MAP White Paper Lite Version

*This article serves as a quick reference of our platform. Please see our full technical white paper for further information.

We intuitively map our personal world, which allows us to navigate certain routines on “autopilot”: we’re familiar with our neighborhood, where to stop for mid-day coffee, and depending on the hour, we know what kind of traffic to expect. We know these things because our memory maintains a working knowledge of our surroundings.

To supplement our spatial knowledge and optimize our routines, we have come to rely on our devices for information, like to know a store’s operating hours and then to provide the best possible routes with their respective ETA.

Our devices adequately resolve these kinds of inquiries, but as we evolve towards automation and augmenting our real world experiences, our location-based apps have fallen short in providing a richer experience.

In order for emerging technology to become part of our daily routines, our devices need an adaptive framework that can maintain a working knowledge of a world that is in constant change. This precise map of the world for our devices does not yet exist, but we believe we are at the precipice of creating a living map that can provide precise indoor and outdoor location and positioning data. Mapping Aggregation Platform (MAP) is our solution.

MAP Overview Video: https://www.youtube.com/watch?v=QDO2x8W2qn4

tldr; What is MAP?

MAP is a conceptual decentralized protocol that enables the gathering and sharing of geospatial data through the blockchain. To create this map, the community contributes mapping information that is used to build a digital representation of the world. To enrich the map, the community embeds location-specific information on the world’s digital representation. Contributors of accurate submissions would be rewarded with MAP Tokens and MAP Token users would then embed digital information onto the blockchain.

Community cartography can be unreliable and impractical, but with MAP, community cartography would be a simple and intuitive process. Anyone can be part of creating the open map of the future.

Current consumer grade mobile devices can now gather 3D information about the world around them. If a user engages an augmented reality (AR) app with the MAP SDK installed, they can gather and submit 3D mapping data as they go about their session, without ever disrupting their experience.

The MAP protocol would consolidate the individual perspectives of millions of devices to create a 3D map that enables every device to understand exactly where they are in the physical world.

MAP is not building a digital representation of the world with images; MAP creates a wireframe of the world by capturing features such as the edge of a door or the surface of a wall.

With this digital representation of the world, anyone can build high-quality and accurate location-based applications and services on top of the MAP blockchain. Information on the MAP blockchain will be fully transparent and open to the public, empowering everyone to access what information is being shared and how it is being used.

A decentralized map serves our individual and collective interests. The up-to-date 3D maps we have now are proprietary, and access (with restrictions) can cost a fortune. This can be a significant barrier for innovators and developers who want to build data-rich and accurate location-based apps. Location-based apps include, but are not limited to, augmented reality, entertainment, social, emergency and non-emergency alert notifications, and pathing for autonomous drones. As for end users, they want access to a myriad of high-quality, location-based experiences without subscribing to a specific platform. The MAP blockchain would ultimately allow for an interoperable “write once, use everywhere” model.

In an ever-changing world, we need the contributions of millions of people to supply a data set that is larger and more complete than what can be mapped by singular entities. Decentralizing our maps will allow all devices of any platform to know more about the the world via a self-sustaining, incentive-supported distributed commons.

What’s Wrong With Our Current Positioning and Mapping Technologies?

MAP is not designed to replace GPS or existing map applications. The MAP concept is designed to address the need to anchor digital content and devices to a decentralized, always up-to-date, digital map that anyone can read and write. With that said, there are a number of reasons why today’s maps are inadequate for AR and autonomous vehicles and drones. Three dominating concerns are:

(1) Accuracy of positional technologies like Wi-Fi, Beacons, and GPS are imperfect.

• Today’s GPS-enabled mobile devices are, at best, accurate up to approximately five meters. This accuracy suffers in any environment where the line-of-sight can be obstructed by anything from tall buildings to tree cover. To compensate, we also adopted Wi-Fi Positioning Systems. But location information gathered from wireless access points is still coarse. For augmented reality experiences, meters in accuracy can make the difference between a useful and inadequate location-based app.

(2) Satellite, aerial and vehicle surveying are out of date from capture to availability.

• Aerial and vehicle surveying are great assets for mapping out our world. However, because a useful AR experience requires a high level of accuracy, surveyed content is out of date from capture to availability. Additionally, these methods cannot capture 3D map data for indoor spaces, where many of us spend at least half our day.

(3) Dominant companies keep map data proprietary.

