This month we announced our new Entrepreneur-in-Residence, David Ripert.

David has a diverse background in technology, including roles at Google/YouTube, where he led EMEA production, programming, and creator initiatives, including with XR experimentation. He also co-founded Poplar Studio, a renowned, VC backed AR start-up. And is the President of the UK chapter of the VR/AR Association, among other industry roles.

This is David’s first thought piece for us on the Industrial Metaverse which is one of our key investment themes for 2024. If you are a founder building in this space we’d love to speak to you! Get in touch here.

While the Metaverse has captured the media’s attention as it relates to the gaming and entertainment sphere, the reality is that its potential extends far beyond this. And CEOs ranging from Satya Nadella (Microsoft) to Jensen Huang (Nvidia) agree with us. The Metaverse has the power to revolutionise how we work, not just how we play. And we’re already seeing a huge number of use cases in action, today.

This article covers some of the most promising areas of the Industrial Metaverse, and how things have moved on since I first wrote about this topic in 2022.

One area that is rapidly developing is the engines that will power the industrial metaverse: Digital Twins.

Digital Twins

Digital twins are specific digital replicas of physical objects, processes, or systems - made up of two key things:

• 3D Technologies

• Data.

These models are a convergence of lots of different metaverse-related technologies 

• Internet of Things (connectivity, sensors etc)

• AI & ML 

• Network Computing

• 3D/VR/AR

All underpinned by data.

Data as a component enables real-time and historical information to create accurate and dynamic virtual representations of physical entities.

For example, a digital twin of an airport creates a virtual model that mirrors the physical airport, including terminals, runways, and baggage systems. This model is updated with live data such as flight schedules, passenger flow, and equipment status.

What Digital Twins Unlock

Everything from construction and automotive, to business decision-making contains real-world data.

As such, there is an opportunity for ‘compute in the real world’ - this is the industrial metaverse.

A digital ‘world’ that through data mirrors real-life, rendered in 3D; machines, factories, buildings, systems, cities, people, and processes.

Every aspect of business operations can happen either entirely virtually or in a hybrid fashion, i.e., virtually and physically.

Jensen Huang, CEO of Nvidia - who amid the AI boom blasted past Alphabet to become the world’s fourth most valuable company - views the Metaverse as an expansive, interconnected universe far beyond mere virtual reality:

 “A whole bunch of robots in the metaverse doing work and communicating with robots that are in the physical world. The physical world and the metaverse can be connected and not just human to humans - it can be machine to machine.”

By blending the physical with the digital, the industrial metaverse will:

• Increase collaboration as people come together in virtual or augmented physical spaces. 

• Enable better decision-making. 

• Change how products and services are manufactured and delivered to customers. 

• Transform how customers and businesses interact, i.e., from generic to personalised, immersive experiences.

The value here is enormous. 

Many companies that originated in the ‘consumer metaverse’ space are now moving into the industrial metaverse space.

In hardware, Magic Leap pivoted towards the industrial metaverse, and a host of flagship VR headset manufacturers (HTC, Meta, and Pico) launched premium-positioned enterprise devices.

Even the new Apple Vision Pro headset is more suited to enterprises than consumers, due to its aspirational $3,500 price tag and productivity-skewed marketing.

Apps like JigSpace, which were a big part of Apple’s press materials and promotion cycle, can help visualise industrial equipment in highly realistic 3D through mixed reality in a space.

Equally in software, Unreal Engine and Unity have carved out a significant position within the industrial metaverse, particularly in fields such as automotive design, architecture and construction, manufacturing and film production.

Unity's expansion into non-gaming sectors is highlighted by the revenue growth in its Create Solutions segment. So much so, Capgemini just acquired Unity’s digital twin professional services arm to accelerate enterprises’ digital transformation through real-time 3D technology.

The evolution of game engines, driven by the gaming industry's quest to deliver increasingly realistic and immersive experiences, has laid the foundation for this shift.

As industries adopt 3D digital twins, they are integrating gaming-native practices throughout their business processes, centering around:

• 3D avatars

• 3D scenes

The applications of these 3D technologies are where we see Metaverse provide true value.

So let’s look at the use cases:

Virtual Representations: Avatars

Avatars (graphical representations of humans) are not new, but they are continuously being developed to better imitate humans.

For context, 300 million Zoom meetings happen every day. Digital identities represented by avatars, will offer more varied and meaningful online interactions than static video conferencing, allowing more collaboration and training opportunities.

Despite the memed Mark Zuckerberg's Metaverse Avatar - Avatars, or graphical representations of users, are in the early innings of sophistication.

NVIDIA's 2022 introduction of Violet, and the Omniverse Avatar Cloud Engine (ACE) for Games, showed cloud-based avatars capable of responding to natural speech showing potential for avatars to function across various industries, including customer service and in-game NPCs. 

More recently, Apple's Vision Pro headset introduced the concept of the "Persona," a digital representation created by scanning the user's face. Personas replicate facial and hand movements during FaceTime calls or in other apps, and although experience reviews are mixed the future of avatar technology will likely be more immersive and realistic.

