AI-driven Smartglass Facades

Smartglass and AI

Your smartphone can adjust its behaviour based on your personal data and local events, right? 

Examples include :

• dimming down the screen when the battery is low, 
• detecting the risk of a heart attack using a medical app, and
• finding the best sushi restaurant, near you, right now.

Similarly, a smartglass system should be able to adjust its behaviour based on its local environment and what it learns from global events as they unfold in real-time.

For this, smartglass systems will require artificial intelligence (AI).

Use Cases

We have seen some excellent examples recently from smartglass manufacturers such in which AI algorithms direct building facades based on changes in daylight. 

This has clear benefits (pun intended) for sustainability, privacy and visual comfort.

Let us now consider some more advanced use cases that would take this idea one step further.

Security Threats in Large Cities

Just as car doors unlock automatically after an accident, smartglass building facades should also default to a transparent state in emergencies.

For example, suppose that an AI social media monitoring system detects ‘chatter’ indicating an imminent terrorist threat in a city during a sunny day. 

It is possible that any building with a smartglass facade would be in its energy-saving ‘dark mode’ to minimise glare and air conditioning costs.

This dark mode however would also impair visibility into the building for law enforcement trying to evacuate everyone.

In this case, a city-wide alert should trigger smartglass facades in the vicinity of the security threat to transition all buildings to an ‘emergency mode’. 

In this mode the system would activate the smartglass to its transparent mode, aiding visual communication and facilitating evacuation.

The same principle applies to smartglass installed on public transportation. An AI-driven system can enable communication with passengers trapped after an accident, or if a vehicle detects a stranded infant suffering dehydration.

Theft of High-Value Items

Whether in a private residence, a high-end jewellery store or an art gallery, smart glass in showcases coupled with AI algorithms can prove essential. 

CCTV cameras can anticipate an imminent theft by identifying known criminals using facial recognition, triggering the smartglass into ‘dark mode’.

This would hide all items from view and slow down a potential attacker from finding the most valuable pieces.

After hours, security guards need to see the displayed items while on patrol. 

However, if security is at risk the sensible approach is to hide everything from view. 

Not just the most valuable items, but everything.

Reducing Colour Fading for Fine Art & Luxury

Global art and luxury retail inventories worldwide can amount to Billions of Dollars in assets. However, excessive light exposure can cause damage to paintings, fashion textiles and even fine wines, hurting their market value.

If the smartglass AI control system knew which materials were on display (and their sensitivity), it could dim the glass down a few notches.

This change in light level would be almost imperceptible since the eye is a logarithmic sensor (Weber–Fechner law), but would reduce the damage due to light exposure.

Future systems could learn to identify which items are most vulnerable to damage using AI machine vision, adjusting their behaviour to preserve market value.

Artificial Intelligence Act

The AI Act is a European Union (EU) Regulation passed by the European Parliament in March 2024. 

It states that facial recognition by public CCTV cameras is banned, unless in the case of a missing child or a potential terrorist threat.

The AI Act also requires us to clearly identify AI-generated content or AI agents. In the case of an AI-driven smartglass facade, your legal counsel should clarify whether such a system could create any compliance issues.

Let’s now take a step back and break down the basic building blocks to understanding such AI-driven smart glass systems.

What is Smart Glass?

Smartglass is not a product, but rather a family of technologies.

Broadly speaking, we can divide the family into two major groups: ‘passive’ and ‘active’.

In both cases, a stimulus triggers changes in smartglass properties, such as transmittance, reflectance or electrical conductivity.

Passive Smart Glass

Passive smartglass reacts to environmental stimuli such as light, temperature or pressure. 

Examples include thermochromic, photochromic and piezochromic materials which react automatically when struck by heat, ultraviolet or pressure, respectively.

Active Smart Glass

Active smartglass, on the other hand, is driven electrically, often by way of sensors or control commands from a building management system (BMS).

Examples of active smartglass technologies include:

• suspended particle devices (SPD-SmartGlass), patented by Research Frontiers Inc.
• liquid crystal technology, such as Merck Eyrise
• electrochromic glass such as View, Halio and SageGlass
• electrophoretic smartglass, e.g. elStar Dynamics

Within the active smartglass category we also include ‘transparent photovoltaic’ glass, which converts sunlight into green, clean electricity. 

Transparent photovoltaic glass can be used as windows in exterior building facades. where it is often referred to as Building-Integrated Photovoltaic (BIPV) glass.

Where does Artificial Intelligence fit into this?

Artificial intelligence (AI) aims to predict future outcomes based on historical data, rather than relying on pre-programmed rules.

The system is ‘trained’ to recognise patterns, behaviours or properties and, over time, can learn what response to execute.

Artificial intelligence (AI) is often seen as an umbrella term, encompassing Machine Learning, Deep Learning and Neural Networks.

Current examples of AI applications include automated stock trading, customer service chatbots, Netflix’s recommendation engine and self-driving cars.

AI algorithms can now drive smartglass building facades more sustainably and preempt dangerous or undesirable circumstances.

Can Smart Glass use AI to learn?

Smartglass itself is just a material, like concrete, stone or wood.

However, when coupled with an AI system, the smart glass system can indeed ‘learn’. However, this depends on there being a dataset to teach the system what is ‘right’ vs. ‘wrong’. 

And this depends on whether there are sufficient sensors capturing the events which feed this dataset.

Based on the use cases outlined above, these ‘sensors’ might include:

• AI bots that monitor social media chatter
• AI computer vision systems
• AI light exposure monitors that predict an increased risk to objects or people.

Outlook

Our journey into adaptive, AI smartglass building facades and AI-enabled smartglass for transportation has just begun. 

We have merely scratched the surface of available use cases.

Continued advances in materials and algorithms will benefit sectors where smartglass is already being used; from transportation to retail; from hospitality to heritage to healthcare.

Global initiatives such as net zero, sustainability and user privacy are currently driving us towards a future that can be beautifully enhanced by smartglass.