Since indoor navigation is one of the hottest topics within the mixed reality industry, numerous companies have developed solutions to address this matter. In this article we want to dive into the different methods of implementing an indoor navigation tool and highlight their limitations. Furthermore we will reveal what we currently consider to be the most effective approach.
Visual recognition solutions work especially well in small sized areas. By sampling the area in question from different angles through a camera lense you create a so-called visual point cloud, which the indoor navigation system will later on use to locate your relative position by comparing your actual view with the previously generated visual point cloud database. However there is a down-side to this method. It fails in areas bigger than 150 square meters.
Markers / QR Codes
Another tracking approach is, as already mentioned above the marker-based approach. In this case you have to install markers across the area you want to cover with your indoor navigation solution. By scanning one of these markers the system will immediately determine your precise location within your covered area. That all sounds pretty straight forward but the problem with this method lies in the details. For starters, customers mostly perceive markers as ugly and installing them is tricky as well as sometimes forbidden. Then there is the issue of people placing objects on top of or in front of the markers or even damaging them which makes them lose their functionality. And last but not least: This method always requires an action by the user, which inhibits acceptance.
Bluetooth beacons as an alternative
One of the recently emerged alternatives are bluetooth beacons, which address some of the negative aspects of the previously described methods. First of all beacons are easy to install as they can be placed on any surface and most importantly on pillars and ceilings. Second of all beacons don’t come with a big price tag and have a battery life usually between three to four years. The difficulty with beacons however is that they only provide a position and no orientation. Using the compass for the orientation is not an option since the compass has massive drifts inside buildings which adds up to a massive drift once you walk a few meters. Beacons are by all means not a solution for a reasonable indoor navigation.
The ViewAR Tracking Fusion System
All this made us to come up with a solution that just works. We combined ARKit/ARCore movement with beacon positioning to create an accurate positioning and orientation system, which we then called the ViewAR Tracking Fusion System. Check out the video below to get a first glance at our approach in action:
It is obvious that our system dramatically enhances accuracy and improves positioning. The system uses the first beacon position and the compass for the initial tracking, then we collect further beacon measurements while walking and continue to optimize the tracking. After a distance of around 20 meters, we reached an accuracy of around 0.5 meters, more than 10 times as accurate as standard beacon positioning.
As you can see in the video we even incorporated our virtual guide called GuideBOT which is a 3D-animated character with an AI chatbot. The interactive bot with voice recognition and text input not only guides users around, it also answers questions and gives advice and recommendations. It therefore adds an incredible feature to any navigation system.
Benefits of GuideBOT
By making it possible to interact with the virtual character, implementing GuideBOT into your indoor navigation system will enhance the customer experience by miles. Furthermore you can collect direct feedback from your clients by letting GuideBOT ask for it or gather invaluable insights into customer journeys and interests.
If you want to learn more about our Tracking Fusion System or are interested in an early beta access, please feel free to contact us here.