Geoff Mair is the CEO of SensorUp. They are a geospatial movement intelligence software company connecting people, equipment, and fleets in the field. Combining the world of geospatial with the Internet of Things (IoT) andArtificial Intelligence (AI) and provides real-time process management.
It’s the ability to connect low-cost sensors over the internet and track devices, equipment, and people in real-time. With all the innovation that’s happening in this space right now, there is a real explosion of these low-cost sensors out in the field, in people’s homes and businesses. It’s the next generation of the internet.
Everything happens in space and time. Thus, everything we track and gets tagged from an IoT and movement intelligence perspective is also keyed on space and time. All other properties vary depending on the type of sensor that’s providing the input. If you’re tracking people, that may be body temperature and heart rate. If you’re tracking a truck, it’s direction, speed, and fuel consumption.
For some things, yes.
If you’re tracking temperature in a home on a thermostat, the home is staying in one place. It’s not moving around. The location becomes a valuable component when you roll it up and look at temperatures in homes across the country to see what’s the average energy spend is on homes. To go from a single device use case to multiple devices, you do need to have location. The real power of IoT comes in when you have assets moving around in space and time.
When IoT collects data and doesn’t have the context of location and time, the value of the data is significantly less. Everything should be tagged in space and time, really. Once organizations start collecting location and time correctly, they get the most intelligence out of their data.
Dr. Steve Liang, worked with the Open Geospatial Consortium, NATO, and the UN to create a standard called the SensorThings Standard. Disparate types of sensor data and other types of non-sensor-related data can be read in real-time and aggregated into a standard format that’s based on observations tagged in space and time.
Think of it as a root observation plus space and time.
These standards are being adopted internationally, and there’s a lot we can do with them now. Public safety and military spaces combine scores of different types of sensors for various use cases and do it across different organizations. They can now use a standard data format that can be read by different systems. IoT connects the many devices and different root data formats into a standard format where everything is aggregated and consolidated.
Public safety is where the roots of the technology came from. It’s all about responding to real-time events and coordinating responses across police, fire, hazmat, and other departments.
To do that, we add several body sensors to first responders and connect them through Android and the SmartHub software we have. This makes the responses more effective and efficient in the case of critical events ̶ saving lives.
Oil, gas, and mining companies benefit in terms of process efficiency and effectiveness. They can track people, equipment, and trucks. By layering on our geospatial movement intelligence software, they can run historical reports showing them where the bottlenecks are in their operations. They automate steps in the process by detecting certain conditions and using the workflow software to automate responses to that.
Suppose you spend a large amount of money on wastewater hauling. If you can track the fullness of your wastewater tanks and predict when they’re going to be full, it’s possible to locate and send the nearest truck and do just-in-time wastewater hauling ̶ saving you money on empty runs.
Rail networks, such as the ones in Canada, track rail cars with our technology to detect bottlenecks and make their logistics more effective. They can trace their rail cars historically and keep a tab on merge fees (fees for cars sitting in a rail yard.)
No. The core is pulling data based on the SensorThings Standard. We build second-order metrics on top of that and provide visualizations. There is also a workflow engine that allows us to build business rules or use more complex AI to detect things going on in the real world and program responses to them. If a piece of equipment has more vibration or heat than usual, a ticket can be generated in the field service management system, and a technician is sent out automatically to check it. We can build those rules in space and time and combine geo-fences with physical things. The rules can be geographically based.
In the case of location, it’s always more accurate to use multiple location inputs and combine them. We can combine GPS data with other location tracking systems and legacy location tracking systems like RFID. By doing so, we get a much more accurate, real-time location than what we would typically get.
In the case of Fred, the fireman, who is part of a response team, he can be modeled in the system as a combination of his heart rate sensor, body temperature sensor, and the location data that’s coming from his phone. The same can be done with equipment, not only humans. A pump that you’re monitoring could have twenty sensors on it drowning you in data. Our software clusters sensors together and customers only get the information associated with the pump as a single unit and not from the individual sensors.
Business rules are then built around “Fred” and the “pump” rather than their individual sensors. This is what’s valuable to our customers.
You can take your observations and uplevel those with second-order metrics. Take raw location and time data for a rail car. If what you really need to know is the average dwell time per car, you can derive that second-order metric from the raw data. You can even calculate what’s the average cycle time per car from the same data set.
Yes. We are directionally investing heavily in AI so that we can move into automating businesses based on things that are not only happening right now but how they are expected to occur in the future.
Is There a Particular Industry That’s Just Waiting to Be Disrupted By The Spatial Internet of Things?
Industrial IoT will be a game-changer andthe next Industrial Revolution. We’ll see massive productivity gains from IoT being adopted in industrial settings in the next 10 years. Oil and gas, mining, and industrial rail logistics haven’t been as fast to adapt to newer technologies, so there’s an excellent opportunity for them to make significant investments in IoT and get productivity gains out of it.
These sensor networks are already present in our lives, sometimes even making decisions for us. Just think of Amazon Alexa or Google Nest. They are consumer IoT – tracking information, getting value out of data, and providing it back to the customers. The innovation that’s going on with low-cost sensors will fuel this massive explosion. The pace of change in the next 10 years will be significantly more than in the last decade because of the availability of low-cost sensors and their ability to help us in our personal and business lives.
People will always have concerns about their privacy, and so they should. There will come a time when the benefits outweigh the risks and concerns, and we are already starting to see that change in public safety.
A few years ago, the average fireman would be defensive about their whereabouts being tracked. These days the attitude is changing, and firefighters and policemen understand that sensors could help them with their health and safety as well as being useful in an emergency event. People are wanting to see this adopted to ensure safety in dangerous situations.
The Gartner hype cycle is a curve that follows an innovation trigger. It goes up steeply to a peak of inflated expectations, and it drops down to the trough of disillusionment only to slowly climb up to what’s called the slope of enlightenment and the plateau of productivity. All new technologies go through this cycle.
I believe we’re currentlyon the downslope with IoT. A year and a half ago, the hype around it was much bigger. The technology is maturing, and we understand it better. Once you get through that trough and reach broader adoption, value gets created. An even more widespread adoption is coming just around the corner, and then we can get into maturity and get better business value out of it.
Standards need to be applied for IoT to be successful, just like the internet gained traction and adoption from standards. If everyone does their own thing, the value they’re creating is lower because you can’t get all the nodes talking, and the data set is incomplete. We must continue to invest in developing and promoting them with our customers and partner organizations.
Would you consider sensoring up, or have you already kitted your home out with Alexa or Nest? Did you have any privacy concerns? Do the benefits already outweigh the risks, in your opinion? I would love to know what you think.
Be sure to subscribe to our podcast for weekly episodes that connect the geospatial community.
For more exclusive content, join our email. No spam! Just insightful content about the geospatial industry.
To put it simply, point clouds are a collection of XYZ points that represent some real world object of nearly any scale.They can be generated in a few ways. As geospatial scientists, we mostly work with LAS/LAZ data collected by aerial LiDAR (light detection and ranging) scanners at varying scales, from landscapes, down to project sites. We may also derive point clouds from highly detailed orthoimagery of an area, such as from the products of a drone flight.
As a data scientist, you don’t just go in and solve problems. You make recommendations to multi-faceted issues so that you get a fantastic model in the end. You’ll also be advocating a better use and understanding of the data while you do that.