Edge computing: SWIM launches AI digital twins that can learn

Edge computing: SWIM launches AI digital twins that can learn

AI startup SWIM aims to democratise both AI and digital twin technologies by placing them at the edge and within reach of everyone’s budget – without the need for large-scale number-crunching. Chris Middleton reports.

With Pure Storage and NVIDIA recently launching an ‘AI supercomputer in a box’, it is easy to believe that enterprise-grade AI is solely about throwing massive number-crunching ability at big data sets and seeing what patterns emerge. But while these technologies are notionally aimed at all types of business, the cost of optimised AI hardware that can be slotted into a data centre may be too high for many organisations, as our recent report suggested.

At the other end of the scale are technologies such as IBM’s Watson and Watson Assistant, which can be deployed as cloud services – and, of course, the numerous suite-based AI tools that are currently offered by Oracle, SAP, Salesforce, Google, Microsoft, and others. 

But for many IoT and connected-device deployments, neither data centre not cloud options are realistic, which is why many AI systems are moving elsewhere, fast.

For time-critical processing – such as when an autonomous vehicle needs to avoid a collision – the edge environment and the distributed core are where the real number-crunching needs to take place. This is why companies such as Microsoft and Dell have announced new IoT strategies that focus principally on the edge and/or the distributed core.

The ability to add AI at the edge is an increasingly important element in the IoT, avoiding the need to transfer large amounts of data to supercomputers or the cloud and back again to IoT networks.

Startup SWIM.AI aims to “turn any edge device into a data scientist”, without the need for big data sets and the enterprise-grade number crunching that goes with them. 

The company recently emerged from stealth mode and announced the release of a new AI edge product, EDX. It aims to supply business and operational insights in real time from what the company calls “grey edge data”.  

Twin solutions

The new product autonomously builds digital twins directly from streaming data in the edge environment.

The system is built for the emerging IoT world in which real-world devices are not just interconnected, but also offer offer digital representations of themselves, which can be automatically created from, and continually updated by, data from their real-world siblings.  

Digital twins are digital representations of a real-world object, entity, or system, and are created either purely in data or as 3D representations of their physical counterparts. The concept is core to platforms such as Microsoft Azure IoT and AWS.

For example, every component of the largest machine in history, the Large Hadron Collider, is stored as a digital twin in CERN’s enterprise asset management (EAM) system. This allows scientists to not only know where everything is and what it looks like, but also how well components are performing and when they need upgrade, repair, or replacement.

But for most organisations, that kind of massive, bespoke programme isn’t an option, and they need something simpler, easier to deploy, and cheaper.

Predictive twins

SWIM’s EDX system enables digital twins to analyse, learn, and predict their future states from their own real-world data, according to the company. In this way, systems can use their own behaviour to train accurate behavioural models via deep neural networks.

The important difference to other AI solutions is that this ability is offered as a service in real time, without centralised, batch-oriented big-data analysis.  

“Our digital twins encapsulate what you really need to know about current and predicted performance of instrumented assets, and the insights can be easily integrated into your ERP logic, ops processes, or workflows to help inform decision- makers, in real-time,” said SWIM in an announcement.

SWIM EDX applications include smart cities, industrial automation, utilities, and IT infrastructure optimisation. “We’ve even used SWIM EDX to power an autonomous swarm of drones, flying a complex mission”, claimed the company.

Read more: Research: NASA to explore Mars with swarm of robot bees

Twin management

Gartner views digital twins as one of the top strategic enterprise trends in 2018. However, a key challenge is how easily enterprises can implement the technology, given their investments in legacy assets.

Read more: Gartner: Four best practices for managing digital twins

SWIM believes that limited skill sets in streaming analytics, coupled with an often poor understanding of the assets that generate data within complex IoT systems, make deploying digital twins too complex for some. Meanwhile, the prohibitive cost of some digital twin infrastructures puts other organisations off.

“Digital twins need to be created based on detailed understanding of how the assets they represent perform, and they need to be paired with their real-world siblings to be useful to stakeholders on the front line,” said SWIM.

“Who will operate and manage digital twins? Where will the supporting infrastructure run? How can digital twins be married with ERP and other applications, and how can the technology be made useful for agile business decisions?”

The company claims that SWIM EDX addresses these challenges by enabling any organisation with lots of data to create digital twins that learn from the real world continuously, and to do so easily, affordably, and automatically.

Importantly, the service can be delivered without new infrastructure, skill sets or expert management, said the company.

Internet of Business says

The edge environment is critical to the success of IoT programmes, but the challenge has long been that systems generally need to be fast, simple, and low cost, which means that processing power isn’t always available locally. And when it comes to AI, IoT systems need to maintain that speed, simplicity and low cost without the need for periodical retraining.

Put simply, edge systems are optimised for proximity, so latency isn’t an option – in other words, real time needs to be real time, and not an approximation of it.

In this sense, SWIM seems to have hit upon an ideal solution with EDX. But unlike the edge systems offered by AWS (DeepLens), Microsoft, Dell, and Google (Android Things), EDX isn’t part of a vendor ecosystem that locks clients into a platform or way of working. That may be an advantage to some organisations, but not to the others that can simply extend their platform to the edge environment.

Read more on the the shift from cloud to edge computing.