Using People Counters to Track Occupancy? There’s a Better Way.
Tracking occupancy in spaces such as buildings, meeting rooms and more is not a new concept. Building operators and facility managers have been tracking occupancy using a variety of methods — either manually or using technology — for decades. Understanding the occupancy for a space is used for a variety of purposes, including better planning, optimizing operations and, more recently with COVID, limiting occupancy.
While these traditional methods may provide a sense of building or room occupancy, they all have fundamental limitations that mean organizations are missing out on the true value of real-time, advanced occupancy data to improve their spaces for employees and customers.
With recent innovations in computer vision, artificial intelligence and edge computing, solutions are now available that deliver real-time and highly accurate occupancy data in ways that have never been possible with traditional methods. These new technologies open a whole new world of opportunity that enable significant energy savings, improved sustainability and enhanced space planning and operations.
This week’s blog post focuses on the various types of sensors used in the built environment today, their limitations and the new alternative to these traditional methods that is more scalable, extensible and cost-effective.
Several options exist in the market today for people counting, including devices that use infrared cameras and LiDAR to detect people entering and exiting a space. In cases where this technology is used to count individuals, it can be accurate within a tightly controlled environment. However, given that these sensors are designed with one function in mind, they are unable to detect any other metrics beyond simply counting people, which significantly limits their utility and scalability. They are also ineffective at measuring large areas which means that customers are required to install many sensors within their spaces, exponentially increasing the investment needed to get an accurate people count.
CO2 sensors are very commonly used in the built environment for demand-control ventilation. By determining the amount of CO2 in a space, they can estimate the number of people in a room and adjust ventilation accordingly to ensure the air is fresh. However, they have a number of limitations. First, they take anywhere from 15 to 30 minutes to estimate the number of people in the room, which can lead to occupant discomfort (a stuffy room). Their accuracy also “drifts” over time and need to be calibrated every six to twelve months which can be costly, and often doesn’t happen leading to incorrect CO2 detection. Their lack of precision means they’ve never been used to generate detailed accuracy data for planning, let alone controlling heating/cooling.
Motion detectors are mostly used in the built environment to turn on and off lighting. While this option is low-cost and broadly effective at sensing if there is movement in a space, it is a binary measurement that only detects movement, not people, which means they cannot tell how many people are in a space. As a result, motion detectors cannot be used to gain any useful information about actual occupancy numbers and cannot be used to collect building usage or more real-time applications such as controlling ventilation and heating and cooling that modern buildings need.
Beacons detect how many Bluetooth devices (phones, computers, watches, earphones, etc.) are located within a set area. While this can be used to estimate the number of devices a space, it does not necessarily translate to a precise count of the people in the space since most people these days have multiple Bluetooth devices on them. To get a correct occupancy count, you need wearers to register one of their devices to be counted, which is often not feasible and raises privacy issues. As a result, Bluetooth has not been used to provide precise data about occupancy for planning or HVAC controls.
Introducing Nomad Go’s Advanced Occupancy Sensors
By leveraging new advances in computer vision and artificial intelligence, there is now a vastly better alternative than the sensors listed above. This new solution by Nomad Go generates occupancy data with a level of detail, accuracy and speed that has not been possible up to now. In addition, Nomad Go uses the same generated data for a variety of additional applications, from better control of HVAC for energy and savings, to more efficient operations and space planning, to gaining other insights about occupants such as mask compliance.
What makes this new solution unique?
- Real-time and highly accurate: data is collected and processed instantly, giving you a much more precise view of what is happening in your spaces and the ability to send real-time alerts about the space.
- Hardware independent: All the other sensors listed above use propriety hardware. Nomad Go’s solution leverage either smart devices from Apple or Google or smart cameras from NVIDIA, Sony and others. This makes them highly flexible, easy to deploy (everyone knows how to use a smart device), are supported by an ecosystem of established software to manage them (Mobile Device managers). Furthermore, there is an endless supply of these devices which removes potential supply chain issues.
- Flexible placement: by leveraging commodity hardware, it means that they are quickly installed in physical spaces and can be easily moved if needed.
- Extensible: This capability is the most important aspect of the solution. Computer vision doesn’t just measure the number of people in a space — it can tell you about everything it sees. This means that you are building a digital backbone for your built environment. You can start collecting one metric such as occupancy, and over time can add mask usage, dwell time, hard hat compliance and so on using the same sensor. Because Nomad Go is an app running on top of commodity hardware, it can be updated with new metrics over time. Contrast this to any of the sensors listed above which are single purpose.
The combination of these features provides the ability to generate real-time data to control HVAC and ventilation, as well as real-time alerts about what is going on inside a building, delivering the most robust set of actionable insights about your spaces.
to learn more about our end-to-end computer vision solution for the built and retail environments and how they can help you streamline operations, lower your energy costs and improve sustainability.
Originally published at https://www.nomad-go.com on February 16, 2021.
