We've reported on Roomba-like robots that mow your lawn. Now, one with a wide-angle camera for conducting under-vehicle security checks helped to patrol the Republican National Convention.
General Robotics' Ferret is a four-inch-high, 14-inch-diameter, disk-shaped robot camera used for inspecting under vehicles. Constructed of anodized aluminum with a high-impact polycarbonate cover, the radio-controlled robot transmits real-time color, wide-angle video images via UHF wireless to an operator up to 100 meters away.
The radio-controlled, Roomba-shaped Ferret under-vehicle inspection robot can transmit real-time video images to an operator up to 300 meters away. (Source: General Robotics)
Operators view the pictures with high-resolution 640 x 480 video glasses, or optional video monitors. The video signal can also be recorded or streamed to a remote location. An extended range option provides a long-range antenna for operation up to 300 meters between the operator and the robot.
The Ferret's movable camera can be tilted to many different angles. Camera angle and the amount of light from its onboard LEDs can both be controlled remotely. Previous methods of searching for bombs and explosives, such as driving a vehicle over a fixed digital scanning ramp, provide only flat images. In contrast, the Ferret is especially useful for seeing up and around drive shafts and into deep wheel wells. (Watch a video demonstrating the robot below.)
At the Republican National Convention, the robot was reportedly used by the Tampa Bay Regional EOD (Explosive Ordnance Disposal) response team and the Secret Service. It has also been field tested by the New York Police Department's bomb squad for first-look applications.
The robot's manufacturer, General Robotics, is a subsidiary of Panoscan, inventor of the MK-3 high-resolution, rotating, 360-degree digital panoramic camera used by the New York Police Department for capturing crime scene data.
The robot, the video belt pack receiver, and the video glasses are all powered by rechargeable batteries. Typical battery run times per charge are about three hours for the Ferret, six hours for the video receiver, and about five hours for the video glasses. The Ferret weighs 8.8 pounds with battery installed and has a one-inch ground clearance. The robot operates on most paved surfaces. Its maximum speed is 3 mph. Its three wheels include one swivel caster and two wheels driven by a direct precision planetary drive.
Rob, these are not consumer prices, since the cameras are much more capable than consumer products and they are designed for organizations needing computer vision, not snapshots. This--cost and capabilities--is especially true of the image sensors they contain. Also, consumers have gotten used to extremely cheap camera prices due to the price declines in CMOS image sensors. I'm pretty sure the Ferret costs less than the Panoscan since it is less capable. OTOH, it's digital, which means it uses CMOS image sensors, and the Ferret is CCD-based. Comparable CCD sensors usually cost more than CMOS sensors, but the two cameras must have very different sensor specs if one can do 360 and the other can (apparently) do 90 degrees.
That explains a lot, Ann. Are these systems also programmed to recognize unusual objects as they make their scans? Do you know if devices like the Ferret feed their scans back to a centralized center where they can be viewed by experts with more knowledge than those in the field?
The article suggested that the camera was removeable. Is that only for replacment or are there other options available, maybe like a chemical sniffer package with less advanced camera?
@Rob, that would be a good idea to pair the unit up with some high end matching systems. I could see in the future having a database of common vehicle undersides and then alert if somesomething doesn't match.
Good point, Jack, on both the chemical sniffer package as well as a database of car underbellies. Seems that would be within reach technically. That could offer a quick detection of something different in the picture.
As we said in the article, the video is viewed by the operator using special glasses or a monitor. Alternately, it can be recorded or streamed to a remote location. Object recognition and analysis would most likely be a facet of vision analysis software, but I don't recall any being mentioned.
Jack, the article says the camera is movable, not removable. I think combining it with various object recognition/analysis software packages, as Rob suggests, might be an interesting idea, or chemical sniffers, as you mention. But the existing robot is designed primarily for on-the-spot human eyeball/brain inspection.
With major product releases coming from big names like Sony, Microsoft, and Samsung, and big investments by companies like Facebook, 2015 could be the year that virtual reality (VR) and augmented reality (AR) finally pop. Here's take a look back at some of the technologies that got us here (for better and worse).
Good engineering designs are those that work in the real world; bad designs are those that don’t. If we agree to set our egos aside and let the real world be our guide, we can resolve nearly any disagreement.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.