• Large mapping companies silo their mapping information, choosing which ways and what methods their information can be accessed and used. This causes unnecessary fragmentation in both purpose and availability; proprietors of map data maintain their dominance over mapping markets, but do so at the cost of everyone else involved.

Why Decentralize Mapping Data?

When we use major location-based apps and services, we accept user agreements that co-opt us into helping these companies build out their proprietary maps. These user agreements exist because companies know there is functional and economic value in crowdsourcing mapping data (humans survey the earth orders of magnitude more than what any one company’s fleet of cars can capture). Yet, despite our assistance in building out these maps, we don’t have access or permission to use our collective location data freely.

As we know, an accurate 3D map of the world has purpose beyond safe autonomous vehicle navigation, and cross-platform persistent augmented reality content; it can facilitate innovation, democratize the development of location-based applications and services, and establish a transparent reading and writing of our Earth’s information. Mapping on the blockchain is a way for us to restore agency and ownership in the creation and usage of our maps, without third party control or censorship.

In addition, when mapping data is gathered and stored in a decentralized fashion, it allows for data generation that reflects the world’s changes. This is great for companies or products that must keep up with a world that is constantly changing. Mapping data is also sizable, which becomes resource intensive to process at a global scale. Adopting a decentralized, open source protocol lifts this burden by changing the distribution of mapped and stored data.

Why Have A Token?

To generate a tangible incentive for community cartography, the MAP network proposes fixed rewards for submitting mapping data from any place to the MAP Blockchain.

To encourage continuous mapping of places, Region Stewards would receive and disseminate bonus rewards for mapping places that experience frequent changes, have not received enough recent attention, or are in demand by Token Users.

Mappers (can be individual cartographers or a developer with an augmented reality application with the MAP SDK installed) would choose to keep, sell, or use their earned MAP Tokens.

MAP Tokens could be used to embed location data onto the MAP Blockchain. The location data would then be leveraged by developers to make their services even more robust (accurate AR content placements) or it can be general information usable by anyone (store menus or hours). The cost for embedding a Universal Resource Identifier (URI) is determined by the Region Steward. A portion of MAP Token transactions are shared with the Region Steward of that location.

Essentially, the core loop incentivizes:

• Mappers to continuously map the world for MAP Tokens.
• Token Users to use MAP Tokens to embed and append location data.
• Region Stewards to ensure the map is up-to-date.
• Nodes to participate in validation, storage, and delivery.

This is the end of the high level overview of MAP.

If you want to learn more, the next part goes into deeper detail of MAP’s use cases and how MAP gathers data, fuses data, and embeds data.

— — — — — — — — — — — — — — — — — — — —

How Would MAP Gather Data?

Major hardware and software companies have released, or have scheduled releases, for consumer products that enable 3D image data. Apple’s most recent 3D-ready consumer devices, iPhone X and iPhone 8, are either equipped with depth-sensing cameras or can infer depth using computer vision. In an effort to further community development of AR experiences, Apple’s iOS 11 update also released ARKit, a framework for building augmented experiences. Google’s ARCore, which provides 3D image data and a new framework for building augmented experiences, has been available to consumers as of last year. Hardware manufacturers like Intel, Microsoft, and Magic Leap are also actively developing and releasing devices with 3D sensors and depth cameras. These powerful consumer ready devices are key in how we could crowdsource and crowdshare an accurate map of the world.

The MAP protocol enables mobile devices to gather and submit 3D features of the environment while the user is engaging an AR app. MAP is not building a digital representation of the world with images. MAP intends to only operate on 3D features to wireframe the world, like the facade of a building or the edge of an entryway. Although features like the facade of a building may not be unique, they become uniquely identifiable when combined with other environmental features.

Submitted 3D data is cross-referenced across the MAP network to verify which features are reported by all participants. Using this framework, anchor points are established, allowing devices to localize where they are in an environment. We compare newly submitted features to these existing anchor points, which makes it easy for devices to constantly update map areas that experience frequent change.

How Would Mapping Data Be Fused?

As the billions of smartphones upgrade to use depth sensing camera sensors, this data can be fused using a constant consensus model.

As multiple devices traverse a space, common natural features are identified computationally. Over a significant sample size, permanency can be determined from the quantity and quality of contributions (lamp posts, statues, trees).

