Hi all –
Hope everyone had a fine time eating turkey. Today we’ll look at activity at the FAA including the ASSURE impact report, tough new regulations in the UK, the unending DJI, hobby registration and an AI smackdown that the carbon unit wins.
But first… I am delighted to have been invited to join the 2018 Energy Drone Coalition Advisory Board. The Board will be working with Sean Guerre and LaDonna Pettit of the Stone Fort Group to shape the agenda for the 2018 Energy Drone Coalition Summit, June 20-22 in Houston. If you are in energy or infrastructure, this is your show. You can read my review of the 2017 event here.
Tuesday the 28th marked the close of the initial application for “lead applicants” for the recently announced UAS Integration Pilot Program (DIPP). In case you haven’t memorized the details, lead applicants are the state, local and tribal governments who “will serve as the primary point of contact with the FAA from start to finish.”
From the FAA website:
- The deadline of 2:00 pm ET, November 28, 2017 to submit a Notice of Intent to become a Lead Applicant has passed.
- By 2:00 pm ET, December 13, 2017: Lead Applicants complete volumes I and II in the application portal
- By 2:00 pm ET, December 13, 2017: All interested entities can request to be included on the Interested Parties List
Miriam McNabb and the team at dronelife.com have been all over it. I particularly like this headline, Tomorrow’s the Deadline: Are State and Local Governments Ready for Drone Integration Pilot?
The answer is, nobody knows, since to no one’s surprise, the FAA has yet to issue a press release announcing how many applications were received. From everything I am hearing, this is an extremely short amount of time for a bureaucracy, especially one without aviation experience, to organize a response. As I have said before, the responses are basically like applying for a waiver…
In a separate but not equal report, NextGov.com writes that “As of this week, about 1,823 separate entities, including individual drone operators, have listed their names on an FAA spreadsheet as “interested parties,” according to a FedBizOpps posting. Which is a little surprising since the posting reminds us that “The Government will not provide any funding for project performance.”
I had a fascinating off-the-record conversation in which I was assured that DIPP was specifically developed as an alternative to the proposed Feinstein Drone Federalism Act that was introduced in May. For proof, I was offered up another McNabb article from a few weeks back, Rep. Jason Lewis: UAS Integration Pilot Program “Doesn’t Go Far Enough” to Protect States’ Rights.
That neither of these had a snowball’s chance of leaving committee seems to be beside the point – in fact the great virtue of DIPP if there is one, is that it end runs Congress which is preoccupied with more taxing matters. Seldom have the wheels of progress turned so quickly.
As to the potential effectiveness of DIPP, one of the more stunning headlines of the week came in Unmanned-Aerial.com where Betsy Lillian reported that Michigan Proposes Its Own Statewide UAS Regulations. BTW “Develop a submission for the Federal Aviation Administration’s UAS Integration Pilot Program” was one of ten recommendations.
However, Congress is not that easily distracted, after all they issued very specific instructions in 2012. On Wednesday, the HOUSE TRANSPORTATION AND INFRASTRUCTURE COMMITTEE, SUBCOMMITTEE ON AVIATION met for a hearing on Unmanned Aircraft Systems: Emerging Uses in a Changing National Airspace – the link takes you to the video.
FAA Deputy Administrator Daniel Elwell’s written testimony is here.
Morning Transportation reveals the potential regulatory pitfalls:
DON’T HOLD YOUR BREATH: FAA officials Wednesday wouldn’t commit to a timeline for issuing a much-anticipated proposed rule governing drone flights over people, despite being peppered with questions from House Transportation Committee members.
In what appears to be a startling piece of obfuscation:
[Elwell testified that] An Aviation Rulemaking Committee focused on tackling issues associated with remotely identifying drones in the skies recently wrapped up its work, and its report is expected to be released soon. Elwell said it’s undergoing “executive review” right now.
Probably trying to figure out what’s left.
Not a bird, but there’s a plane: This tidbit from Elwell’s written testimony didn’t get much play during the hearing, but …the number of complaints received from pilots spotting unmanned aircraft nearby so far in 2017 “is significantly higher than the number received in 2016 and 2015.”
SPEAKING OF BIRDS: Research released earlier this week showing that drone strikes could be more dangerous to aircraft than birds prompted the Air Line Pilots Association to weigh in. President Tim Canoll said the study “further demonstrates the need to allow public comment.”
