Unmanned Systems’ Impact on Society

I believe that unmanned ground systems will have the biggest impact on society over the next two decades. Why? Because ground systems are the ones that are most likely to perform functions that people would otherwise do, and the biggest impact will occur when peoples’ jobs are changed or replaced by machines. An obvious and often-discussed area likely to experience some of the greatest impact is that of the transportation worker, such as taxi and limousine drivers, bus drivers, railroad engineers, and others (Tracy, 2015).

Aside from the direct economic impact of jobs being changed or eliminated, another reason that ground systems will have the largest impact on society is that they’ll probably be the ones that most people own and interact with directly on a regular basis. For example, it doesn’t appear likely that there will ever be a time when most people own an unmanned aircraft system, but it does appear likely that there will come a time when most people own a car that’s capable of driving itself (Chuang, 2017). That level of ubiquity and familiarity is likely to drive innovation and reveal possible uses in the area of unmanned ground systems that are difficult or impossible to predict today, just as has happened with computers that are now used in ways that few could anticipate decades ago when most people didn’t own one.

Unmanned ground systems also have the potential to impact society by being used as a vehicle for some people to promote their political beliefs. Just as journalists have leveraged fear of military drones in their anti-war reporting, many are attempting to use peoples’ fear of job loss to “robots” as justification to promote the concept of a socialist “universal basic income” (UBI) that the government would pay to all citizens (Darrow, 2017).

If we apply “unmanned system” in its broadest sense of including not just vehicles but all machines and especially those that are capable of replacing people (often referred to as “robots”) the potential impact becomes even greater and includes things like fast food workers and others in the restaurant industry (McGee, 2017), bricklayers (Sklar, 2015), cashiers and toll both operators (McFarland, 2017), telemarketers (Mahdawi & Chalabi, 2017), security guards (Metz, 2014), bartenders (Wright, 2017), and bank tellers (Grothaus, 2017).

In reality, the changes to jobs and job opportunities are likely to be gradual and less serious than most of the dire predictions. For example, it has been suggested that jobs in the law profession will be significantly impacted by “robots”, but the reality is that the potential for lawyers, paralegals, and others to be replaced is very limited. In fact, research has shown that automation typically increases the number of jobs in a given industry and that it tends to transform jobs rather than eliminate them due to the fact that while specific tasks can be performed by machines, it’s rare that an entire job can be (Markoff, 2016). So while it’s true that ground systems are likely to have a large impact on society, it appears unlikely that they’ll result in the kind of mass unemployment often alluded to by sensationalist reporting.

As with other types of machine, I believe the overall impact of unmanned ground systems will be very positive. Transportation, for example, will become faster, cheaper, and safer for people as machines increasingly take responsibility for tasks they’re able to do better. Changes will occur, but they’ll happen gradually and be less a matter of eliminating jobs than changing them, and people who are willing to adapt will find work doing tasks that machines can’t yet do. In fact, it’s that adaptability that ensures that people will never fully be replaced by machines.



Chuang, T. (2017, March 26). When Can You Buy — Or Try — A Driverless Car? The Denver Post. Retrieved from http://www.denverpost.com/2017/03/26/driverless-car-options/


Darrow, B. (2017, May 24). Automation, Robots, and Job Losses Could Make Universal Income a Reality. Fortune. Retrieved from http://fortune.com/2017/05/24/automation-job-loss-universal-income/


Grothaus, M. (2017, January 19). Bet You Didn’t See This Coming: 10 Jobs That Will Be Replaced By Robots. Fast Company. Retrieved from https://www.fastcompany.com/3067279/you-didnt-see-this-coming-10-jobs-that-will-be-replaced-by-robots


Mahdawi, A., & Chalabi, M. (2017, June 26). What Jobs Will Still Be Around in 20 Years? Read This to Prepare Your Future. The Guardian. Retrieved from https://www.theguardian.com/us-news/2017/jun/26/jobs-future-automation-robots-skills-creative-health


Markoff, J. (2016, January 4). The End of Lawyers? Not So Fast. The New York Times. Retrieved from https://bits.blogs.nytimes.com/2016/01/04/the-end-of-work-not-so-fast/?_r=0


McFarland, M. (2017, September 15). Robots: Is Your Job at Risk? CNN. Retrieved from http://money.cnn.com/2017/09/15/technology/jobs-robots/index.html


McGee, C. (2017, March 28). In a Decade, Many Fast-Food Restaurants Will Be Automated, Says Yum Brands CEO. CNBC. Retrieved from https://www.cnbc.com/2017/03/28/in-a-decade-many-fast-food-restaurants-will-be-automated-says-yum-brands-ceo.html


