- Construction engineer, U.S. Dept. of State »
- ELECTRICIAN »
- Senior Director of Facilities »
- Sr. Building Systems Engineer »
For Smart Buildings, Artificial Intelligence Is Still a Journey
March 13, 2019 - Building Automation
By Ken Sinclair
Disruption is everywhere, and open is on our mind and in our software and hardware. Autodidactic, DIY is how we learn. The discussion below and this “Just Do IT” review provides resources to help kick start your digital transformation. Focus: We are the intelligence in AI not the artificial.
The term "AI" — artificial intelligence — seems to be troweled on top of everything these days. But I feel we are a long way from its true definition.
Our industry is on a journey towards “automated intelligence” and that, for now, is the "AI" we are talking about. The interlocking "AI" — autonomous interaction — occurs when we start automating intelligence. It is early days, but we need to learn how to walk before we can allow true "AI" to run our buildings.
Our existing buildings all need to be smartened, retrofitted as exemplary examples of automated intelligence an interaction.
Automation is basically making hardware/software that is capable of doing things automatically. AI is all about trying to make machines or software mimic human behavior and intelligence. Automation may or may not be based on artificial intelligence.
We are still struggling with self-learning, machine vision, voice interactions which all need to be solid before we can enter the true world of AI. I love this definition: "AI is whatever hasn't been done yet."
We are now in the exciting era of automated intelligence because we are doing what can be done with today's tools while leaving what hasn't been done for future definition of artificial intelligence.
I feel our journey is close to that of the autonomous car, a vehicle that is capable of sensing its environment and moving with little or no human input. I believe that we are striving to achieve autonomous interactions in our buildings. This interaction is a part of what I am suggesting might be "building emotion."
We have much to learn from automotive industries reinvention.
I asked Sudha Jamthe of IoTDisruptions for advice and she provided these comments,
"1. AI today is all about building the technology to capture what we do as humans and trying to automate it using machine learning to build models based on training data of past behaviors. Another industry term for this I see is "augmenting intelligence." It is about using AI to help us do our work better instead of replacing us, e.g., make decisions faster with a huge volume of data.
"2. I like your point about the comparison with the journey of autonomous vehicles. I see autonomous vehicles as setting the vision for how vehicles will drive themselves and free us of the risks of traffic deaths and the cost of owning cars. But getting there is a journey which is more exciting for me because the car is not going to get there alone. The city infrastructure, buildings, parking, energy sources all have to become smart to interact with the autonomous car. That journey is going to create a connected world which I call the driverless world.'"
Join Jamthe for coverage of #StanfordEnergyWeek and the #EnergyAI interview with Rebecca Wolcoff, #MachineLearning engineer from Enel X, as she talks about real-time energy storage for #SmartBuildings .
Over a year ago I wrote this editorial: Building Brains: AI is the brain, IoT the body speaking with the face and voice of digital transformation.
“Machine learning or artificial intelligence ("AI") is poised to take our industry quickly, in directions never before conceived. ‘Building Brains’ with its IoT body, face, and voice is changing our world faster than we can imagine; actually, digitally transforming it.
“Building Brains will evolve rapidly as more and more intelligence devices come with their own machine learning and AI. The challenge will be to manage these bit brains on the edge and use that intelligence for the greater good of the building, the campus, the city, and the world. The new task is to build building brains with the myriad of edge brains creating a virtual intelligence digital transformation community for our purpose. In the past, machine learning and AI only resided in the cloud but with the development of open source learning, voice platforms, and low-cost processor/memory at the edge, bit brains are now everywhere.”
More on AI from the March issue of AutomatedBuildings.com: Cutting Through the Hype Surrounding Artificial Intelligence in Smart Buildings:
“The talk of AI might always sound like the distant future but the technology, in different forms, is already around us.
