Investors: Press Release

GeckoSystems' New Video Demos Microsoft Kinect Enabled Personal Robot

CONYERS, GA--(Marketwire - January 24, 2011) -  GeckoSystems Intl. Corp. (PINKSHEETS: GOSY) announced today that their recent video demo of their latest CareBot™ incorporates two of Microsoft's Kinect sensors. GeckoSystems is a dynamic leader in the emerging mobile robotics industry revolutionizing their development and usage with "Mobile Robot Solutions for Safety, Security and Service™."

This video shows a GeckoSystems' CareBot™, equipped with a pair of Microsoft Kinect sensors, navigating through a narrow passageway cluttered with various obstacles. This represents the worst case for in-home and/or commercial navigation. GeckoSystems recently announced GeckoImager™ incorporates multiple Kinects for a sufficient field of view (FOV) to be safe, collision proof, and capable of automatic self-navigation.

This new unedited video may be readily viewed at: http://www.youtube.com/watch?v=kn93BS44Das 

In it one will observe static and dynamic obstacle avoidance into and back out of a cluttered and narrow "gauntlet alley." One may determine that the movements are smoother than what most people could do using a joystick, human controlled mobile platform. One will witness three low levels of obstacle avoidance: reactive, proactive, and contemplative. No prior mapping or path planning was performed before or during the demonstration run. GeckoNav, the AI savant doing the guiding, controls all mobile activities during this demo. You will witness subsumptive AI behaviors occurring as obstacle avoidance subsumes way point seek and vice versa with an emergent behavior expressed as errand running down a cluttered hallway and back without any human intervention.

Microsoft recently announced that over 8 million Kinect sensors were sold within 60 days of its launch in November 2010, making 2010 the biggest year in XBox history. Microsoft sold around 1.9 million XBox 360 consoles in December 2010 alone, according to NPD, a market research company.

"Even though we are in very delicate and sensitive discussions with various Microsoft representatives, nonetheless due to the founder of Microsoft, Bill Gates' article, 'A Robot in Every Home' (http://www.scientificamerican.com/article.cfm?id=a-robot-in-every-home), gives us basis for belief that their interest is not cursory and simply polite. Microsoft has invested many man-years creating their mobile robot software development environment, entitled Microsoft® Robotics Developer Studio 2008 R3. Microsoft RDS is a Windows®-based environment for academic, hobbyist, and commercial developers to easily create robotics applications across a wide variety of hardware," stated Martin Spencer, President/CEO, GeckoSystems.

In a recent Nextgov.com article, "Gates: Robots are next hot health technology," (http://ht.ly/37XrJ) he is quoted as saying: "Advances in robotics will greatly enhance health care delivery over the next decade." The computer is "learning to see, it's learning to talk, it's learning to listen . . . and it's learning to move around," he said, predicting that ambulatory machines would be relatively cheap in about five years. "The dexterity things are maybe five years behind." The comments came at a conference to showcase advances in mHealth, or mobile health, organized by the National Institutes of Health, the independent nonprofit Foundation for NIH and the mHealth Alliance.

GeckoSystems' recently announced GeckoImager uses sensor fusion interpolates structured light machine vision data from the Kinect sensor with sonar range finding data, complimented by GeckoOrient's™ solid-state compass, accelerometer, and odometry sensor fusion. This provides their automatic, self-navigation AI software, GeckoNav™, with sufficient and timely data to achieve actionable situation awareness resulting in a very safe, loose crowd level of mobile robot autonomy that is "collision proof."

"The demonstrated CareBot automatic self-navigation is accomplished using three high level GeckoSystems' AI savants: GeckoImager, GeckoNav and GeckoSuper. The three savants are spread over two low cost, low power, off-the-shelf dual core Intel Atom motherboards running at 1.66 GHz located on the robot. The computer running GeckoImager is running Ubuntu Linux 10.10 while the computer running GeckoNav and GeckoSuper is running Microsoft Windows XP. Two low-level GeckoSavants, GeckoOrient™ and GeckoMotorController™ run on their robot controller board, the GeckoSPIO. All of the capabilities depicted in this video are running on board on this prototype mobile service robot. I am very pleased to be part of the engineering and programming team that achieved this demonstrable accomplishment," stated Kevin O'Connor, Sr. EE Roboticist, Research and Development, GeckoSystems.