Metz, R. (2014, November 13). Rise of the Robot Security Guards. MIT Technology Review. Retrieved from https://www.technologyreview.com/s/532431/rise-of-the-robot-security-guards/


Sklar, J. (2015, September 2). Robots Lay Three Times as Many Bricks as Construction Workers. MIT Technology Review. https://www.technologyreview.com/s/540916/robots-lay-three-times-as-many-bricks-as-construction-workers/


Tracy, S. (2015, June 11). Autonomous Vehicles Will Replace Taxi Drivers, But That’s Just the Beginning. The Huffington Post. Retrieved from http://www.huffingtonpost.com/sam-tracy/autonomous-vehicles-will-_b_7556660.html


Wright, B. (2016, March 28). Robots Are Coming For Your Job. The Los Angeles Times. Retrieved from http://www.latimes.com/opinion/op-ed/la-oe-wright-robots-jobs-data-mining-20160328-story.html





Unmanned System Implementation Strategy

To discuss an implementation strategy that addresses the concerns of privacy, ethics, safety, and link loss I decided to ponder how I’d do that for an application that I think will and should become much more commonplace. Specifically, I decided to discuss how to address those limitations when using an unmanned aircraft system (UAS) as part of a search and rescue (SAR) operation involving a missing person lost in an area such as a large wooded park.


Concerns and demands regarding “privacy” are usually vague and poorly defined, probably because the person insisting on having their privacy respected hasn’t seriously considered what it is they’re legally entitled to or can reasonably expect. There’s generally no right to privacy from being observed by other types of aircraft (Friedersdorf, 2014), but people are demanding — and in some cases state and local governments are granting — protection from being observed by drones while on private property (Mather & Chang, 2017). For public property such as a park there may be restrictions against flying a drone recreationally, but there’s generally no legal right to not be observed, photographed, or videotaped from a public place (Estrin, 2012), including the national airspace.

In any case, in our lost-in-the-park scenario, privacy could be respected by simply not flying over areas where campers or other visitors are known to be located. This probably wouldn’t have a significant impact on the effectiveness of the drone use because a missing person would presumably have already revealed himself or herself to anyone who happened to be nearby.


Most ethical codes would suggest that it’s inappropriate to knowingly anger someone without a good reason, but ethics is a double-edged sword and any reasonable ethical code includes not only what a person shouldn’t do but what they should. By most reasonable standards, for example, being annoyed at having a drone flying over you and “invading your privacy” while camping out doesn’t outweigh the importance of finding someone whose life may depend on being found. More to the point, there’s a moral obligation to use whatever tools you can to accomplish that goal even if doing so is likely to mean that some people won’t be happy campers.

However, to address any concerns that the drone would be misused, it could be made mandatory for someone other than the drone’s pilot to review any photographs and video taken by the drone while it was in flight. In addition, the specific areas to fly over and for which images would be captured could be identified and agreed upon ahead of time by those in charge of the operation. As long as these actions were taken, any objections someone might have to being incidentally flown over and even recorded while in a public place would be no more compelling than the complaints of a person included in a photograph taken while they were walking down a public street.


Safety is a relatively easy concern to address partly because regulations have already defined by the Federal Aviation Administration (FAA) that, when followed, are very effective at keeping the national airspace and those on the ground safe. Some specific regulations that are likely to be applicable during a search and rescue (SAR) operation are to yield to manned aircraft (Operation near aircraft; right-of-way rules, 2016), maintain line of sight (Visual line of sight aircraft operation, 2016), avoid flying over people (Operation over human beings, 2016), and to only fly during the daytime (Daylight operation, 2016). As long as these and other applicable regulations are followed the risk of using a UAS in the SAR effort is likely to be minimal.

Lost Link / Loss of System Control

The problem of how to handle a lost link between a UAS and its controller is one that has already been addressed for all but the most inexpensive off-the-shelf drones. For example, some models are designed to record a “home point” location at the time of takeoff and return to and land at that spot if the connection to the controller is lost, a feature sometimes referred to as “return to home” (RTH).

Addressing the possibility of signal loss could involve simply using one of these systems and ensuring that it’s configured to return to home in the case of a lost connection. However, another important and related configuration option allows the pilot to specify a minimum height above ground level (AGL) to which the drone should ascend before it returns home (“Spark User Manual”, 2017). The purpose of the RTH height setting is to help ensure that the drone doesn’t run into obstacles during its return flight, which could destroy the drone itself or even endanger people and property on the ground. In this case, the RTH height should be set so that it will allow the drone to fly well above the tree line and any other nearby obstacles in case its connection to the controller is lost.


