System Bits: Dec. 11

Calculating the costs of autonomous vehicles
The development of autonomous vehicle technology commands a lot of media coverage. Little reporting has been devoted to the costs of operating AVs, a subject that developers don’t discuss in general.

The Houston-Galveston Area Council’s website recently divulged contract figures with two startups, Drive.ai and EasyMile. For Silicon Valley-based Drive.ai, the company charges $14,000 a month, which works out to $168,000 a year, to provide one van vehicle, with the company assuming operation and maintenance costs. If a public agency makes a five-year commitment to using the AVs, that figure is lowered to $12,900 per month, or $154,800 a year. EasyMile, a French provider of autonomous shuttles, charges more than $27,000 a month for each small shuttle for cities that sign up for one year of service, for a yearly total of $324,000 or more. For a five-year commitment, the monthly charge falls to about $8,000 a month per shuttle, or around $96,000 per year.

A Drive.ai self-driving van in Frisco, Texas. Credit: Kaveh Waddell, Axios.

Local governments are using federal grants to afford such charges in their trial programs with automated driving. Rhode Island is paying $800,000 for the first year of a shuttle service in Providence during 2019, with $300,000 of that total coming from a federal grant. The city of Arlington, Texas, a Dallas suburb, selected Drive.ai to provide three on-demand, self-driving shuttles in its entertainment district. Drive.ai is charging $435,000 for a yearlong program; the city will pay one-fifth and the federal government is picking up the tab for the remainder.

Greg Rodriguez, an attorney in Washington, D.C. specializing in AV law, told Axios, “Most cities think that there will be no costs related to a pilot project with [a driverless] shuttle company.”

Raj Rajkumar, a Carnegie Mellon University professor, said, “I will be very surprised if cities will pay for AV car services even at relatively low prices, let alone steep prices.”

Experts are concerned that adopting AVs will endanger funding for existing public transportation, such as buses and subways, which are used by low-income residents.

“There’s certainly going to be consequences for mass transit in the same way as even the development of rideshare has,” says Darrell West, director of the governance studies program at the Brookings Institution. Government investment could serve as the catalyst for “broad economic benefits,” which would be followed by private investments, he adds.

ETRI succeeds with exchange of quantum information
The Electronics and Telecommunications Research Institute of South Korea reports successfully exchanging quantum information in daylight using a free-space quantum key distribution (QKD) paired with ETRI’s polarization encoding microchip. QKD is said to be a promising secure communications technology, encoding information in a single photon.

ETRI researchers look at the QKD system. Credit: Electronics and Telecommunications Research Institute.

ETRI says its free-space QKD works in daylight, while similar systems have failed to operate in those conditions due to the substantial amount of noise photons generated by sunlight. By developing and adopting elaborate noise-filtering technologies, ETRI’s QKD system achieved the secure key rate of 142.94 kilobits per second with a quantum bit error rate of 4.26% in daylight over the free-space distance of 275 meters.

Secondly, the ETRI QKD system is significantly reduced in size compared with conventional QKD systems thanks to the polarization encoding chip. This makes the QKD system suitable for such applications as autonomous vehicles and unmanned aerial vehicles.

The institute is now working on applying its integrated chip technologies to other optical components to create miniature QKD transceiver modules. ETRI is also looking to extend the transmission distance in daylight.

Oxford spinout automates 3D digitization of objects
The UK’s University of Oxford reports that an Oxford spinout company, PalaeoPi, has developed the MKIII TablePi photogrammetry platform, which can digitize small to medium-sized objects in a short amount of time, taking more than 200 high-resolution photographs in five minutes with three digital SLR cameras. The spinout is aiming the MKIII TablePi at the archaeology field, followed by applications for CGI designers, developers of video games and virtual reality, and related animation.

PalaeoPi’s MKIII TablePi photogrammetry platform in action. Credit: PalaeoPi.

PalaeoPi’s technology was developed by Richard Benjamin Allen, research support officer of the Oxford University School of Archaeology and lab manager of the Henry Wellcome PalaeoBARN, and by Dr. Ardern Hulme-Beaman, a Leverhulme Trust Early Career Fellow in the Department of Archaeology, Classics, and Egyptology at the University of Liverpool. PalaeoPi is the first lean spinout developed by Oxford University Innovation, an arm of the university.

Allen said, “I set up PalaeoPi to offer a different paradigm. PalaeoPi is helping institutions and companies to develop and maintain their own expertise in photogrammetry. We are doing this through customer-led R&D, publication of research with university partners, and long-term support of an automated platform that is easy to customize, easy to repair, easy to retrofit, and modular with backwards compatibility in mind. We are equipping people with the tools and knowledge they need to make good choices in their research and beyond.”