This week ASSURE (Alliance for System Safety of UAS through Research Excellence), a consortium of universities that conduct research for the FAA Center of Excellence for Unmanned Aircraft Systems, issued their long awaited The sUAS Air-to-Air Collision Severity Evaluation Final Report. This is the briefing deck. Note that this was Phase One of Four.
Dear readers, I hope you are not surprised that after simulating “over 180 impact scenarios in a twelve-month period,” the study found that drones are potentially more damaging than bird strikes (birds being ‘gelatinous’.) The things that I found most surprising are:
- The windshield is not the most vulnerable point.
- Fixed wings pose a greater threat than quads.
- There is a real possibility of a LiPo fire.
The unassailable, no getting around it conclusions:
- sUAS can introduce severe damage to aircraft structures in case of airborne collision scenarios as shown in this research.
- Even non-severe structural damage levels in the Level 1 to 3 can introduce a significant economic burden to aircraft operators: due to downtime
For a summary, see UASVision.com’s ASSURE Publishes UAS Airborne Collision Severity Evaluation Report.
For more from Canoll and ALPA, see the sUAS News story here.
“Congress must allow the FAA to apply proven safety regulations to hobbyists and help prevent future UAS-aircraft collisions that could hold dire consequences for the traveling and shipping public.”
Think about it – ladies and gentlemen this is your captain speaking…
In another McNabb scoop, FAA Explains the LAANC Program, Miriam notes that:
The FAA article comes several weeks after providers Skyward and AirMap announced that the system was available for testing. These early announcements were the source of a controversy in the industry, as competitors not involved in the initial testing committee claimed that the companies were taking an unfair advantage of their unique position to gain an early market advantage.
Meanwhile as January 6th nears, Administrator Huerta is firming up his plans for an extended ski trip. Still no word about who the new Administrator might be – and do remember that he or she must be approved by the Senate…
Sandy Murdock writing in the JDA Journal offered up Mr. Administrator, Whoever You May Be, Here Is The Collective Wisdom Of A Few FAA Consiglieres On Your Challenges. It’s a thoughtful article, filled with a lot of good management advice. I bring it to your attention for context. In the very concise 1,200 words there is exactly one mention of drones.
THE NEW UK REGS
This is worth reading closely. It reflects what we all know – drones need to be more carefully regulated. sUAS News reports that:
“The UK’s Department for Transport and Civil Aviation Authority (CAA) have announced measures [link is to the press release] that will start to address drone safety concerns in its draft ‘Drone Bill’, scheduled to be published in
- All drones over 250 grams must be registered.
- Anyone wishing to fly a drone must pass a written ‘safety awareness” exam.
- Police officers will be given the right to order drone operators to land their vehicles.
- Police officers will be able to seize drone parts to verify whether the vehicle was used to commit an offense.
The most startling is that:
- Mandatory use of accurate safety apps, such as Guardianor Drone Assist, to check airspace and other local restrictions, weather and for nearby air traffic.
Which is like a license to print money…
There are a number of articles but I found Altitude Angel – Thoughts on the UK’s Proposed Rules to be the most interesting. If you don’t know them, Altitude Angel is a group that has partnered with Microsoft and the Imperial College of London to “enable the next era of aviation.” They are part of GUTMA and the developers of the Drone Assist application.
The following statement is in my opinion equally relevant to the LAANC.
In practice, this means that UK Government should also act to ensure that:
- Only verified suppliers of data to the drone industry can publish so-called ‘safety apps’ and;
- Such suppliers must meet rigorous, new safety standards and have strict, end-to-end data integrity controls built into their platforms.
It is vitally important that any proposed legislation that mandates the use of a technical solution does so in a manner that enables open competition while ensuring that a minimum level of technical compliance is met.
The SUAS-Global Newsletter [from the UK] puts a different spin on the bill, writing that “The Government is clamping down on nationwide use of unmanned aerial vehicles, both commercial and consumer systems.”
In addition, they note that:
The Civil Aviation Authority (CAA) is threatening already squeezed British commercial drone operators with another licence fee hike from April, piling another 40% onto their costs. [A commercial drone license is currently US$250/yr.]
…Prompting anger from the drone community, many of whom have claimed the regulations are not enforced – meaning those already licensed face fee hikes while the great unlicensed continue to remain fee-free.
It seems that enforcement is a universal challenge. By the way, SUAS is an interesting group. “Our main objective from the outset, to provide an online go-to resource for all Drone operator professionals within the Surface, Underwater, Air and Space sectors.” Clever.
