The basis of the Autonomous Economy is the promise of Autonomous services such as fleets of Autonomous vehicles, Autonomous drone swarms and Autonomous robot workers. This vision is driving the future of IoT using 5G and is a tantalising prospect for both end–users, enterprise customers, MNOs and IoT platform providers as the suite of killer applications that will enable both step-change in value-creation (productivity gains, new revenue streams, new business models) for the users and rapid monetization of the heavy 5G investments by suppliers and providers. However, significant challenges exist today that is preventing potential users for embracing Autonomous services built over IoT platforms and delivered over networks, both 4G and 5G.
This paper released today at the 5G North America in Austin, Texas where Kumardev Chatterjee, CEO of Unmanned Life is speaking alongside industry leaders Verizon, AT&T and others in the session titled “Future of IoT”, sets out how Unmanned Life in conjunction with major MNOs on both sides of the Atlantic and leading global equipment providers is addressing these challenges using the combination of Network slicing and Unmanned Life’s ground-breaking Autonomy-as-a-Service architecture and AI driven IoT SAAS platform that implements this architecture.
Fulfilling the promise of 5G for the Autonomous Economy with Network Slicing and Autonomy-as-a-Service
To be able to be deliver the exciting IoT use cases that are driving the User interest in 5G and to deliver the monetization potential of 5G for the MNO, simply building the core and network is not nearly enough. To achieve the Future of IoT over 5G, it is crucial to deliver both Network Slicing on the infrastructure side and Autonomy-as-a-Service on the edge. The combination of these two is able to provide that simultaneously high-throughput, low-latency, resilient, reliable and scalable platform that can orchestrate multiple complex IoT use cases such as a swarm of autonomous cars providing an autonomous ride-sharing service or a swarm of drones supporting real-time deployment of life-saving emergency rescue services.
Network slicing is expected to play a critical role in 5G networks because of the multitude of use cases and new services 5G will support. These new use cases and services will place different requirements on the network in terms of functionality, and their performance requirements will vary enormously. For example, an autonomous drone communication which requires low latency but not necessarily a high throughput. A streaming service watched while the drone is in motion will require a high throughput and is susceptible to latency. Emerging IoT / IoFT use cases of these types can both be delivered over the same common physical network on virtual network slices.
Autonomy-as-a-Service is the crucial glue that binds the network capability to the IoT layer and manages the capabilities to make it possible for the User to easily deploy and manage autonomous use cases integrated with their enterprise information systems and applications. Autonomy-as-a-Service enables companies with little experience in automation to utilise fully autonomous multi-robot systems like autonomous car fleets or drone swarms they would be otherwise unable to manage. For example, autonomous drones could be used for delivery from web-stores, retail phone orders and restaurant deliveries, speeding up the delivery, while reducing traffic and fuel consumption. Autonomous mail and shipping systems would drastically cut transport and transit delays, at the same time reducing dangerous work where injuries are common.
5G Network Slicing as the enabling Network Layer for Distributed IoT Services Orchestration
5G Network Slicing enables distributed IoT services orchestration:
Network slicing is a form of virtualisation much like software defined networking (SDN) and network functions virtualisation (NFV) in fixed networks. It aims to create multiple networks that share the same physical infrastructure which would allow operators to dedicate a portion of their network to a certain functionality and should make it easier for MNOs to deploy 5G-enabled applications, partitioned into virtual elements that can be linked through software. Network slicing allows multiple virtual networks to be created on top of a common shared physical infrastructure. The virtual networks are then customised to meet the specific needs of applications, services, devices, customers or operators.
For 5G an MNO will have 3 pieces as part of E2E network slicing – 1. RAN Network Slicing 2. Transport Network Slicing and 3. Core Network Slicing.
When the traffic from the RAN slices travel over the transport slices to reach their targeted virtualised core network a complete E2E slice is achieved. Each virtual network (network slice) comprises an independent set of logical network functions such as speed, capacity, connectivity and coverage that support the requirements of the particular use case and can be optimised to provide the resources and network topology for the specific service and traffic that will use the slice and allow it to be independently managed and orchestrated by the operator or customer.
Autonomy-as-a-Service as the Integration and Delivery Layer for 5G enabled IoT Services
Unmanned Life’s Autonomy-as-a-Service Architecture allows for the integration of drones of different types (aerial, ground, hybrid) and with a variety of capabilities to work together as robust autonomous fleets, in conjunction with other IoT devices and hardware, powering their deployment over the cloud via a single, multi-device compatible user interface. The architectural approach ensures the de-coupling between business information systems, physical infrastructure, networks and the autonomous swarms allowing scalability on-demand.