For example, Meta's Codec Avatars project aims to create photorealistic avatars for remote communication, using machine learning and potentially simple smartphone scans to generate lifelike representations.

As technology becomes more accessible and affordable, such advancements will likely become mainstream, offering high-quality, realistic virtual interactions to a wider audience.

While calling an avatar a digital twin is a stretch. The increased use of real-time scanning to create these representations and more excitingly the integration of AI to bring these avatars to autonomous life stretches close to traditional digital twins.

Manufacturing

BMW announced with NVIDIA at GTC 2023 that it’s expanding its use of the NVIDIA Omniverse platform for building and operating industrial metaverse applications across its production network.

The BMW team can aggregate data into massive, high-performance models, connect their domain-specific software tools and enable multi-user live collaboration across locations. All of this is possible from any location, on any device.

Here BMW can use virtual twins of their production sites to trial changes and determine where problems are (for example, supply chain backlogs) and how to solve them, which will become a blueprint for reducing risks and ensuring success before committing to massive construction projects and capital expenditures.

Increasing the use of 3D models and integrated data will pay off. Putting in optimizations on existing facilities is extremely costly and causes production downtime. So having the ability to pre-optimize virtually eliminates such costs.

Agriculture

Nemo’s Garden, a startup that tests the viability of growing fruit, vegetables, and herbs underwater, uses digital twin technology to iterate designs and test concepts without having to do so physically, which would be very difficult and expensive due to its particular environment.

A number of limitations, including short summers and lengthy physical testing, meant that although the prototype for Nemo’s Garden was a success, the team was restricted to one innovation cycle per year. As a result, it would take them years before they could turn the prototype into a global solution.

To speed up the process and scale the operations, Nemo’s Garden partnered with Siemens to build a twin of Nemo’s Garden biosphere.

The twin simulates the biosphere's conditions and the impact of equipment on the aquatic environment and the surrounding area. They also developed a machine learning algorithm for real-time plant monitoring, eliminating the need for divers to collect data.

Nemo’s Garden Co-founder Luca Gamberini emphasised the uniqueness of Nemo's Garden and the advantage of modelling environments virtually to anticipate and tackle challenges efficiently.

Construction

In the construction sector, Digital Twins are impacting everything from architecture and design to the retrofitting of existing buildings, enabling more efficient construction processes, cost savings and reduced carbon footprints.

By integrating Digital Twins with IoT devices, buildings can achieve smarter facility management—optimising temperature control, reducing energy consumption through intelligent lighting systems, and improving space utilisation based on real-time footfall data. This approach is particularly prevalent in regions like the UK and the Nordics, driven by government mandates for sustainability.

Companies like Siemens and startups such as Twinview are at the forefront, offering solutions that enhance building performance and sustainability. These technologies not only streamline the construction phase but also contribute to the long-term operational efficiency and environmental responsibility of buildings.

Healthcare

Medicine, at its core, is human maintenance.

Digital twins act as virtual counterparts of the human machine, enabling healthcare professionals to monitor, simulate, and optimise human health.

They allow for predictive analysis of medical conditions, personalised intervention strategies, and real-time monitoring mirroring the proactive and analytical maintenance strategies employed in manufacturing.

Pharmaceutical giants like AstraZeneca, Pfizer, Roche, Novartis, and Sanofi are already harnessing digital twins to refine drug development processes.

Digital Twins simulate the behaviour of drugs within the human body, offering insights into efficacy, safety, and optimal dosage. The ability to predict how a drug interacts with the body in a virtual environment before clinical trials can significantly expedite the drug development process, enhancing the precision of drug formulation and reducing the reliance on traditional trial-and-error methods.

London startup AIBody creates digital twins from echocardiograms. Called Nova Heart, the system turns heart scan data into a precise 3D digital twin of an individual's heart. This model accurately displays heart chamber activity, pressures, and rhythms in real time. This model offers cardiologists precise insights and provides patients with a clearer understanding, paving the way for advanced treatments.

Smart Cities

A smart city is a technologically modern urban area that uses different types of electronic methods and sensors to collect specific data. Information gained from that data is used to manage assets, resources and services efficiently; that data is used to improve operations across the city.

Helsinki, Finland, showcases one of the world's oldest and most advanced uses of digital twin technology, aiding in urban planning and environmental goals. Starting with CAD mapping, evolving through 3D simulations, and now utilising photorealistic digital twins with comprehensive data layers, Helsinki's approach serves as a model for sustainable city planning.

This digital twin, enriched with semantic data and powered by gaming engines, supports the city's objectives towards carbon neutrality by 2035 and a recycle-only waste system by 2050, demonstrating the potential of digital twins in urban development and environmental management.

Similarly, "Metaverse Seoul" represents an innovative application of digital twins and VR to enhance city services, planning, and virtual tourism, creating a digital replica of Seoul for improved municipal governance and service delivery.