Features that are recently identified multiple times but did not exist before can be represented as objects temporarily residing or passing through the space (parked cars, garbage bins, emergency cones). There are several layers to this, and it would be best explained using San Francisco’s Moscone Convention Center as a hypothetical:

• We can determine which features are permanent because they are present in every submission. If we were to map the Moscone Center, features like the convention center’s floor, building facades, and door are features that remain consistent over a long period of time, and because they are present in every scan, they are the most reliable for positioning devices.
• While the Moscone’s frequent hosting of events with varying décor may sound like it will throw off the positioning of our devices, these new, semi-permanent additions are actually helpful in positioning our devices. Over several days or rounds of consensus, we can identify booths, banners, and the stages as being present at the Moscone. We can use these features to identify where we are, but because they are new and may not be around long, these are not logged as permanent features.
• Finally, how do we fuse data when there are people present on the show floor, roaming robots, and half-empty drinks? We know these are temporary features, so these kinds of features are filtered out during each round of consensus and from each device.

Consensus requires a minimum number of user submissions. If not enough submissions are received, the submitted data rolls over to the following round of consensus. The ongoing nature of MAP’s updating model establishes and provides for a density-correlated update frequency, where commonly visited areas are mapped and updated more frequently.

MAP’s model allows us to establish the permanency of an environment (anchor points). We can then build off of those established anchor points. In higher population density areas, where more devices frequently move through the space, data is more likely to be updated but are contextualized by existing anchor points. The consensus model favors more recently submitted data for presentation, allowing for the most recent representation of the world to be accessible for use. The consensus model accomplishes this while keeping a history of the world.

How Would Knowledge Be Embedded?

To be useful, a map needs to provide more information than telling you where you are.

When MAP perfects a reliable digital representation of the physical world, MAP Token owners would embed location-specific information to exact physical locations. MAP also provides an easy way for establishments to verify their embedded content, which allows developers and users to filter information they trust and/or want to see.

When we compare MAP framework to existing positioning systems like GPS, the differences may seem negligible, but these differences become significant with use cases of the future.

With MAP, content could be oriented correctly and can appear within a meter of where it is supposed to be. It’s the difference between having your groceries list appear on your refrigerator door versus on the building outside your kitchen window.

Precise positioning is vital to improving the ease and the quality of experiences we have with the world around us.

Sample use cases for embedding content:

• Logistics and supply chain companies can plot autonomous vehicle and drone paths and waypoints onto the digital representation of the world. This provides autonomous machines with the means to communicate their whereabouts while, simultaneously, updating the map for other autonomous machines.
• Retailers can drive foot traffic with unique in-store experiences that include but are not limited to: embedding 3D content for products in their display windows, limited-time AR photo opportunities, and hosting personalized shopping experiences.
• Spaces that are, by nature of their purpose, large (shopping complexes, convention centers, universities, etc.), can provide a seamless experience for visitors with AR guided navigation and/or relay information about a landmark.
• Advertisers and marketers can run truly synchronous location-based campaigns (indoor and outdoor) at scale for thousands of users to engage, without the use of markers all around the world.
• Decentralized AR applications can implement use case specific triggers tied to smart contracts, instantly providing a compatibility layer for application content; an AR app can be built on the MAP Blockchain without a server.

Quality, Precision and Access

Applications that utilize geospatial data are able to operate at their greatest capability when given accurate and up-to-date 3D data about the world around us. As we evolve towards implementing autonomous vehicles and drones and augmented reality into our communities and routines, positionally accurate location data is critically important in providing spatial context for quality experiences. Unfortunately, the map we need does not yet exist. MAP is our solution for future systems that rely on geospatial data to function at their highest caliber of safety and quality.

MAP is a 3D map where any device using a location-based app can communicate their whereabouts, update the map as they traverse the world, and localize themselves accordingly.

In a sense, everyone’s devices could be embedded with a mental map that mimics the way humans have an intuitive and working comprehension of our surroundings. With MAP, our devices will also be flexible enough to comprehend their understanding of the world and act in accordance with the unpredictable environmental changes around them. Their new powers of 3D understanding can be fused into a digital singularity.

Providing a consensus based, always up-to-date map of the world requires nuance that crosses an array of fields and disciplines. To meet this challenge, we need a cross-discipline team. The MAP team is comprised of computer vision doctorates, blockchain experts, augmented reality pioneers, mapping professionals, and large scale mobile application developers.