The Unmanned Life platform is an AI driven IoT Software-as-Service platform that implements this architecture to enable users to deploy and manage autonomous robot fleets (i.e. drones and rovers) and complex autonomous missions from a single management interface, on-demand and scale up cost-effectively. The platform connects these autonomous industry 4.0 solutions with business information systems and enterprise AI, as well as enables them to work both indoors and outdoors, with or without GPS, over Wi-Fi, 4G LTE or 5G. The AI driven Platform manages the entire autonomous deployed solution as an integrated system of systems with intelligence based near real-time decision making to drive coordinated outcomes. The advanced AI takes care of the fundamentals required for autonomous workforces from three-dimensional mapping, navigation and location accuracy using advanced optical sensor capabilities for collision avoidance and human-detection. By definition, the platform is Hardware-agnostic and controls any standard UGV (rover) and UAV (drone) as well as robotic hardware components like grippers and a range of sensors allowing it to orchestrate and deploy various use case specific solutions without redevelopment of the core.
Industry Study: Autonomous control and management of drones using 5G network slicing to drive new mission-critical services globally
On February 21, 2018, BT, Verizon, Ericsson and Unmanned Life demonstrated autonomous control and management of a fleet of drones in central London, which was launched from the US by Verizon, on a dedicated 5G network slice within BT’s network. The use cases were centred around disaster response scenarios including delivery of emergency kit or rescue equipment to a disaster area using drones, search missions with HD imaging in disaster recovery zones, specifically areas of difficult access and coordinated missions where multi drone fleets from multiple countries could be used to inspect an emergency area or perform specific tasks (e.g. load handling).
These use cases were enabled by Unmanned Life’s AI platform. These techniques can deliver a vast array of applications, including mission critical services that require ultra-low latency and high availability to achieve real-time feedback loops that enable applications like remote operations regardless of geographical location.
See the full video here: https://www.ericsson.com/en/videos/2018/2/autonomous-control-and-management-of-drones
Use Case: Fire Emergency Rescue using Autonomy-as-a-Service Architecture
(featured by the BBC: http://www.bbc.com/news/business-43906846)
Multiple emergency services respond to a fire emergency and need live data in to make decisions and coordinate the response. However, due to the nature of the emergency, different teams with different command structures, need different sets of data about the unfolding emergency which they simply don’t have the tools to obtain, making it is difficult to get the data rapidly and safely enough to plan, execute and coordinate the response. This leads to miscommunication, uncoordinated and conflicting instructions, misdirection to the trapped people and delayed response time. All this leads to larger number of deaths, damage and longer-drawn out resolution. The environment of the emergency area also deters and slows down the ability of teams to respond. Equally, such emergencies lead to localised telecommunication network failures due to over-burdened or damaged equipment.
Unmanned Life’s Autonomy-as-a-Service architecture and platform allows multiple different types of drones with different data gathering instruments such as telecommunication small cells, high-resolution HD cameras, thermal imaging, sonar based structural damage detectors and others to be deployed simultaneously, autonomously and most importantly in a coordinated and reliable manner, with no human pilots or intervention required. The drones can go near and around the building on fire and capture the data, without putting at risk the responders, and the AI software can control individual drones as well as coordinate the swarm of drones, such that there is coherence in the data received to allow the emergency teams to align their actions and respond better based on joint analysis of the current data received. This is possible because the software platform is integrated with the telecommunications networks using the Autonomy-as-a-Service Architecture.
Unmanned Life’s Autonomy-as-a-Service architecture implemented through its AI driven IoT SAAS platform and Network Slicing is the combination of the infrastructure and edge paradigms required to enable that simultaneously high-throughput, low-latency, resilient, reliable and scalable platform that can deliver Autonomous services over IoT platforms such as a swarm of autonomous cars providing an autonomous ride-sharing service or a swarm of drones supporting real-time deployment of life-saving emergency rescue services and many others.
Both the future of IoT services as well as the monetisation of 5G depend on such innovative mechanisms to bring the benefits of 5G to all. Unmanned Life is actively working with leading MNOs on both sides of the Atlantic as well as leading IoT platform providers and network equipment makers to bring to life the promise of the Autonomous economy.