Microsoft Patents Bad Neighborhood Detection

With the grant of their US Patent #8090532 Microsoft may be attempting to corner the market on GPS systems for use by pedestrians, or they may have opened a fertile ground for discrimination lawsuits. We won't really know until actual devices incorporating the designs in this patent begin to appear.

Described as a patent on pedestrian route production, the patent describes a two-way system of building navigation devices targeted at people who are not in vehicles, but still require the use of such a device to most efficiently route to their destination. Microsoft feels that this patent addresses an unmet need for assisted navigation on foot.

Loosely the patent breaks down into three categories; gathering data, analyzing data and user requirements, and route generation. For example, the user inputs their destination and any constraints or requirements they might have, such as a wheelchair accessible route, types of terrain they are willing to cross, the option of public transportation, and a way point such as the nearest Starbucks on the route.

Any previously configured preferences are also considered, such as avoiding neighborhoods that exceed a certain threshold of violent crime statistics (hence the description of this as the "avoid bad neighborhoods" patent), fastest route, most scenic, etc. The system then analyzes potential routes, considers constraints, compares data sources, and delivers the appropriate instructions in the form selected by the user, such as a scrolling map, turn-by-turn direction, or spoken guidance.

So far, this sounds pretty much like any GPS device you currently use, albeit with a few pedestrian specific bells and whistles. However the patent also covers a few components not currently found in navigation systems. First, a device built to these specifications is continuously gathering data about routes commonly traveled by the user; it’s a two-way, real-time process that requires connectivity of some sort, as opposed to the self-contained, annual update nature of most GPS systems. Through that connection it can gather security information (which includes crime statistics and presumably real-time news reports), traffic data, weather conditions, and additional user goals, such as a stop at a highly-rated lunch cart.

These evaluations can allow the user device to recommend "take the subway" rather than walk if, for example, weather conditions were such that the subway was a better alternative. It also describes mechanisms where the choices might be between driving and taking public transit to sporting events, for example, including evaluative factors such as comparing the cost of parking and potential traffic issues to the cost of public transit and time required to walk and take the bus to complete the trip.

There is also an artificial intelligence component, which would analyze data, evaluate the reliability if the source and apply it to the requested directions. This can also include contextual data derived from previous and current usage; the device should be able to evaluate the behavior of the user; do they saunter or walk with a purpose, do they frequently stop to enjoy the sights, or are they a head down power walker. This contextual evaluation can be used to provide the route that would be most enjoyed by the device user.

And since any successful technology needs a clear path to monetization, the patent includes mechanisms for inserting both obvious advertisements and weighted recommendations into the routes being generated, including mechanisms that would allow for payments based on the success of getting device users to actually stop at locations that have inserted advertising or recommendations into the navigation system.

Clearly the potentials described in this patent cover many useful aspects that would be of value to urban dwellers or those that often travel and go sightseeing at their destinations. Given the benefits of a continuous net connection to the described technology, it’s incorporation into smartphones seems an obvious use and a possible significant influence on a purchase decision.

The patent is surprisingly easy to read and contains much more than we have outlined here. If you find this technology interesting, spend a few minutes reading the patent and let us know what you think.

With more than 20 years of published writings about technology, as well as industry stints as everything from a database developer to CTO, David Chernicoff has earned the term "veteran" in the technology world.