Can Google provide us with a new approach to how we commute to and from work and travel to other destinations?
It’s something that they appear to be working upon.
In one of my previous lines of employment, I had a twenty-minute commute that often took closer to an hour, based upon traffic.
One approach was to take public transportation, which was very convenient because catching the bus was nearby both home and work, but this took a fairly long time. Another was to leave early in the morning to avoid the crush of traffic and to take less-traveled alternative routes in the evening. Of course, leaving early meant arriving early, but it was better than sitting in traffic. Unfortunately, taking alternative routes sometimes resulted in surprising delays and travel times that weren’t significant improvements.
Imagine being able to quickly and easily gauge how much traffic was on different routes at different times and to be provided with viable alternative routes. That’s the focus of a new patent application from Google.
A computer-implemented method of providing personalized route information involves gathering a plurality of past location indicators over time for a wireless client device, determining a future driving objective using the plurality of previously-gathered location indicators, obtaining real-time traffic data for an area proximate to the determined driving objective, and generating a suggested route for the driving objective using the near real-time traffic data.
Here are some of the highlights:
- The system is envisioned to work with cars, trucks, and mass transit.
- It can be used with computers, wireless handheld devices, and navigation systems.
- Real-time traffic conditions are part of the information that is transmitted by the service.
- GPS and wireless devices may be useful in helping the system “learn” about travel routes when their users place those devices in a “learning” mode. A person may also manually enter information about starting and stopping routes that they often take while commuting.
- Time of a journey may be considered in the learning aspect of this system, as well as necessary stops along the way, such as pulling over for a cup of coffee or picking up dry cleaning.
- Travel profiles are generated during the learning process, and navigational points are recognized by the system, enabling it to offer alternatives when necessary and still get a driver or commuter to the places that they want to go.
- Real-time traffic data is included as part of this system, and it may include “data relating to non-automotive modes of transportation, such as rail, bus, and ferry speeds and schedules.”
- It may be possible to include local search information in this system to help a driver find places to stop for dining or other purposes.
- The system could also be used in conjunction with a scheduling program, such as Outlook, to help plan travel routes based upon destinations noted in a person’s schedule.
- Real time speed of traffic may be identified through cameras or inroad sensors.
Google’s 2004 Acquisition of Zipdash
In 2004, Google acquired Zipdash, which was working on providing aspects of a service like this, and it looks like they have not ignored the acquisition. Zipdash (Internet Archive link) covered the following areas with their traffic coverage:
- SF Bay Area
- Los Angeles
- San Diego
The patent application mentions the company by name:
The information may reflect the speed of traffic flow at particular points in a transportation system and could include a service such as Zipdash. NIS 51 may take the information received from traffic service 56 and combine it with route information computed for a user to compute a suggested travel route, as explained in more detail below. In this context, real-time traffic conditions are intended to include sufficiently recent or accurate conditions to have relevance and be of assistance.
The patent document provides a much more detailed look at how this system might function, and I’ve only brushed upon parts of what it covers. However, it certainly expands upon the type of information that you might expect from a search engine. The incorporation of real-time information seems to be something that we are starting to see more of in some of the patent filings from Google, such as the one I wrote about last week on shopping.
It appears that Google’s mission of organizing the world’s information includes information about things that are happening at present.
In his days at Carnegie Mellon University, Shumeet Baluja wrote or co-authored many papers on robots, transportation, and automated highway systems. Some interesting reading there. Here are links to a number of those papers:
- Multiple Adaptive Agents for Tactical Driving (pdf)
- Evolving an Intelligent Vehicle for Tactical Reasoning in Traffic (pdf)
- Evolution of an Artificial Neural Network Based Autonomous Land Vehicle Controller (pdf)
- Prototyping Intelligent Vehicle Modules Using Evolutionary Algorithms (pdf)
- Artificial Neural Network Evolution: Learning to Steer a Land Vehicle (pdf)
- A Massively Parallel Road Follower (pdf)