This is commentary about a Kickstarter campaign for a radar retroreflector integrated into a bicycle taillight assembly.
An image from the Web site:
Image from Ilumaware Web site
One nice thing I can say about the product is that it is quite inexpensive, so I’ll say that first. The reason is that this is not a high-tech product. This is a low-tech component of a system whose high-tech component is in cars.
Retroreflectors work by concentrating light (or in the case of a radar reflector, radar signals) back toward the source. The product is a single cube-corner retroreflector. Optical retroreflectors are the insect’s eye version, with multiple smaller reflective elements, so they work at the much shorter wavelengths of visible light. The technology is described on another Web page.
As to the effectiveness of this product, I have no doubt that it improves the visibility of a bicycle to radar — but…
The product’s Web site repeatedly uses the term “OTR technology”, without ever spelling out the meaning of the acronym. I couldn’t find a definition anywhere online, either. This term makes the product appear more high-tech than it is. Indeed, the site claims:
Stealth techniques use radar reflection to make an object less visible and/or “invisible” to radar. We have reverse engineered this technique into a product used by a cyclist to make you more visible to a car. This is a revolutionary application of radar technology.
Reverse engineering is correctly defined as analysis of an undocumented product to develop specifications for a duplicate or similar product. Examples are the Wright brothers’ reverse engineering the flight characteristics of birds to design aircraft, and Linus Torvald’s reverse engineering the proprietary Unix computer operating system to construct the Linux operating system. The Kickstarter campaign uses the words “reverse engineered” inaccurately, so as to mislead people who do not understand it, as if to mean design of a product to have the opposite effect of an existing product. And when that product is a stealth bomber — wow, now the new product must be extremely high-tech! Again, the product is a simple cube-corner radar retroreflector, as has been used in boating for decades. The designers describe design and optimization of their product, but this is plain vanilla engineering, not reverse engineering.
A radar retroreflector which works in all directions is more desirable, (though it still will not always work, even if a car has radar, because the radar beam may not be aimed in its direction, and there may be a line-of-sight obstruction).
Radar alone as a robotic aid to a human driver is possible, but not very practical. Only a small percentage of cars have radar as of yet. A human driver uses visual cues. A fully-robotic car also must, because not every potential obstacle will be as large or reflect radar signals as well as a bicycle — think potholes, cats and dogs, etc.
The product, as shown on the Web site, includes an active taillight, but no optical retroreflector — though installed in the same location on the seatpost which is usual for one — following in the long tradition of new products promoted as a panacea for cyclists’ conspicuity problems while ignoring basic legal and functional requirements. Most states require a retroreflector or taillight, but any taillight can go out without the bicyclist’s being aware of that, and so any bicyclist who rides after dark should have a rear-facing retroreflector, not only a taillight.
The online promotion entirely fails to mention the need for a headlight, or the legal requirement for one. The Web site shows a bicycle with no headlight.
A bicyclist must always use a headlight at night, because an optical forward-facing reflector does not alert pedestrians or drivers who do not have headlights aimed at the bicycle (cars backing out of driveways, at stop signs in side streets, other bicyclists without headlights, etc.) Still, unlike the optical retroreflectors on bicycles, a forward-facing radar retroreflector is likely to be effective, because a car’s radar is likely to scan in more directions and its pulsed output is immune to interference from other sources. But the retroreflector here is only rearward-facing.
The online promotion also makes a number of inaccurate statements.
Riding with a tail light [sic] is important regardless of the time of day.
While a very bright taillight can help to alert drivers — human or robotic — during daytime, reducing the probability of a collision somewhat, there is no law requiring a taillight (or rear-facing optical retroreflector) when riding during daytime.
* “In 2015, more than 35,000+ collisions occurred between cars and cyclists in the U.S. Approximately every 3 minutes, world-wide, 6 people die and nearly 285 people are injured in collisions involving cars and bicycles. The majority of these accidents are from behind because drivers didn’t see the rider and it is NOT because they did not have a tail light.”
This is wildly inaccurate. While rural car-overtaking-bike collisions are disproportionately serious and fatal, only approximately 7% of car-bicycle collisions in the USA are car-overtaking-bike collisions. A very large percentage of these occurs to cyclists riding at night without a taillight! In urban areas, most of the serious and fatal collisions involve turning and crossing movements. No rear-facing conspicuity equipment — optical or radar retroreflector, or taillight, will prevent most of these. Sure, many if not most car-overtaking bike crashes could be avoided, day and night, by use of a radar reflector, if cars have radar connected to a robotic crash avoidance system — but again, as of yet, only a very small percentage of cars is so equipped. Which takes me to my next quote:
* “In 2016 … there are 470 out of 566 unique car models sold in the U.S. equipped with radar (83%).”
This is very seriously overstated. Saying that a model is equipped with radar is not the same as saying that radar is standard. Adaptive cruise control is still often an expensive option. Only some adaptive cruise control systems include automatic crash avoidance. Some systems use laser ranging rather than radar. The fleet of motor vehicles turns over slowly. More even-handed estimates are found in this article in the Detroit News. Quote from that article:
IHS Automotive forecasts 7.2 percent of vehicles produced globally by 2020 will feature adaptive cruise control, up from 2.2 percent in 2014.
More details and a list of vehicles are on Wikipedia.
Why do promotions like this occur? Fundamentally, because regulation of bicycle equipment in the USA at the Federal level, where equipment standards are set, is a Wild West situation, harkening to the interests of the bicycle industry. That is another story, too big to cover here.