2014
03.27

Awards to well-known corporate names—and a few new ones—for improved speech recognition, autonomous vehicle control, landing a space vehicle in the ocean, and improving Website analysis through behavioral portraits featured prominently these past two weeks.

Google received 103 awards, among which were three for controlling autonomously driven vehicles and three for improvements in speech recognition. The former will help the company’s self-driving car initiative and the latter for potential improvements to Google Glass.

The three patents associated with autonomously driven vehicles are 8,676,430 (“Controlling a Vehicle Having Inadequate Map Data,”) 8,676,427 (“Controlling Autonomous Vehicle Using Audio Data,) and 8,676,431 (“User Interface for Displaying Object-Based Indications in an Autonomous Driving System.”) The one I found most interesting was the one that addressed inadequate map data. We are all dependent upon GPS directions as we drive. They are from our smartphones or dedicated devices such as those made by Garmin. These devices are in turn dependent upon updates that account for new roads, streets, directional changes, and the myriad of other changes that occur daily in the U.S. Those updates depend upon the map-generating companies receiving the information from local sources as changes occur, which is problematical.

Human drivers compensate for unexpected routing disruptions using visual cues and logic. We do this quickly and even if we err, we can determine how best to overcome the issue independently of the GPS.

So imagine if you are a passenger in an autonomously driven car, and there is an inaccuracy in the map data. What will happen? This is where the process described in the patent comes into play. The patent stipulates that the autonomous car has sensors that help it detect obstacles, road conditions and other driving inputs. When a map-based error occurs, the control system employs the data from the sensors to determine the corrective action. The hierarchy is map-based data then sensor data to control the course of the car.

Google’s three patents for improvements in speech recognition are 8,682,659 (“Geotagged Environmental Audio for Enhanced Speech Recognition Accuracy,”) 8,682,661 (“Robust Speech Recognition”), and 8,682,663 (“Performing Speech Recognition Over a Network…”). All three have potential application to Google Glass, which incorporates speech commands to control applications. Speech recognition improvement is very important to commercial and industrial applications that intend to be deployed on Google Glass. Commercial and industrial workplaces tend to be noisy environments. As an example, voice-directed work applications in warehouses and distribution centers, which may include freezers, conveying systems and other high-decibel generating machinery, depend upon ruggedized devices and advance noise-canceling headsets to overcome the impediments to consistent speech recognition. Anywhere there is enhanced background noise, efforts to improve speech recognition are essential if wearable devices expect to gain a foothold, and displace the specialized devices that are presently used.

Related to this is the award Google received to improve augmented reality, for which Google Glass is designed. Patent 8,681,178 (“Showing Uncertainty in an Augmented Reality Application”) provides for a means to alert the user that a specific part of the his view, say an area around which the user sees a red circle, has a degree of uncertainty as to what the system thinks is there, such as merchants in a building.

In my most recent blog, MODEX 2014, I commented on Amazon’s efforts to recruit knowledge workers for its distribution center operations. Amazon received three patents for automating warehouse operations. They are 8,682,751 (“Product Dimension Learning Estimator,”) 8,682,473 (“Sort Bin Assignment,”) and 8,682,474 (“System and Method for Managing Reassignment of Units Among Shipments in a Materials Handling Facility.”) While it seeks to reduce the low-skill labor component in its distribution centers through continuous introductions of robotics and sensor-enhanced machinery, it is aggressively recruiting the high-skill workforce that will be required to keep the automated warehouses up and running.

Here’s a company I’m sure you have never heard before: 7 Billion People, Inc. (Austin, Texas). It was awarded Patent 8,682,741 (“Behavioral Portraits in Web Site Analysis.”) The process describes a method “for determining a website user behavioral portrait based on navigation on the Website and dynamically reconfiguring Web pages based on those portraits. In accordance with the method, data relating to the progress of a user through a Website is recorded, and an ongoing behavioral portrait of the user is built based on the data. The portrait is then used to dynamically reconfigure Web content.” This is intended to benefit eCommerce merchants that want to refine offerings that more closely match the user’s interests, based on the user’s behavior on the site. Here’s the thing: The company was started in 2006, shows investor funding through 2012, and then news about the company dries up. And, its Website is not functioning. Is this an example of an interesting patent with nowhere to go?

Finally, Blue Origin, LLC (take a guess as to what blue origin refers), was awarded Patent 8,678,321 (“Sea Landing of Space Launch Vehicles.”) The patent covers a more cost effective way to land a spacecraft in the water in a manner that will allow for its reuse. Civilian space initiatives are shaping up and have a future, and the creative thinking going into ways to bring costs to launch, recover and reuse spacecraft are to be applauded. Here’s what continues to be a puzzle to me: Why do Americans land spacecraft in the ocean, while the Russians bring them back to a terrestrial landing?