The Defense Advanced Research Projects Agency (DARPA) helped lay the foundation for autonomous ground vehicles with a 300-mile challenge race. DARPA also pushed AI-powered, software-defined radios to their limits in the three-year Spectrum Collaboration Challenge that culminated at last year’s Mobile World Congress Los Angeles, and now the underlying technology is being further explored in 5G test beds.
The National Institute of Standards and Technology’s Public Safety Communications Research Division is hoping to spark similar innovation for public safety technology with a series of challenges funded out of $300 million in research funding that it has to spend by 2022. Public challenges, according to Division Chief Dereck Orr, put PSCR in touch with experts, organizations and innovators that it might not otherwise come across.
PSCR recently concluded its “Tech to Protect” challenge, which included public safety hackathons in 10 cities across the U.S. and awarded $800,000 worth of prizes in May. The challenge was unique in that it wasn’t a single-day hackathon series, but was designed to bring application development along over time. The winners from May received $30,000 each in “seed” money, and in November, they can come back to PSCR for up to $70,000 each in additional funds, depending on how much progress they’ve made on a “growth strategy” get their solutions closer to market — and in the hands of first responders. This particular challenge was also aimed at supporting public safety solutions that could be brought to reality quickly; PSCR usually focuses on research that will ultimately support tech developments are further out, on the scale of 5-10 years away.
The Tech to Protect challenge is just one of PSCR’s current challenges and planned 2020-2021 challenges. There are six others, which include:
-The ongoing 2020 CHARIoT Challenge, which centers around integrating information from IoT devices, smart buildings and smart cities into augmented reality interfaces to help inform first responders and improve situational awareness; as well as emulating smart city data sources for disaster response scenarios. The four-phase challenge will award up to $1.1 million in prizes; the first round is already complete.
-The 2020 UAS Endurance Challenge, focused on advancing drone design for public safety search and rescue use. In those circumstances, long operating time (90 minutes+) and the ability to carry a heavy load are useful features, but can pose design challenges. This is a four-stage competition, and first-stage winners have already been announced. Walk-on entries are still being accepted through mid-December.
-2020 Automated Streams Analysis for Public Safety (ASAPS) Prize Challenge. This challenge will involve competitors developing algorithms to handle live streams of unstructured public safety data and pick out emergency events, derived from an “extensive and unique staged dataset designed to stimulate a day in the life of a busy city,” according to PSCR.
-The upcoming Enhancing Computer Vision for Public Safety Challenge, which will center on creating image datasets that include common camera image problems, like glare or a dirty lens. A roadblock to the deployment of computer vision and video analytics is the myriad problems cameras experience when deployed in real world environments,” PSCR explains. Computer vision systems need images on which to be trained so that they can be trained to spot and account for such problems, so that public safety computer vision use is more accurate. The two-phase challenge will have prizes of up to $240,000.
-Planned for the winter/spring of 2021 is a Mobile Fingerprint Capture for First Responders Challenge, in which PSCR said it wants competitors to explore “ways to utilize already-deployed mobile technologies to provide a portable, on-the-scene fingerprint capture solution that would save public safety personnel resources and potentially save lives.”
-PSCR is also planning a follow-up to a 2018 challenge related to differential privacy algorithms. In a public safety context, being able to identify people’s locations and track them over time is valuable for emergency and disaster response, evacuation, personnel placement and more — but tracking people’s locations over time also represents a privacy risk. The upcoming 2020 NIST PSCR Differential Privacy Temporal Map Data Challenge seeks to “develop algorithms that preserve data utility while guaranteeing individual privacy is protected.” That challenge is expected to begin this fall.
PSCR noted that more information about the related issues to privacy and data-sharing will be among those highlighted at its virtual annual stakeholder event, being held this week.
Information about more U.S. federal challenges is available here.
