This Unmanned Plane Can Land on a Helo Pad

Courtesy of Latitude Engineering via Popular Mechanics

Popular Mechanics published about two weeks ago detailing a project in which Latitude Engineering is developing a hybrid quadrotor-airplane.  This seems like a cool project; the full scale 25-lb model is supposed to have 12 to 15 hour endurance.  It seems they made an impression at AUVSI this year as well.  I have a couple of questions and comments though:

  • It looks like from their photo (below) that the vertical lift is electric and the forward propulsion is gas.  I would wonder if they could still get the advertised endurance having to carry both fuel and batteries in addition to the surveillance payload?
  • This does seem like a much more reasonable design than the StopRotor Hybrid Rotorwing.
  • I’m not sure that the VTOL capability is worth the performance tradeoff, as an aircraft this small would need just a postage stamp of runway to operate.  However, if this 25-pounder is simply a stepping stone to the larger model under development, this may make more sense since it won’t require a fancy recovery system a la Scan Eagle.
Courtesy Latitude Engineering

Act on data (if you have it), not on anecdotes

Being a huge football fan (and former player in high school), I have been both concerned and annoyed with the current intense interest in CTE and concussions of former football players.  Concerned because, if the head trauma associated with the game as it is today causes real, debilitating long-term damage, the players we enjoy watching may be literally bashing themselves to death.  Annoyed because it seems that a certain segment of the public has taken a few very visible anecdotes of players with problems that seem to have resulted from long-term brain damage (notably, Junior Seau), and extrapolated their cases to the whole sport.  Now, there is a clamor for possibly banning the game–even the president said, “If I had a son, I’d have to think long and hard before I let him play football.”

The problem is, this kind of thinking ignores the big picture.  I read two articles (here and here) earlier this week that describe the results of actual research, using science, which show that while NFL players have a higher-than-normal rate of neurodegenerative disease, in the aggregate they live longer, healthier lives than the average person.

Worse still, football “abolitionists” take their interpretation of the data to the extreme conclusion–because of the injury of the few, ban the game for all.  If head injuries are a significant problem (which they probably are), another option is to modify the game–or in this case, start enforcing rules that are already on the books by penalizing players who lead with their heads.

The lesson to be taken here is to take anecdotal claims–especially those that are alarming enough to make the headlines–with a dose of skepticism.  A very clear case where this is so is the reaction to UAVs, which I’ve written about before.  Are UAVs a tool of international war crime or a technology that will enable us to cheaply see things from the sky or pick up a gallon of milk without leaving the house? Wait until the data bears out the answer!


Pizza Delivery From The Sky? Drone Distribution Networks Held Back By Regulators

I’ve criticized the current state of regulation on commercial UAVs before, primarily on the basis that they don’t make much sense (you can fly a quadrotor for fun, but not to do something useful or valuable?)


However, this article from Singularity Hub highlights the potential value of UAVs for micro-deliveries: pizza, groceries, meds, anything that can fit in a shopping bag. I had thought of small UAVs as having potential for aerial photography, but I always thought of aerial delivery as gimmicky (Dominos or tacos). This author points out, though, the immense environmental and safety value and savings on infrastructure by taking a car off the road every time you need a gallon of milk or loaf of bread.

I’m in agreement with this guy–one the FAA and the states get this thing figured out, there will be mass of startups doing aerial micro-deliveries and startups to support them. (If the stars align, maybe I’ll be one!) Lets hope that it’s soon!

Sikorsky Launches Autonomy Research Program


Credit: Ted Carlson/Sikorsky Aircraft

Aviation week reports that Sikorsky Aircraft Corporation has taken a big step forward in unmanned aircraft technology. The short, short version is that they’ve equipped an S-76 (which they’ve named Matrix–or SARA depending whom you talk to) with an autonomous architecture that allows unpiloted, fly-by-wire, or manual control.  This airframe is one of the first, and biggest, steps in a research program to supply large, expensive, low-altitude air vehicles with autonomous capability previously limited to high altitude aircraft (i.e. Predator and Global Hawk) or small, cheap UAS’s.  Without re-hashing the whole article, here are the key points:

  • Reliability for the system is now about 1 failure per 1000 flight hours, which results in very high expected replacement costs.  The target is 1 failure per 100,000 flight hours, which is on par with current manned helicopters.
  • The architecture is being designed to be platform-agnostic, enabling rapid implementation not only on S-76 and Blackhawk, but I assume also (when the time comes) the S-92, CH-53, Raider, and Schweizer models.
  • While autonomous systems cannot yet do all the stuff a pilot can, it is surely more consistent than a human pilot, insusceptible to stress or fatigue. It also has access to information from sensors that can see what the pilot can’t. This means safer flight in night, poor weather, and poor visibility conditions.
  • The objective system will be able to handle actuator failures in a way that a human pilot cannot; a stuck collective for example would be cause to abort mission and make an emergency landing.  The autonomous system would be able to reconfigure flight control and evaluate the risk of continuing the mission with reduced performance.

I, of course, am thrilled–the approach seems to be the right balance of ambition and prudence, and if they succeed it could open the floodgates for UAV acceptance and employment.

