“Predicting the future isn’t magic, it’s artificial intelligence.”  ~Dave Waters

 

Edge Computing and AI: What it is and how it’s a Game-Changer – A discussion with Kiran Venkatesh, Co-founder and CEO, FrontM Limited

 

What is Edge AI

Very simply, Edge Artificial Intelligence is the application of Edge Computing and AI together.

Machine intelligence when deployed on local computing devices and servers, as against centralized cloud computers, delivers Edge Intelligence.

 

Why Edge AI for Maritime?

The modern digital world is heavily reliant on centralized cloud systems. It is, in fact, the viability of cloud computing and data architecture today that has truly kicked off the AI revolution and automation across industries.

But the number of connected personal devices, nodes, IoT edge devices are rapidly growing in adoption and are soon going to be in their trillions. With the advent of 5G and 6G, the application of smart tech is expected to grow exponentially.

However, in maritime shipping and other ocean-based industries, a key barrier to digitalization is access to unlimited, uninterrupted, guaranteed quality  internet.

Due to the global nature of the industry and remoteness of operations, connecting via space is the only viable way to meet the demands and reap the benefits for modern tech like AI. Indeed a gold rush is seen in upstream space innovation with new entrants like SpaceX and OneWeb challenging the status using LEO mesh satellite constellations.

Yet, that alone is not enough.

The pace of advancement and physical limitations mean, bandwidth demands will always remain 10x higher than supply and the ROI on Satcom will always remain incomparable to other terrestrial connectivity solutions due to at least 100X higher costs at any point in time.

With VSAT connected vessels, although bandwidth is increasing, factors such as the line of sight, weather fade, localised congestion, traffic detection, latency, and dynamic prioritization, and mobility contribute to limitations in securing uninterrupted cloud access.

So, while land-based manufacturing industries and land transportation are accelerating and already hold the highest market share of digitalization and IoT applications at 27% in 2020, Maritime and ocean-based industries are in contrast yet at a very nascent stage. The solutions landscape is yet up and coming.

So how can we drive appropriate solutions to advance the common themes in Maritime such as smart shipping, sustainable operations, and crew welfare?

 

We believe Edge AI holds the key!

Edge AI enables an always-on, low latency, reliable solution architecture removing the risks from lost connectivity and delays in reaching the cloud. The architecture involves leveraging distributed computing paradigm to make necessary computation and data storage closer to the users and devices where it is collected. Today, user devices are quite capable to compute the locally ingested data. Equally, ships have installed compute servers with virtualised run time specifically for hosting 3rd party solutions.

 

Application of FrontM Edge AI

FrontM is a developer platform with frontm.js a native javascript-based low-code developer API. Using this, we build micro-applications that are hosted on our partner branded apps and/or our super-apps. All apps are multi-platform (iOS, Android, Web, Desktop). FrontM uses the same distributed computing architectural paradigm to run the micro-application code in the nearest node to the source (user or data device) first, before deferring to the cloud. The nearest node may well be the user’s own smart device or laptop, and/or the local onboard servers.

Our initial use cases are quite simply enabling offline workflows. This is applied today on use cases such as,

a) Telemedicine, to gather user inputs and securely ingest user data

b) Crew Welfare, to provide essential information to crew seamlessly cached on users devices for offline consumption

b) Field Service Maintenance, to semi-automate and empower field engineers to access ERP data and perform services, using locally held intelligence

d) Bunker and Lube procurement, to orchestrate offline capture of order requests and process them seamlessly when Satcom is available

 

We are in parallel integrating Machine Learning capabilities into frontm.js to enable the next set of use cases, such as,

a) Offline object detection, be it for class surveys, remote assistance, or crew welfare

b) Predictive maintenance, to establish detection of a developing problem

c) Situational awareness, to detect threats from collision or localized developments

d) Enhanced crew welfare, to pre-empt fatigue by augmenting data from wearables

 

Whilst the initial low-hanging use cases are barely scratching the surface of the significantly vast possibilities, the always-on, ready, reliable, and intelligent qualities of the applications with the use of Edge AI are proving the relevance and value of applying this technology to Maritime.

 

Do you have an idea or a concept that you’d like to initiate an experimental development for?

Our developer tools are built to de-risk and accelerate concepts in the market, do get in touch for initiative a conversation, we’re all about moving the maritime industry forward via collaboration and partnerships.

Since we’re all about moving the maritime industry forward via collaboration and partnerships, do get in touch and initiate a conversation.

Kiran Venkatesh

Kiran Venkatesh

CEO at FrontM. Software, Data and AI thought leader. Passionate Learner. Doer. Team Builder.

Leave a Reply