(untagged)

Optimizing Traffic for Emergency Vehicles using IOT and Mobile Edge Computing

Intel Corporation

13 Feb 2018

In this paper we focus on how Vehicle-to-Infrastructure connectivity can help to optimize city traffic for emergency vehicles, and provide personalized treatment to patients in need.

This article is in the Product Showcase section for our sponsors at CodeProject. These articles are intended to provide you with information on products and services that we consider useful and of value to developers.

Abstract

The ‘Internet of Things’ as a technology has opened up automation opportunities in almost every aspect of our daily lives, be it at home, in healthcare, in industry, with transport, wearable tech, and more. These opportunities have not only helped ease our day to day life, but also generated additional revenue models for businesses.

Any modern vehicle contains many sensors which are used for creating a smooth riding experience. A few years back the automotive industry started centralizing and analyzing this sensor data to improve on their designs and safety features, thus creating better user experience for the end user. Various connectivity protocols are available for vehicles, which are used to collect data (such as real-time analytics) from the internal and external environment. This helps create the next generation of intelligent vehicle systems. Standards developing organizations like ETSI* and 3GPP* are working together to standardize the communication protocol for vehicle-to-vehicle, and vehicle-to-infrastructure technology. This allows technology from different vendors talk to each other, creating opportunities for intelligent system integration into Smart City networks, and Intelligent Transport Systems.

In this paper we focus on how Vehicle-to-Infrastructure connectivity can help to optimize city traffic for emergency vehicles, and provide personalized treatment to patients in need. We will use a oneM2M* compliant ecosystem, and ETSI Mobile-edge Computing technology, thus creating a true smart city.

Introduction

By 2020, there will be billions of connected devices communicating over interconnected networks. To support this market of exponentially growing M2M devices, a sustainable and scalable infrastructure is required. Most IoT solutions have their own proprietary framework which little interoperability. For service providers, deploying multiple frameworks/ecosystems for same service is neither a feasible nor a profitable solution.

Across the world, SDOs (Standards Defining Organizations) are working with industries to standardize Internet of Things technology and improve interoperability with a transport layer. OneM2M has adopted ETSI M2M standards and is now working with different industry segments to develop interoperable standards which not only support standardized communication protocols, but also work under a true heterogeneous environment.

With millions of devices already connected, there is ever increasing demand from end users for personalized service, better performance and a better user experience. Businesses want detailed information about their consumers, easier and secured access to devices, and greater flexibility for provisioning new services. This continuous growth is predominantly driven by mobile phones and connected devices.

For a smart city, an intelligent, secure and integrated Traffic Control system is required to ensure smooth operation of day to day traffic as well as manage emergency vehicle movement with minimal delay. Among available traffic control systems, none can currently automatically clear traffic congestion in case of emergency, allowing vehicles like ambulances, fire engines, and police vehicles to pass. With exponential growth in number of vehicles on the road, there is a strong need for Intelligent Traffic and Transport Systems for hospitals, fire stations and police stations.

Mobile-edge computing technology can help solve this problem by introducing IT cloud infrastructure computing resources. Compute nodes, along with mobile radio stations, allow deployment of latency and bandwidth hungry applications. Pre-authorized edge applications can use real time network context information, provided by a MEC node, and artificial intelligence analytics to create a next generation intelligent system which can be used to solve day to day problems that are difficult for humans.

TCS (TATA Consultancy Services Ltd) has designed and developed an ecosystem based on OneM2M and ETSI Mobile-edge computing standards which can help deploy/ implement highly scalable and mission critical use cases. This system takes advantage of RAN (Radio Access Network) proximity and Internet of Things features. The Intelligent Traffic System will be discussed in detail in later sections, in particular how it utilizes MEC and IOT technology to optimize traffic to allow a free passage for emergency vehicles as well as connecting to network hospitals.

Mobile-edge Computing Overview

Mobile-edge computing technology utilizes existing IT infrastructure and opens it up for network application developers. It can also be utilized by service providers to deploy applications at the RAN edge to generate new revenue models.

Figure 1: MEC Overview

There are two main services in the MEC server, which are used by application at runtime,

RNIS (Radio Network information Service) provides authorized applications with radio network related information, such as network condition, user location, user allocated bandwith, measurement and statistics information related to the user plane at the relevant granularity.
TOF (Traffic Offload Function) provides routing capability based on the policies and routes GTP-U traffic towards applications. Application can decide to terminate the connection or forward it back to EPC, or back to eNodeB.