Transport and Infrastructure

Autonomous driving is gradually moving from single-vehicle intelligence to an internet of vehicles, where traffic participants can share the traffic flow information perceived by each other.

Autonomous vehicles, such as Waymo's with Lidar sensors, contribute to intelligent transportation by forming a network that collects traffic data. This data, added to other city-wide sensor data, like live video feeds from cameras, can support a traffic digital twin model, which can predict conditions to optimise autonomous driving and transportation. It aids in strategic traffic planning and boosts traffic intelligence.

According to the UK’s National Highways who are investing £27BN into their strategic road network, digital twins of England’s roads will reduce traffic jams, decrease the amount of time and money it takes to inspect roads and reduce emissions by 50%.

Twins can help simulate the benefits of autonomous driving, including on the environment. E.g., MIT modelled the impact of the introduction of 4000 autonomous taxis in NYC, using current NYC taxi data. That algorithm could reduce car trips by 98% within 3 minutes, saving gas and pollution.

Automotive firms themselves have been earlier adopters of digital twins and 3D technology.

Many modern vehicles show a 3D twin of the car on the dashboard to highlight issues, and live vehicle information such as traffic position, or open doors.

At the other end of the automotive lifecycle, General Motors (GM) has created digital twins to predict maintenance issues in their equipment. By collecting data about the equipment's performance, they can identify potential issues that may arise. Tesla also uses digital twin technology for maintenance and repairs. The automotive company famously creates digital twins of all the vehicles it sells to determine where issues might occur and reduce the likelihood of car owners having to go to a servicing station.

Maintenance

A digital twin that replicates a building’s state in real time can detect building errors, pinpoint trends, and test different scenarios, like how a cooling system would perform if there were more people in the building than usual.

The sports and entertainment arena in Los Angeles, SoFi Stadium, has a digital twin that helps with daily building management. Besides providing management with information on the current conditions of the building and facility equipment, the digital twin also simplifies game day operations by showing new staff, contractors, and vendors where exactly they’re supposed to be on the day.

Venture-backed Israeli startup, Resonai is the creator of Vera – a computer vision platform for commercial real estate that converts any building into an Intelligent digital twin. Now the building can communicate with any device and property managers can create digital twins of their buildings and visualise and interact with them from any location at any time.

That means they can more efficiently control “Internet of Things” devices, for example, leveraging the integrated data to do a better job of managing energy usage, identifying maintenance issues, enhancing the security of the building, and much more.

The Future of The Industrial Metaverse

The future of the industrial metaverse is poised to further leverage the power of digital twins through AI, and robotics. This hinges on the seamless integration of several key processes and technologies where there are opportunities for startups to aid efficiencies and sustainability in industrial operations.

Automation in Digital Twin Creation

The process of creating digital twins—capturing real-world data, cleaning it up, and creating a digital replica—currently requires significant manual effort. 

The future points towards the automation of these tasks. Advanced sensors (including drones and robots) and IoT devices will increasingly automate data capture, while AI algorithms will streamline data cleanup, making the process more efficient and less labour-intensive. Even capture itself is getting easier, Gaussian splatting allows for the capture and rendering of high-fidelity 3D models from simple 2D imagery, providing a more accurate digital representation.

These advances will enable real-time updates to digital twins, ensuring they accurately reflect their physical counterparts at all times.

AI-Powered Data Analysis

The role of AI will expand beyond data cleanup to power the entire data analysis process within the industrial metaverse. AI algorithms will be trained on vast datasets to recognise patterns, predict outcomes, and suggest optimizations.

This knowledge base will be industry-specific, and highly vertical.

For instance, in manufacturing, AI could predict equipment failures before they occur, schedule maintenance, and optimise production schedules for efficiency.

Robotics and Digital Twins Convergence

Robotics technology will converge with digital twins, leading to unprecedented levels of automation and efficiency.

Robots in the physical world will execute tasks based on simulations and decisions made within the digital twin environment. This synergy will enhance precision in manufacturing, reduce downtime, and allow for the exploration of new manufacturing processes and product designs in a virtual environment before physical implementation. This is already particularly good inside built environments, but still difficult in the real world. However, as we get closer to twins of the whole world, robots will be better trained on that complete synthetic data and perform better in the natural environment (from deliveries to other tasks that can be automated).

Configuring and Simulating for Process Improvement

The ability to configure and simulate digital twins in a virtual environment will drive process improvements across industries.

By experimenting with different configurations and operational strategies in a risk-free virtual space, companies can identify the most effective approaches before implementing them in the real world. This will lead to significant cost savings, enhanced performance, and the ability to quickly adapt to market changes or operational challenges.

-

As the industrial metaverse evolves, it will become an integral part of the industry, transforming how businesses design, manufacture, and maintain products and infrastructure.

The convergence of digital twins, AI, and robotics will not only improve existing processes but also enable new business models and opportunities that were previously unimaginable.

If you’re working on the industrial metaverse, let us know!