This Small Business Innovation Research Phase I project is developing a smart phone application that creates indoor maps through crowd-sourcing and data fusion technologies. Commercial and enterprise applications require indoor maps to support accurate location calculations and location display. In this project a global indoor map database is created though the contribution of a community of volunteers or any team that is interested in collaboratively creating indoor maps of a building, campus or city. The individuals will have a mobile app on their smart phone. As they walk through indoor spaces the application automatically creates a map using motion sensors in the phone by detecting building features such as hallways, stairways, and elevators based on the users motions. The technical challenge is to merge and fuse sensor data coming from many users carrying commercial cellular devices to incrementally update a global map of indoor spaces. The commercial challenge is to deliver an enabling technology for a broad spectrum of location-based applications. The proposed solution will provide the following: - Global data base of indoor spaces - Mapping API for mobile app developers to access to indoor maps - Map intelligence about patterns of pedestrian movement for third party companies
The broader impact/commercial potential of this project is the ability to create the indoor maps of navigable passageways that are required for a broad array of enterprise and consumer applications without requiring professional survey teams or enterprise involvement to deliver the mapping data. The TRX solution will significantly reduce the cost and time of creating and updating maps and will increase the availability of indoor mapping data. TRX will use these maps to deliver mapping data for indoor location systems to support location calculations and display users or other information inside buildings. Application developers will include providers of software to consumers, urban planners, law enforcement, first responders, and retailers. Companies interested in our mapping application will include both location-based services companies and dedicated mapping application providers: (1) Infrastructure-based Navigation providers: Navteq (Nokia subsidiary) has enabled a mobile indoor navigation system for airports by installing antennas throughout the building. Qualcomm is developing a similar Wi-Fi based solution. These solutions count on maps provided by the enterprise. (2) Indoor Map creators: Point Inside and Micello are manually creating indoor maps of important buildings by converting floor plan images into interactive maps. Again this requires enterprise involvement for establishment and maintenance.
In this Small Business Innovation Research Phase I project we have begun to develop software and a system that creates a global database of indoor maps through crowd-sourcing and data fusion technologies. This indoor map database will be used for improving indoor location estimates and routing information, as well as for displaying users’ location on mobile devices (e.g. smart phones). This project combines state-of-the-art techniques for pedestrian navigation with methods for map generation and map merging. Commercial and enterprise applications require indoor maps to support asset and personnel management including accurate location calculation and display, and indoor routing. The process of building indoor maps is different from the process of building outdoor maps since neither satellite imagery nor GPS traces for referencing the images are available for indoor spaces. A growing number of companies (e.g., Google, Micello, Point Inside) are now creating maps of indoor spaces with the majority focused on high use environments such as malls, airports, hospitals, and museums, or in creating and maintaining enterprise-specific data. The current state of the art in indoor mapping requires a manual labor intensive process to create and then to maintain maps as they change over time which is difficult to scale. The visualization files for each venue include high level shapes - such as stores within a mall - but the navigation files often do not have information on entrances, exits or other structural features required to support effective indoor navigation. Also missing are maps of navigable passageways and most likely routes required for effective indoor route planning. Populating and maintaining these navigation files requires a new scalable approach. TRX has developed patent pending technology to infer map features based on measuring a user's movement through a structure using the accelerometers, gyroscopes and other sensors available in the modern Smartphones. This work is being extended under this funding to provide a crowdsourced mapping solution. The technical challenge is to merge and fuse sensor data coming from many users carrying commercial cellular devices to incrementally update a global map of indoor spaces. This data is complementary to the data created by indoor mapping companies, augmenting it with navigation data that will further improve location estimates. In phase I we have succeeded in demonstrating the feasibility of developing a system that would allow an individual or a team with no mapping or cartography expertise to contribute to the creation of indoor maps. Using our mapping App on a Smartphone, any user can contribute to the creation of 3D indoor maps by simply walking in buildings carrying their Smartphone. As the user walks in a building the data from the suite of embedded Smartphone sensors (including accelerometers, gyroscopes, magnetometers, barometric pressure, light, GPS) is used to classify users’ movement such as walking, climbing stairs, or riding in an elevator. This information is used to create a track for that person and to infer the location of building features (stairs, elevators, rooms, entrances/exits as well as other building signatures). The tracks from multiple users are fused together to create a navigable map of buildings that include hallways, rooms, doors, building exits, elevators, and stairs and their connectivity relationships. The structural data provides essential data within map visualization files that maintain a picture of a venue or building. All the collected map data (both structural elements and sensor signatures, e.g. magnetic and radio frequency) can be used in the "navigation layer" which is a set of map files that complements the visualization files. The commercial challenge is to deliver an enabling technology for a broad spectrum of location-based applications. We have begun to work with strategic partners to ensure we can meet this challenge. Aside from the large markets around commercial location based marketing and business personnel and asset management, there many applications that will result in improved safety and situational awareness based on improved location and mapping tools. For example, as firefighters or police walkthrough in a building, our mapping tool can dynamically create a 3D map of the building in real-time that is immediately accessible to other emergency responders. Additionally, the mapping tools developed in this program can be used to improve navigation guidance in indoor spaces for blind and low vision users. Expanding indoor maps - in particular the essential map navigation data - from only high use locations to virtually all public spaces will also dramatically improve accessibility of such buildings, supporting enhanced location for a broad array of specialized third party applications.
