This section discusses works by various authors using wireless robotic vehicles in different application areas. It concentrates mainly on two major applications: smart parking systems and surveillance systems.

Smart Parking Systems

In today’s world, vehicle parking is one of the key challenges faced when people visit shopping malls, movie theaters, restaurants, etc. As a result, researchers have come out with smart car parking prototypes. These still face difficulties, however, in manipulating the data received from sensors, including the fusion and filtering of data. "Data fusion” refers to the accumulation and consolidation of data from many sensors. "Data filtering” refers to a process that removes unused data transmitted bv the sensors, to reduce the delay time in data transmission. Both data fusion and data filtering remain open challenges for today’s researchers.

Another major challenge lies in choosing appropriate algorithms, to process the information gathered. In addition, all of the node sensors require exceptional functionality, to prevent errors as much as possible.

In 2018, Wael Alsafery et al. [1] proposed a system wherein initially, the sensed data is obtained from various sensors distributed inside the parking lot and also outside, for on-street parking. The data obtained from the sensors is then manipulated and handled locally, with the assistance of IoT edge gadgets, thus making the system

General IoT parking system

FIGURE 3.1 General IoT parking system.

work in real time, because it continuously receives data from the sensors. This cumulative sensor data is analyzed using artificial intelligence (AI) algorithms, as shown in Figure 3.1, through pre-characterized conditions defined in the algorithms. This framework is easy to understand and consists of a versatile application (app), which encourages clients to effectively discover the closest empty vehicle leaving zone by declining conceivable traffic check through Google API, which gives a constant perusal of traffic status.

Muftah Fraifer and Mikael Fernstrom [2] have proposed a cloud-based intelligent vehicle parking system for developed cities. The entire arrangement consists of three main layers: the sensor, communication, and application layers. The server searches for the finest vacant zone, based on the user’s preference. After searching, the server sends the user the route that he or she has to drive, to arrive at that vacant parking zone. The prime idea of the paper is to use the CCTV camera systems already installed. Initially, the authors aimed to make the existing CCTV cameras smarter, but later, they proposed a low-cost and reliable model of an intelligent parking control system, using an embedded micro-web server, with IP connectivity for remote evaluation and monitoring. This system has improved reliability in finding the preferred parking zone without wasting time.

Vaibhav Hans et al. [3] have proposed a smart parking system that is very convenient for the users. Their main objective is to find nearby vacant parking zones and allocate them at the entry point itself, identifying the vehicle through a mobile app. They have designed their system in such a way that payment for the parking can be processed through an online transaction. Also, they give first preference to aged/ physically challenged people, so that a parking zone can be reserved at the earliest point in time and near the elevator. They use image processing techniques to recognize the license plate, to identify the vehicle. IBM Blue Mix applications have used for cloud storage [4].

IoT-based sensor-empowered intelligent parking system

FIGURE 3.2 IoT-based sensor-empowered intelligent parking system.

Abhirup Khanna and Rishi Anand [5] have proposed an IoT-based intelligent vehicle parking system, which uses a mobile app to extract information through sensors about available parking spaces and send the information to a cloud platform in real time, as shown in Figure 3.2. Ultrasonic sensors were used to determine the availability of parking spaces. This information was sent to the cloud through a Wi-Fi chip attached to the Raspberry Pi. w'hich acted as a gateway to the cloud, using the Message Queuing Telemetry Transport Protocol (MQTT) protocol. This is a lightweight data-centric protocol. It is 93 percent faster than the Hypertext Transfer (HTTP) protocol. Thus, by using the mobile app. a user can see vacant parking locations. But the execution charge of this framework is very high, because each sensor is connected to the Wi-Fi chip.

В. M. Mahendra et al. [6] have proposed an IoT-based sensor-empowered smart car, as showm in Figure 3.3. The proposed system is customer authenticated. It also excludes the false charges that happen when a customer parks a car in a zone other than the one he or she reserved. The overall implementation cost is less than for some other systems, since the application uses low-cost IR sensors, which proves to be advantageous.

Yanxu Zheng et al. [7] have proposed a parking vacancy forecast system, wherein each parking slot is sensor enabled, to forecast the standby time based on such parameters as temperature, humidity, weather, day and time, etc. Using these parameters, the standby time is forecast by algorithms such as a regression tree, support vector regression, and neural network [8,9,10].

Juan Rico et al. [ 11] have proposed a simple parking system, using the framework information of the city. The proposed structure allocates four parking conditions: vacant parking zone, booked parking zone, in-use parking zone, and load/unload parking zone. Here, the amount due can be paid wirelessly through near-field communication technology. The four parking conditions can be done by using geomagnetic sensors, w'hich help to detect the presence of a vehicle. The main drawback of geomagnetic sensor-based vehicle occupancy detection is that the sensors’ reactions are prone to magnetic intervention.

Overview of a smart parking system

FIGURE 3.3 Overview of a smart parking system.

F. Zhou al. [12] have proposed an intelligent parking assortment. In this method, magnetic and ultrasonic sensors are used wirelessly, to detect the presence or absence of vehicles in the parking zones accurately. They have also described a customized version of the Min-Max algorithm, to discover vehicles using magnetometers.