Commence jokes about Sikorsky taking the red pill!

Sikorsky Launches Autonomy Research Program | Aviation Week & Space Technology

Sikorsky Unveils MATRIX™ Technology to Advance VTOL Autonomy to Higher Levels of Capability and Safety | Sikorsky Press Release

The $4 Million Teacher

Kim Ki-hoon (Courtesy Wall Street Journal)

The gentleman you see above is possibly the most well-paid teacher in the world.  Pulling in $4 million a year as a teacher seems like a fairy tale, certainly for teachers in the US.  But according to this article in the WSJ, it is reality in the Republic of Korea.

Apparently, given the extreme pressure felt by Korean parents to get their children into the best universities, there is a flourishing market in private tutoring business, called hagwons, in which teachers are paid according to their performance.  If the students are inspired, if the teacher is well prepared, and if the students end up performing well on entrance exams and getting into universities, the teacher is able to draw more students and higher rates.  The poor performers experience the converse, with the worst leaving the market.

This is really intriguing, and made me wonder how this could inform education in the US.  The article addresses a few of the things (such as the fact that one hagwon owner “fires about 10% of her instructors. By comparison, U.S. schools dismiss about 2% of public school teachers annually for poor performance.”) But I wondered:

  • Is there any chance that US parents, at least those in all but the top tax brackets, would shell out bucks en masse for private tutoring? Based on my experience, it seems that many Koreans still remember what it was like when Korea was a poor third-world country–they’re hungry (in the sense of ambition) and driven.
  • How does this phenomenon (at least in case of guys like Kim Ki-hoon) relate to the emergence of the MOOC and resources like Khan Academy?
  • Is this tutorship system providing students with real analytic skills and creativity? Or are they fancy cram systems that teach the test?
  • How can the US manage to stay on top when hundreds of millions (sorry. Billions.) of Asians are willing to sacrifice so much to acheive?

What I wish I knew about finding the right graduate advisor

This is not news, but: A good graduate school experience hinges upon finding the right advisor.

This can be a foreign concept to students who have only completed their bachelor’s degree–I know it was to me. In undergraduate studies, students typically have a checklist for success–get certain grades in your classes; maybe a capstone project, and ba-da-bing: you graduate. This is not quite so with a master’s degree and even less so in a PhD program.

A PhD program, as I can best describe it from my own shoes, is an awful lot like an apprenticeship in a trade guild. There are some classes, to be sure, but they are merely a means to (hopefully) develop some skills you need for the main thing: novel research, and a thesis about that research. There are no classes that can really teach you how to succeed at this endeavor.

Instead, there is a PhD advisor who is there to help you develop and grow into a professional who can conduct a research program themselves. The relationship will start as that of master and apprentice, but end as a collaboration between colleagues.

In light of that, here are some things to consider when you search for your advisor:

  • Confluence of interest and expertise. First and foremost, you need an advisor who can give you technical advice. That means they have to be interested and knowledgeable in specifically what you want to research.
  • Resources available. Your advisor is going be the gateway for you to access experimental equipment and data–you’ll probably have to have access to a wind tunnel to do aerodynamic research, for example.
  • Comfort in communicating. Since you will probably collaborate with your professor more than any other colleague in your career, communication with him or her has to be natural and comfortable, especially in the stressful times when you’re overwhelmed and can’t seem to get anything to work right.
  • Hands off vs hands on. You’re there to learn how to do research on your own, so your advisor has to be able to give you wide latitude to succeed and fail. But, a good advisor won’t let you just flounder for very long. It’s important you find the right balance for your needs.
  • Young vs established faculty. Young professors tend to be energetic and hungry, but have fewer resources to draw from. Established professors aren’t necessarily as interested in moving fast but tend to provide more stability, especially in funding. This is not as important a factor because there are always exceptions to the rule.
  • Employment. Keep in mind that in many cases (as in mine) the advisor is also the supervisor or employer. Ask yourself if you want to work for this person.
  • It’s not just about what’s best for you. Potential advisors are looking for students who can advance their research programs and produce publication-worthy research, thereby increasing their prestige, ability to procure future funding, and for young faculty, obtain tenure. Are you going to be a good investment?

I have to admit, I’ve been fortunate–I didn’t really think much of these things in the process of getting ready for graduate school, but in hindsight, it could have turned out quite poorly.

Drones to fly U.S. skies, FAA approves 1st civilian UAVs

It seems that the FAA has certified two UAV models for commercial use: the Boeing-Insitu ScanEagle and the AeroVironment PUMA. This is really awesome, as it indicates that the FAA is slowly warming up to the idea of unmanned systems in the NAS. On the other hand, this raises a host of questions which none of the articles I checked answered:

  • What does this certification mean, exactly? Since they are still writing the rules for UAV use in the NAS, is it simply that these aircraft can fly in uncontrolled airspace in a for-profit/for-pay role?
  • If there is now a certification process for commercial UAVs, what is the threshold for who has to be certified? One of the draws of UAV use is that they’re cheap, and having to type certify each of the tens of thousands of designs out there seems a bit ridiculous. Aerovironment and Boeing can foot the bill, but what about startups?