Machine vision is a combination of structured lighting, a detector, and a computer to precisely gather and analyze data. Scanning the object with the light constructs 3-D information about the shape of the object. This is the basic principle behind depth perception for machines, or 3D machine vision. In this case, structured lighting is sometimes described as active triangulation.

Structured light is the projection of a light pattern (plane, grid, or more complex shape) at a known angle onto an object. Although other types of light can be used for structured lighting, laser light is the best choice when precision and reliability are important. This technique can be very useful for imaging and acquiring dimensional information. Fanning out a light beam into a sheet-of-light generates the most often used light pattern. When a sheet-of-light intersects with an object, a bright line of light can be seen on the surface of the object. By viewing this line of light from an angle, the observed distortions in the line can be translated into height and/or distance variations.

This fundamental technological approach has many applications in mobile service robots. Heretofore, cost, size and complexity of set up has been prohibitive. Microsoft's recent contribution to the advancement of low cost, truly utilitarian personal robots advances their founder's timeline notably and takes their new product, the Kinect, far beyond only the living room.

Service robots can be used to provide domestic aid for the elderly and disabled, serving various functions ranging from cleaning to entertainment to remote monitoring. The high cost of labor in developed countries and the increasing need for assisted living has led to the development of the service robotics market. As service robots are in greater proximity to humans, the technology involves more safety concerns over human-machine interaction. However, developments in the manufacture of intelligent and safer robots by GeckoSystems address the issues of safety, manipulation, and sensing. Thus, GeckoSystems is well prepared for the day when every home will have a robot.

A test drive in a house was not done as the first demo since most homes are not as cluttered as the "gauntlet alley" depicted. An actual home video, using the prior CompoundedSensorArray™ is at: http://www.geckosystems.com/

The design strategy for incorporating the Kinect units into the CareBot overall shroud design is to keep integration costs and manufacturing costs at a minimum. Just as the Ford Model A was better than that of the Model T, nonetheless several millions of Model T's were sold due to the compelling value proposition of selling for the same money as a motorcycle, but capable of carrying up to four people in a near weatherproof transport.

More detailed information explaining what the CareBot is seeing and thinking, while beneficial, would be misleading since machines do not "see and think." They read sensors and then compute responses. The "appearance" of seeing and thinking is an organic capability, not a machine's. So the mobile robot platform gathers distance data to objects in its environment, determines if they are obstacles, and then computes the necessary wheel speeds to successfully avoid that obstacle. 

GeckoSystems has several breakthrough technologies -- not just the GeckoImager -- in concert with the Kinect, and all the other necessary mobile robot solutions, such as their GeckoSavants, to have a complete, cost effective and multifunctional product. Due to the robustness of GeckoSystems' biological hierarchal architecture, GeckoNav being "sensor loving," and the recent invention of GeckoImager, the company has strengthened its "first mover" position in this emerging trillion-dollar industry.

Further, scanning laser range finders provide a similar level of data flood as does the Kinect. Robot experimenters have used scanning laser range finders for nearly ten years. At a cost of $2,500 to $50,000 each they are prohibitive. The PrimeSense Kinect solution is a watershed event due to its low cost and extremely robust functionality. It is consumer pricing for an industrial solution that has been investigated and improved for over 30 years.

Some believe that if the hobbyist could build a better personal robot, they would have videos up in many places on the Internet. Most hobbyists and many professional experimenters are hardware centric. Mobile robots are 80-90% software in their difficulty. GeckoSystems is AI software centric and only develops hardware on an as needed basis. The CompoundedSensorArray (CSA) was invented to reduce machine vision costs. The GeckoImager and Microsoft's Kinect enable a fifth generation CSA for GeckoSystems with significant cost reduction while improving overall performance.

To state simply, just because you have low cost and efficient fuel injection, does not mean that your racecar will win at Daytona. There are many, many more critical requirements besides "fuel management" to win NASCAR's Sprint Cup, eg.

"We are very pleased with the hard work and long hours invested by our team of engineers and programmers these past few months at achieving this higher level of value to our marketplace. This breakthrough usage of Microsoft's low cost Kinect range finding sensor system may enable some near term business-to-business niche retrofit markets for us. We continue to expect technology-licensing revenues to precede revenues from product manufacturing and sales and have two to three prospective licensees in active discussions. This augurs well for increased ROI and shareholder value for our nearly 1400 investors," concluded Spencer.

Download Full-Text PDF