An autonomous lawn mower from iRobot may interfere with radio astronomy's use of methanol's radio signature as galactic markers.
The first iRobot Roomba floor cleaner unit was a tremendous breakthrough when introduced in 2002. It made the vague and confusing concept of a useful robot real and tangible. It did so by providing a sophisticated, autonomous robot that offered a meaningful, demonstrable function and made it into a consumer-market success.
Since then, iRobot has introduced many consumer, commercial, and military robots for a variety of applications including cleaning rain gutters, searching through collapsed buildings, and investigating explosive devices. The Roomba also showed how far the various underlying technologies embedded within it—motion control, sensors, algorithms, power subsystems, and more—had advanced.
Their latest specialized iRobot model is one which mows lawns, which seems to make a lot of sense: it's a task which is definable, done over a limited area, and which most people would rather not do (or elect to pay someone to do for them). Such lawn mowers are already offered by other vendors; the article "Astronomers See Robotic Mowers Invading Their Space in FCC Fight" has some interesting numbers on their world-wide sales. For any autonomous mower, the problem must literally be "bounded", with the area to be mowed delineated in some way.
That's where there's a potential problem. iRobot has asked permissions from the FCC (Federal Communications Commission) in the US to set RF beacons operating between 6.240 and 6.740 GHz as markers on the periphery of the lawn or field. Competing mowers used buried wires to form a boundary loop (similar to the invisible fence used to keep dogs in the yard). A typical area to be mowed would need between five and ten of these beacons, each mounted on a pole about 2 feet (2/3 meter) high.
The problem is that astronomers such as those at the Green Bank, Virginia, National Radio Astronomy Observatory (NRAO) as well as other locations, use 6.650 to 6.6752 GHz as the spectral band for "seeing" methanol in space (Figure). Methanol (CH3OH), with a 6.66852 GHz spectral signature, is associated with star-forming activity, which is used as a beacon to estimate distances and track the path of what is called "galactic evolution." Obviously, that methanol signature is determined by the laws of physics and can't be shifted. The received signal flux from various radio sources is almost unimaginably tiny, on the order of 10−26 watts per square meter per hertz (called one Jansky in radio astronomy metrology) with extremely difficult SNRs (signal-to-noise ratios), which are below -100 to -150 dB.
The Green Bank Telescope is 485 feet (148 meters) high. It weighs 17 million pounds (7700 metric tonnes) with a dish area of 2.3 acres (1.7 hectares). It is the largest moving structure on land, and the largest fully-steerable telescope in the world (photo provided by NRAO).
The NRAO astronomers would like the beacons kept at least 55 miles (90 kilometers) from a radio telescope. The area around Green Bank is already restricted to be a very quiet RF zone by regulation: no cell phones, no transmitters, nothing that might cause RFI in the truest sense of the phrase. The specifics and recommendations for RF levels near radio telescopes are even called out in ITU documents, see References.
The request from iRobot is for an exemption to the FCC declaration that the methanol band is protected spectrum. The company says the fears are exaggerated, and once hills and trees are factored in, no robot lawn mower will operate within interference range. The company adds that the potential for interference is limited due to the low power, intermittent use, and low radiation angle of the beacons. They have even offered to put notices about appropriate beacon locations in the user's manual. (You can read the various back-and-forth statements: the waiver application from iRobot, the comments from NRAO, the response from iRobot, and the reply from NRAO.)
Of course, RF beacons on stakes are less of a headache and much more flexible for users than the buried wire loop. I can only speculate why iRobot decided against a GPS-based marker system: was it due to one or more issues of cost, system complexity, and ease of installation and configuration? Even so, why go for 6.240 to 6.740 GHz, instead of some other band slivers: was it related to propagation, wavelength, convenient antenna sizing and configuration, or other?
What's your view on this? Are the NRAO fears legitimate, or overstated? Have you ever had a super-sensitive RF measurement corrupted by an unexpected source which was difficult to locate?
ITU Recommendation ITU-R M.1583, "Interference calculations between non-geostationary mobile-satellite service or radionavigation-satellite service systems and radio astronomy telescope sites.
ITU Recommendation RA.769-2: "Protection criteria for radioastronomical measurements” (ITU 2003),
""Single Dish Radio Telescopes," Chapter 3, Jayaram, N. Chengalur.