R. Shyam et al. [13] have proposed a technique that utilizes the current structure of a cloud server, such that a boundless amount of space might be incorporated, with no change in the code. The developed mobile application (app) can run on Windows, Android, and iOS. In addition, the code can be utilized for multiple boards, making the proposed framework cost efficient, reliable, and resourceful.

R S. Saarika et al. [14] have proposed a system wherein the sensor’s data is collected at the fog controller (decentralized computing/edge computing), as shown in Figure 3.4. The collective data is sent to the edge devices that are close to the clients’ area, to begin examining and taking care of information. This refined/processed information is then sent to the corresponding clients, to call attention to the adjacent vacant space least obstructed by traffic. The client also receives a response from the cloud showing the route with the least traffic, heading toward the parking space in the client’s locality.

Burak Kizilkaya et al. [15] have proposed a system using the progressive placement algorithm with a binary search tree (BST) to formulate and identify the nearest possible locations. Initially, they locate parking zones nearby. Once the nearest parking space is discovered, it can be used in the hunt for empty parking spaces. This progressive technique with BST extensively improves the inquiry system, in terms of the required hunt time and energy effectiveness. The main aim of their proposed system is to reduce time in finding vacant parking zones. By utilizing a progression in placement algorithm, less time is expended in finding the nearby vacant zone in a

Overview of a smart transportation system

FIGURE 3.4 Overview of a smart transportation system.

car park. A decrease in time while parking a vehicle involves less power and fuel usage and even creates less CO,.

Felix Jesus Villanueva et al. [16] have proposed a framework using a magnetometer sensor, which is implanted in a large portion of the present cell phones. They proposed an idea where they designed an app that empowers individuals to share their sensor data, in exchange for receiving data on empty stopping zones, whenever needed.

Surveillance Systems

Surveillance cameras are widely accepted and acknowledged today and comprehensively utilized to detect unlawful activity and conduct inspections, such as in shopping malls, airports, bus and railway stations, individual assets, etc. The surveillance camera is a proficient development whose purpose can be properly utilized in all locations.

Nowadays, government agencies recommend and emphasize the importance of using these cameras to make the transportation system easy, as traffic details are reported on radio stations all the time [17]. Traffic’s random conditions are well known to regular drivers, who would abstain from routes that frequently have more traffic obstructions. The major advantage in this regard is to convey safety measures to the public. Surveillance cameras also should to be correctly placed. This alleviates the problems of misbehavior and the burglary of vehicles in houses, strip malls, and cafes. [17]. The next sections present some of the methodologies previously adopted for smart vehicles used in surveillance.

Poonsak Sirichai et al. [17] have proposed a system wherein, during the early development phase of the car, the camera and radio receiver are deployed to ensure safety. Along with the main purpose of driving, this also serves the purpose of monitoring vehicle disasters, investigating offenders, and direction-finding. Figure 3.5 presents a block diagram of the proposed system. Using radio receivers, drivers can

Block diagram of a smart car with a surveillance camera for native land protection

FIGURE 3.5 Block diagram of a smart car with a surveillance camera for native land protection.

also evade commuter traffic jams. Besides these, the information collected can be kept undisclosed until the examination is introduced by the administration authority required for giving reasonable judgment between two gatherings. In addition, alarms indicating suspected wrongdoing can be added to the framework, which is helpful for safety networks, homeland security, and the insurance business.

Chein-Hung Chen et al. [18] have proposed a smart camera that is a split mobile cloud model for deep learning, capable of shooting and processing videos. It then refines the vital information from the video clips and sends it to the cloud platform. The authors also showcased the usage of the Git protocol in portable cloud design.

In their work, Sirichai et al. [19] propose a technique known as a vehicular ad hoc network (VANET), as shown in Figure 3.6. VANET is an intelligent network technology, where each car acts as a mobile node to share information, such as road status and traffic congestion, without any central access point as a safety measure. Thus, mobile vehicles can assist transportation safety by avoiding congestion in traffic systems through VANET technology. In the future, VANET technology can help capture pictures that can be recorded, analyzed, and utilized for transportation security and broadcasting purposes. Furthermore, for security purposes, the transmitted pictures can be encrypted through algorithms, which can be accessible only by the authorities.

As the literature shows, the data or images communicated over a wireless environment can be made available remotely, over a cloud platform. Then, these datas or images are processed using various algorithms, either locally or in the cloud environment. However, the problem or gap in these treatments is how far the data are secure. That is the billion-dollar question to be answered.

This research gap applies to how securely our data/images are available in cloud platforms. Our focus is to have secure data communication in a wireless environment. Data security is the need of the hour. Protecting information from malicious

Structure of VANET

FIGURE 3.6 Structure of VANET.

attacks in networks or by data hackers, both on the server side and the client side, is the prime concern.

Our main focus is to develop a surveillance robot (spy-bot), whose movement can be controlled remotely and which captures images that are stored and wirelessly communicated over a secured Windows 10 Azure cloud platform. This platform provides platform-as-a-service (PaaS) cloud computing services, which allows customers to develop, run, and manage applications without the complexity of developing and maintaining the infrastructure. Azure cloud has built-in security services that include unparalleled security intelligence to help in identifying rapidly evolving threats at an early stage [29].

< Prev   CONTENTS   Source   Next >