Pre COVID-19, Indoor Mapping was already one of the highest growth location analytics use cases, with hospitals, shopping malls & airports all investing in technology to gain deeper insights on consumer & patient behavior inside their properties.
Today practically all smartphones, wearables and vehicles have an advanced Global Navigation Satellite System (GNSS) onboard allowing us to accurately pinpoint our location and use a variety of location-based services. Due to the nature of the system this accuracy is very high whenever we use such technology outdoors but can often suffer when moving indoors. As the expectations of consumers change with the prevalence of food delivery, ride sharing and other apps using location, they are demanding more context based and accurate information. We have perhaps experienced this ourselves when requesting a taxi only for it to arrive far from where we actually are due to the inability of our device to accurately pinpoint our indoor location.
Using an industry example, the holy grail for retailers is to understand where exactly customers are within a store, what specific products they are viewing and for how long. Why did they pause at a specific point or follow a certain path within a shopping center? This is similar to the sort of analytics common amongst online retailers and is currently being replicated in store environments using radio (WiFi, RFID, BLE, UWB) and light (LiFi, IR, CCTV, cameras) based systems.
In order to improve the accuracy of Indoor Mapping Software in these and other scenarios, Focal Point Positioning and others, take an approach of ‘sensor fusion’ by intelligently combining the wealth of data being tracked by the myriad of sensors now commonplace within the technology we use, carry, and wear. This means that instead of using simply radio and light based systems as described they can also leverage data from sources such as the barometer, microphone, and inertial sensors resulting in a more ubiquitous outdoor and indoor relationship and therefore richer context.
Crucially this means that no additional infrastructure is required and as can be seen in the map below, analyzing this data with human motion modeling results in more accurate and reliable positioning. The area being shown is Canary Wharf in London, an area dense with high rise buildings which can often cause issues for GNSS devices. The yellow line shows the route as tracked by a very high end military class GNSS device representing the ‘ground truth’. The red line shows a standard smartphone with unassisted GPS with the blue line showing the improvements made by combining sensor fusion, machine learning and signal processing.
Precision has never mattered more when considering the current climate regarding social distancing guidelines with the key advice to ensure that a 2m distance is kept between those outside of our households. With regards to contact tracing apps the most common technology being utilized is Bluetooth. Again, sensor fusion can play a role in ensuring the effectiveness of these types of apps given that the Bluetooth sensor may determine a nearby phone is within 2m but it may not be the case they are sharing the same air (being separated by a pane of glass for example). Therefore being able to use additional sensors, such as comparing the sound around the devices or barometer data can provide greater precision and context.
Looking at an example within the retail industry, the following map shows the route taken by a consumer with a standard Android smartphone within a shopping center. The red dots represent the Android fused location estimates, which is a combination of GPS, WiFi and cellular positioning. The blue line represents the human motion modeling system by Focal Point Positioning. When outside the system uses GPS data to tune various aspects, learning step bank and stride length and context the phone is carried in. When moving inside and with the loss of GPS, data from the accelerometer, gyroscope and barometer is translated by the system into a centimeter level reconstruction of how the smartphone and therefore consumer moves through space.
The blue line is able to fill in gaps where the smartphone went through areas with no WiFi with much richer detail as to how this individual moved through the center. For example for this example of a shopping center in Cambridge we are able to pick out a loop where a visit to the Tesla store was made as well as the route within the nearby Apple store.
The traditional use of existing WiFi infrastructure and beacons makes it a challenge to determine which floor and room the device is located since the radio signals propagate through walls and ceilings. By again using the inertial sensors and human motion modeling a 3D model can be reconstructed of the motion through the building and better understand the transition between floors.
This can be important in a number of use cases including retail, to determine where people are spending their time, healthcare, to locate those in an emergency and construction, to move from paper based interactions to a digital paperless construction site (particularly challenging given the constant evolution of such a site and requiring an infrastructure free approach).
This is something that can be achieved with the integration between CARTO and Kepler.
Want to try indoor mapping yourself?Sign up for a free account
We’ve worked with lots of different customers and partners (such as Situm) to help them optimize their indoor decision-making with location data. We’ve listed some of the industries that can benefit from indoor mapping, along with typical questions that can be answered.
During the webinar we demoed an indoor mapping solution for department stores which included a look at;
To help you answer these questions and more, our expert Professional Services team has years of experience building Indoor Mapping applications, as well as going beyond visualization to apply spatial analysis for large facilities.
In these times, as we return to indoor spaces other than our own homes, both in the short term to offices and in the medium term to larger scale indoor events and gatherings, Indoor Mapping and working to make it as precise as possible has never mattered more.
Want to know more?Watch the full webinar
Like many people who love trees and work in the geospatial field, I was fascinated (and disheartened) by a recent article I read in the New York Times called Since When Hav...Use Cases
Most Data Scientists and Analysts understand that visualizing datasets can be a crucial way for users to engage with data. Knowing where median household income is across a...Use Cases
The urban growth of metropolitan areas around the world can be affected by a number of factors. During the industrial revolution the explosion in job availability fueled mu...Use Cases
Please fill out the below form and we'll be in touch real soon.