“While there is still some debate, leading experts generally divide their anticipated evolution of AI into three generations, defined by the limits of their capabilities. The technology we experience in our smartphones, tablets, automated building systems, or other platforms is restricted to artificial narrow intelligence (ANI). In the future, we should expect more advanced forms when we reach artificial general intelligence (AGI), and then artificial super intelligence (ASI). There is still a lot of benefit to be gained from ANI and, as is the nature of AI, even narrow systems learn over time to become more and more ‘intelligent.’”
"In a world awash with data, AI can help to find ‘the signals in the noise’ by identifying anomalies and patterns, then drawing out actionable insights. AI’s advantage over human intelligence is that it can process huge volumes of data that a person or team could not feasibly analyze in a reasonable timeframe. By using historical patterns to predict future data quality outcomes, businesses can also dynamically detect anomalies that might otherwise have gone unnoticed or might only have been found much later through manual intervention."
Buildings are the destination of the autonomous car. So they will need to look like them and share similar attributes: “While there is rapid innovation in automotive technologies, the infrastructure to support the above vision of sustainable transportation including automatic traffic light sensing by the vehicle, instantaneous traffic data, vehicle to infrastructure (V2I) communications, vehicle to vehicle (V2V) communications, managing and delivering electric power wirelessly to EVs (including shuttles and people movers) while in motion, managing traffic dynamically based on real-time and historical data, detecting and avoiding pedestrians, and others, are in their early stages at best. This makes it challenging for vehicle manufacturers to innovate towards fully autonomous and all electric vehicles beyond a point. Innovations in the areas of communications, sensors, GPS, software, cloud computing, controls, energy storage, power management, battery technology, wireless EV charging during operation, cybersecurity, big data, AI, machine learning (ML), data science and block chains are paving the way to create the smart urban transportation infrastructure that would enable the above vision of a modern sustainable transportation future.”
Google Home's Assistant could one day know your Mood. Take that Alexa. “In the next five years, Huffman suggests, the Assistant could achieve the basics of natural human conversation, which, from a computer science standpoint, are anything but basic. He says wake words like ‘Hey’ or ‘OK’ are ‘really weird.’ He wants the Assistant to understand your mood and tone, and detect if you're frustrated. He wants the software to remember an exact discussion you had with it yesterday, so that today you can pick up where you left off.
“I ask him about the vision for 10 years from now. Maybe, he muses, physical robots — not just bots you can talk to, but robots that move and do stuff — will become household products, and digital assistants could integrate with them.”
Now that is the building emotion I am talking about.
New ways to interface Messaging as a Platform: “The state of human to machine communications conversational user experiences, in the form of chatbots and voice interfaces, are overtaking many of the traditional ways in which we interact with machines. Since the rise of computers, human-machine interfaces typically had some form of graphical user interface (GUI) which enabled direct (if limited) interaction with devices and their programs, for instance via software installs, mobile apps, and web-based applications such as software as a Service (SaaS). No matter how ‘beautiful’ the respective interface, this GUI is now more and more replaced by a conversational user interface (CUI).”
Conversational User Interfaces – Text, Voice, and More! “These CUIs come in many shapes:chatbots on popular messaging apps such as WhatsApp, WeChat, Telegram, and iMessage, and, more recently, voice-activated devices and personal assistants such as Alexa, Siri, and the Google Assistant.”
Here is a very good resource for our journey to socially acceptable autonomous buildings: “The IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems is launching the second version of ‘Ethically Aligned Design: A Vision for Prioritizing Human Well-being with Autonomous and Intelligent Systems’ (EADv2), the most comprehensive, crowd-sourced global treatise regarding the ethics of autonomous and intelligent systems available today
Without smart buildings, a truly smart city can’t exist, says Dave Hollander, Bluetooth Sig: “As self-contained structures, smart buildings address the need for automation, control, and monitoring, while serving as the building blocks of a scalable foundation towards the larger smart city. Without smart buildings, a truly smart city can’t exist.”