I think that I am going to quit reporting the puts, takes, accusations and denials. I am certainly not well enough informed to opine on theories espousing a relationship between DJI and the Chinese government. Nor do I have any insight into whether anyone is desperate enough to forge Department of Homeland Security documents, or if said document is real.
What I am absolutely certain of is that DJI’s credibility and the security of their products is increasingly in question, as is the security of their customer data. Even if all of this is completely misinformed, or if you like conspiracy theories, the result of a deliberate smear campaign; their responses have been less than helpful in allaying concerns.
As everyone has long “known,” DJI currently enjoys ~70% of the market. This was recently confirmed by Colin Snow at Skylogic Research who surveyed 2,600 users worldwide. And by Dan Gettinger and Arthur Holland Michel at Bard College who analyzed the entire FAA registration database – 929,406 as of October 31st.
There is no way out of the public eye for the company. Odds are good that if something bad happens with a drone anywhere in the world, it will involve one of their products. Such is the flipside of success.
Here’s why it matters. Our industry desperately needs standards and leadership. Something the market leader should help to provide. DJI has unmatched R&D and engineering capabilities. Friday’s announcement of voluntary pilot identification is a great example.
Let’s be serious. At this point can you imagine either the FAA ARC or ICAO embracing any of it? It is hard to see the inability of the world’s leading UAS supplier to work collegially and cooperatively with the worlds regulatory organizations as anything but a giant leap backward regardless of the whos and whys.
The problem, of course, is that the regulators need voluntary compliance. It is hard to see a winning long-term strategy.
As an aside and a plea for sanity… Can all of you well-intentioned, self-anointed industry ‘leaders’ please stop with the caterwauling and hyperventilating about how registration will ruin the market and destroy the rights of 14-year-old boys. This is no longer 2015. There are almost one million registered drones. There will be massive confusion when the Defense bill passes and registration is reinstated.
Here too, we need standards and leadership. Let’s look at what the UK is doing and encourage our legislators and the FAA to take the long view. Maybe, just maybe, there is still a chance to close the barn door and accomplish the original objective which was to ensure that everyone who flies a drone actually understands a few simple rules… Who knows, working together we might even save some lives.
THERE’S MORE TO AI THAN
File it under, “Things I’ve always wondered about.” NASA pitted an AI drone against a carbon pilot.
The race on Oct. 12 followed two years of AI research by NASA’s Jet Propulsion Laboratory (JPL), in Pasadena, Calif. JPL’s success in spacecraft navigation attracted Google, who funded the research on drone autonomy.
The results, reported on the NASA website are not entirely surprising. The human pilot cut 20% off of the AI time – cutting the course from 13.9 to 11.1 seconds on his best lap.
While the A.I. and human pilot started out with similar lap times, after dozens of laps, Loo learned the course and became more creative and nimble. … But the drone was more consistent overall. Where Loo’s times varied more, the A.I was able to fly the same racing line every lap.
Which is how you win races…
Air & Space reports on a demonstration Flocking ScanEagles where three ScanEagles executed a mission on their own.
Marshall Williams, the Boeing Advanced Systems program manager who taught this small flock of ScanEagles how to fly, says it’s all about creating programs that respond to the way humans communicate, instead of training people to translate commands so computers can understand them. “It’s a whole new way of telling them what to do,” he says. “You tell them, I need pictures of those things in this space.”
Of course, all things AI are not gold. CNBC reports that Elon Musk tweeted “Got to regulate AI/robotics like we do food, drugs, aircraft & cars. Public risks require public oversight. Getting rid of the FAA [wouldn’t] make flying safer. They’re there for good reason.”
Which I guess qualifies as a shout out to the folks on Independence Avenue.
Throwing some oil on the water, MIT Technology Review writes that Progress in AI isn’t as Impressive as You Might Think.
“Key advances have the capacity to dazzle the public, policymakers, and investors into believing that human-level machine intelligence may be just around the corner. But a new report [The 2017 Artificial Intelligence Index], which tries to gauge actual progress being made, attests that this is far from true.”
Finally, Recode explores the social implications with an opinion piece that opens “Artificial intelligence doesn’t have to be evil. We just have to teach it to be good.”
Thanks for reading and sharing. Back issues of Dronin’ On can be found here.
follow me @dronewriter