TCS Edge Node (MEC Server) provides a well documented SDK for application developers for easy development and integration with MEC server. This SDK provides APIs for registration, subscription and notifications of attribute changes. It also provides a management layer which helps create and deploy virtual machines on OpenStack*, and deploys applications from a single management solution without a need for multiple orchestration tools.

OneM2M* Architecture Overview

OneM2M uses a resource-based architecture which exposes functionality through set of APIs over all reference points. A specified, established communication channel shall be used to execute operations over the resources. Below is functional overview of OneM2M architecture, illustrating their end to end ecosystem and reference points.

Figure 2: OneM2M Functional Overview

This model is comprised of three layers: Application Layer, Common Services Layer and the underlying Network Services Layer. The network service layer is the central hub and works as router and consolidator for all traffic from sensors to remote application management nodes.

The gateway, which works as aggregation point, is typically deployed at the end point of the use case. It aggregates the data plane traffic from different sensors and runs personalized algorithms for localized analytics, or routes the traffic to the network hub. These gateways can be deployed on specialized hardware or as a software components in the cloud, depending on the deployment model and use case.

The application service layer collects the data from sensors using predefined granularity and forwards it to a preconfigured gateway or network hub. The application entity also implements logic for remote application connectivity and operations.

Solution Overview

For a person in urgent need of medical care, every second counts. TCS CETO (Centralized Emergency Traffic Optimization), an Intelligent Traffic and Transport System, is designed for smooth passage of emergency vehicles in extreme traffic conditions thus reducing the emergency vehicle time as much as possible. This system provides connectivity with Network Hospitals and an individual’s personal devices to provide personalized first aid, creating a completely automated system.

Figure 3: Solution

This system consists of multiple nodes:

Connected Traffic Junctions is collection of traffic lights which are end sensor based devices (as per OneM2M Standards). Each direction of traffic is connected with a camera to collect current traffic images and calculate density using image processing techniques. This information is sent to an IoT gateway, using the 2G radio network deployed at the edge of a Radio cloud.
MEC Server, the edge computing platform, provides cloud infrastructure for the deployment of IoT gateways and the Local Traffic Analytics Engine.
IoT Gateway, deployed as edge application, collects traffic density at predefined granularities and passes this information to the Local Traffic Analytics engine for further calculation. It also provides the operating sequence to each traffic junction with a time interval based on the traffic density feedback from the Local Traffic Analytics Engine.
The Local Traffic Analytics Engine takes regular input from each traffic light and executes machine learning algorithm to generate traffic patterns, which are used as a feedback system allowing an IoT gateway to decide the operating sequence.
The Centralized Traffic Controller is a central hub (NSE as per OneM2M Standard) deployed in the cloud which connects to all traffic junctions, as well as traffic lights, individually. It provides a common interface for deployment of new traffic lights, configuration and management of operating sequences, fault management and device management.
Network hospital servers contain the databases of all the registered patients and their medical history. It also connects with the Centralized Traffic controller using a REST API to provide sources and destinations of ambulances.
Registered Patient’s device application. This application is either a mobile application or wearable application can detect a person’s medical status, using sensors like heart rate monitors, pacemakers, etc.

There are two scenarios where this system helps ambulances to reach their destination in the quickest time possible.

If a patient is registered, the connected device is used to call for an ambulance, then a request message is sent to network hospital with the patient’s unique id and current location. The network hospital system finds an available ambulance nearest to patient’s location and sends the ambulance’s details (Ambulance unique id, source and destination) to the traffic controller to find fastest route, as well as optimizing traffic along the way. It also pulls the patients’ medical history, and sends it to the ambulance for the support staff to administer first aid.

Figure 4: Flow Diagram – Registered patient
If an ambulance is called for non-registered patient, then network hospital system finds an available ambulance nearest to patient’s location and sends the ambulance’s details (Ambulance unique id, source and destination) to the traffic controller to find fastest route as well as optimizing traffic along the way.

Figure 5: Flow Diagram: Non-registered patient

In either case, the traffic controller route algorithm remains the same. A critical component of centralized, traffic management for road system is signal phase timing for signal-controlled intersections. This is an extremely challenging problem. Currently available algorithms either use brute force, or simplistic assumptions to compute the fastest path which is not efficient, and generates sub-optimal optimal routes in real-time.