“Smart Buildings as a Service” – sometimes called “servitization.” “Data from these smart building systems give a facility’s infrastructure a brain and a voice. This data is put to work through smart controls for buildings — whether in the public sector or commercial — which give buildings a ‘central nervous system’ that balances and reconciles competing interests such as energy minimization, occupant comfort, and grid stability.”
2019 is the year we get smarter about smart technology, smart equipment and smart solutions. “The built environment has been changing dramatically over the past five years. Technology shifts along with changing value propositions are making an impact on how we operate and manage our smart buildings. Our dialogue is no longer about the potential — rather it is about the reality of economics through driving operational value and business outcomes. The business of smart, connected commercial buildings and facilities has moved beyond the unknown and into the new era of real and relevant. 2019 will surely be another transformative year, not only in terms of technology but also in a fundamental shift in the conversation driven by digital transformation, connected data, and new mindsets.”
On closing this chapter please read the following piece of history but replace the word web-convergence with next level "automated intelligence and autonomous interactions." Good advice from our history. Looking back while looking forward from 2002— our industry created a stable of "Awful Mated Buildings."
“The first few years of the DDC revolution our industry created a stable of ‘Awful Mated Buildings.’ I fear this could occur again with our first attempts at web convergence.
“The large building automation industry understood early the concept of web convergence, but I do not believe that we envisioned ourselves as having such an active part in it. Originally, we perceived that we would take our products and services to a web level and magically our services would become part of the clients web-based enterprise. It is now becoming clear that the companies that are most likely to succeed are developing the ‘magic’ to take our automation interfaces to the next level in an elegant convergence model that adds tremendous value to our clients' enterprises.
“In Tom Hartman's prediction from last month's building automation column ‘We the People’:
“I predict a very strong movement to 'occupant integrated' HVAC controls within the next decade. My prediction is that by the second decade of this century, most class 'A' office spaces will be required to offer individual control of thermal and lighting levels. This integration will most likely be Internet-based. Also, in the same time span, I see much building operations and maintenance will also become Internet-based with fewer operations personnel in individual buildings. Such personnel will become more specialized and perform a smaller scope of services in many more buildings.
“My call to action: Our industry has a limited window of opportunity to control the successful convergence to our web-based clients. We have the knowledge to make these transitions work better than anyone. We know that we must significantly increase our knowledge of web-based ways and add ‘web heads’ to our corporate structures. We can hire or create within but our focus for the next few years must shift from hardware integration to successful web integration and convergence. It isn't that the evolving hardware integration standards aren't important, it is an issue of, if we do not complete our connection to the web world, we run the risk of losing that large market in the future. As the cost of hardware drops and engineering of DDC systems becomes automated, our industry will shrink if we do not move quickly to the next level. There is much work to be done at this level and this is likely the closest we will ever get to the ultimate interface with the end users of our building automation systems. The importance of this and the speed at which we move is cardinal. Our industry is about providing client comfort services. Our clients' information flow model is becoming web-based and the transition to web-based solutions is not only desirable it is becoming mandatory.”
As an industry "AI" no matter how we define it is the next big thing we need to integrate into our services and products. We need to envision ourselves as the intelligence in "AI."
Finally, a couple of noteworthy events coming up:
Controls-Con 2019 Program Details Released! "More than 20 industry thought-leaders join the speaker lineup at this Smart Building and Building Controls Conference to educate attendees on the latest strategies to make dumb buildings smart" — Kristina Reid, Marketing Manager, Cochrane Supply & Engineering.
Project Haystack Organization Launches New Marketing Website "This new website demonstrates the growing acceptance of the value of the Haystack methodology” — Robin Bestel, Marketing Manager, Project Haystack.
Ken Sinclair is the founder, owner, and publisher of an online resource called AutomatedBuildings.com. He writes a monthly column for FacilitiesNet.com about what is new in the Internet of Things (IOT) for building automation.