman filtering,” in [Wireless and Mobile Computing, Networking and Communications, 2007. WiMOB
2007. Third IEEE International Conference on], 21–21, IEEE (2007).
[10] Saab, S. S. and Nakad, Z. S., “A standalone rfid indoor positioning system using passive tags,” IEEE
Transactions on Industrial Electronics 58(5), 1961–1970 (2011).
[11] Alhadhrami, S., Al-Salman, A., Al-Khalifa, H., Alarifi, A., Alnafessah, A., Alsaleh, M., and Al-Ammar, M.,
“Ultra wideband positioning: An analytical study of emerging technologies,” in [Proceedings of the Eighth
International Conference on Sensor Technologies and Applications, SENSORCOMM], 1–9 (2014).
[12] Gunia, M., Protze, F., Joram, N., and Ellinger, F., “Setting up an ultra-wideband positioning system using
off-the-shelf components,” in [Positioning, Navigation and Communications (WPNC), 2016 13th Workshop
on], 1–6, IEEE (2016).
[13] Hazas, M. and Hopper, A., “Broadband ultrasonic location systems for improved indoor positioning,” IEEE
Transactions on mobile Computing 5(5), 536–547 (2006).
[14] Fukuju, Y., Minami, M., Morikawa, H., and Aoyama, T., “Dolphin: An autonomous indoor positioning
system in ubiquitous computing environment.,” WSTFES 3, 53 (2003).
[15] Aitenbichler, E. and Muhlhauser, M., “An ir local positioning system for smart items and devices,” in
[Distributed Computing Systems Workshops, 2003. Proceedings. 23rd International Conference on], 334–
339, IEEE (2003).
[16] Lee, C., Chang, Y., Park, G., Ryu, J., Jeong, S.-G., Park, S., Park, J. W., Lee, H. C., Hong, K.-s., and Lee,
M. H., “Indoor positioning system based on incident angles of infrared emitters,” in [Industrial Electronics
Society, 2004. IECON 2004. 30th Annual Conference of IEEE], 3, 2218–2222, IEEE (2004).
[17] Kim, J. and Jun, H., “Vision-based location positioning using augmented reality for indoor navigation,”
IEEE Transactions on Consumer Electronics 54(3) (2008).
[18] Subhan, F., Hasbullah, H., Rozyyev, A., and Bakhsh, S. T., “Indoor positioning in bluetooth networks
using fingerprinting and lateration approach,” in [Information Science and Applications (ICISA), 2011
International Conference on ], 1–9, IEEE (2011).
[19] Yoon, P. K., Zihajehzadeh, S., Kang, B.-S., and Park, E. J., “Adaptive kalman filter for indoor localization
using bluetooth low energy and inertial measurement unit,” in [Engineering in Medicine and Biology Society
(EMBC), 2015 37th Annual International Conference of the IEEE ], 825–828, IEEE (2015).
[20] Ko¨uhne, M. and Sieck, J., “Location-based services with ibeacon technology,” in [Artificial Intelligence,
Modelling and Simulation (AIMS), 2014 2nd International Conference on ], 315–321, IEEE (2014).
[21] Crockford, D., “Json: Javascript object notation.” http://www.json.org (2006).