We propose an algorithm that is efficient, as well as robust, to compute the optimal path in real-time. The goal is to computing a path that will allow the emergency vehicle to reach the hospital as fast as possible, but at the same time cause minimum inconvenience to other vehicles.

Figure 6: Route Algorithm Sample Route

Algorithm:

The signals operate in two modes: default (MODE: DEFAULT), and override mode (MODE: OVERIDE).

Motive:

In-route : Reduce inflow , increase outflow [if rate-inflow > rate-outflow]
Determine the signals to be controlled - Towards the Hospital (H)

Algorithm:

START

Set the Signals to MODE: OVERRIDE { When the signal is received from the EV}
Determine the distance (D) between Emergency Vehicle (EV) and Hospital (H).
Compute the co-ordinates of a circle of Radius D (Area = pi*D*D) with center at H.
Calculate D as below
Set all the all signals (S) to Red in this area.
Set the signals (S1….Sn) which intersect the line (outgoing) drawn from the circle centered at H to Green.
Determine the signals to be controlled - En-route to Hospital
Compute the time Taken by EV to each of the signals en-route to Hospital.
The time taken to reach signals en-route Se1= te1, Se2=te2,… Sen=ten.
Determine the distance (d) between Emergency Vehicle (EV) and Hospital (H).
Compute the co-ordinates a circle of Radius D (Area = pi*D*D) with center at H.
Calculate d as below
Set all the all signals (S) to Red in this area.
Set the signals (S1….Sn) which intersect the line (outgoing) drawn from the circle centered at signals en-route to Green.

END

Set the Signals to MODE: DEFAULT

Repeat: Start to end – Until the Emergency Vehicle reaches the hospital.

Conclusion

The majority of the most populated cities in the world are old enough that there historically has been very little or no planned roads and intersections. The sight of slow moving, or stuck ambulances in traffic is very common in most of these cities. There is a strong need to create an intelligent traffic management system which not only prioritizes the passage of emergency vehicles, but also takes into account the existing infrastructure.

The above mentioned algorithm took only 3 way and 4 way junctions into account as well as only considering traffic for junctions within a small area, but there is room to improve and enhance it for larger areas and multiple emergency vehicles at the same time.

Mobile-edge computing technology can help to provide an ecosystem at the RAN edge which ensures that there is no, or minimum communication delays between traffic lights, controllers and ambulances. It also opens up opportunities for intelligent applications for smart cities and other IoT use cases.

Author

Anurag Agarwal, Solution Architect (Internet of Things / 5G Networking)

TATA Consultancy Services Ltd | Hyderabad | INDIA | anurag.a@tcs.com

References

About TCS Telecom Industry Solution Unit

TCS’ Telecom Business Unit is the second largest vertical contributing higher percentage to the overall TCS revenues. With a dedicated pool of professionals and an accumulated experience and ongoing associations with world-class Telecom service providers and equipment manufacturers, TCS has acquired unparalleled understanding of the Telecom domain. TCS helps wireline, wireless, broadband, and cable service providers redefine their markets with innovative solutions that help them become more agile, reduce fixed operations costs, and introduce next generation services. TCS sets customers apart from their competitors with instant access to industry solutions, best-in-breed technology, assets, and frameworks.

Contact For more information, contact global.telecom@tcs.com

TCS.com RSS: http://www.tcs.com/rss_feeds/Pages/feed.aspx?f=wFeedburner:http://feeds2.feedburner.com/tcswhitepapers

About Tata Consultancy Services (TCS)

Tata Consultancy Services is an IT services, consulting and business solutions organization that delivers real results to global business, ensuring a level of certainty no other firm can match. TCS offers a consulting-led, integrated portfolio of IT and IT-enabled infrastructure, engineering and assurance services. This is delivered through its unique Global Network Delivery Model^TM, recognized as the benchmark of excellence in software development. A part of the Tata Group, India’s largest industrial conglomerate, TCS has a global footprint and is listed on the National Stock Exchange and Bombay Stock Exchange in India.

For more information, visit us at www.tcs.com

License

This article has no explicit license attached to it but may contain usage terms in the article text or the download files themselves. If in doubt please contact the author via the discussion board below.

A list of licenses authors might use can be found here