In short, by taking some simple and reasonable steps it’s possible to ensure that concerns regarding privacy, ethics, safety, and link loss are addressed before a UAS is used as part of a SAR effort.



Daylight operation. 14 C.F.R. § 107.29 (2016).


Estrin, J. (2012, August 14). Criminalizing Photography. The New York Times. Retrieved from https://lens.blogs.nytimes.com/2012/08/14/criminalizing-photography/?mcubz=3


Friedersdorf, C. (2014, August 28). Why Police Don’t Need Warrants to Snoop With Drones. The Atlantic. Retrieved from https://www.theatlantic.com/politics/archive/2014/08/california-lawmakers-back-a-restraining-order-on-police-drones/379267/


Mather, K., & Chang, C. (2017, August 23). Should the LAPD Test Drones? Police Get an Earful From the Public. The Los Angeles Times. Retrieved from http://www.latimes.com/local/lanow/la-me-ln-lapd-drones-feedback-20170823-story.html


Operation near aircraft; right-of-way rules. 14 C.F.R. § 107.37 (2016).


Operation over human beings. 14 C.F.R. § 107.39 (2016).


Spark User Manual. (2017, June). DJI. Retrieved from https://dl.djicdn.com/downloads/Spark/20170621/Spark+User+Manual+V1.2.pdf


Visual line of sight aircraft operation. 14 C.F.R. § 107.31 (2016).




You Should Have Opened the Pod Bay Doors Like I Told You To

Looking around I found a number of articles discussing the machines-versus-humans-for-space-exploration debate, but the one I decided to write about is titled, “Six Essential Reasons Why We Need to Send Humans to Mars” (Carberry & Webster, 2017), which is actually somewhat of an inaccurate title since it actually lists ten reasons: six “societal” and four “political and commercial”. I don’t agree with all the reasons listed there but I’ll touch on a handful that I feel make the best case for manned space exploration.

I’d probably combine the second, third, and fourth reasons from the first list (“inspiration and innovation”, “prosperity and national morale”, and “security and diplomacy”) into one because they’re all so closely related. Even today countries like India and China use their space programs as a point of national pride (Hunt, 2017) to try to measure up to the “big boys”, which shows that inspiration and national morale are as relevant as ever. In fact, the issue of space debris that came up in this week’s reading material was highlighted in 2007 when China created a lot more debris with an anti-satellite test described in the media as “muscle flexing” and “signaling its resolve to play a major role in military space activities” (Broad & Sanger, 2007). Obviously operating in space is still widely perceived as a benchmark of prosperity as well as technical and military prowess and isn’t just some outdated concept from the Cold War.

Of course that doesn’t directly address the main question of whether people or just machines should be sent into space, but sending machines into space barely even makes it into the headlines unless there’s some kind of novel achievement. For example, in 2014 the European Space Agency (ESA) landed a spacecraft on a comet (Roberts, 2016) but the ESA’s prestige is still well below that of the American and Russian programs. Why? Because just sending a machine into space is easier and cheaper than sending people. In other words, the reason that manned space flight is so much more prestigious is because it’s relatively difficult and expensive, so if you want the benefits that come with the prestige you have to send people.

Another compelling reason mentioned in this article is the first one in the list: discovery and scientific knowledge. In my opinion some of the most interesting, difficult, and probably scientifically rewarding challenges are those related to protecting the human body from the effects of deep space travel. For example, an article titled, “How Our Immune Systems Could Stop Humans Reaching Mars” (Knapton, 2017) talks about how ill-suited the human body is for space travel and covers some of the same issues discussed in the “Living and Working In Space” section from this week’s “The Space Environment” reading (“The Space Environment”, n.d.). It seems certain that in solving these problems that we’ll learn more about the human body, and it’s conceivable that along the way we might find treatments for medical problems faced by people who will never leave the planet.

In my opinion, though, the best reason to send humans into space is because it’s the most practical choice. There are machines that are stronger than people and those that are faster than people. There are also machines that can perform certain tasks — like arithmetic or spell-checking — better than people; in fact, I’m using one as I write this. But when you go into a harsh, poorly understood, and remote environment like that of, say, Mars, what you need to in order to succeed is a jack of all trades, not a master of one or a few. For all the drama queen hand waving about how machines are going to leave us unemployed, the reality is that we’re a very long way from creating a machine with the creativity, flexibility, and adaptability of a human being either physically or intellectually, much less both.

Probably the best way to demonstrate why we need people on our spacecraft is to remember the events just prior to Apollo 11 landing on the moon. Instead of being above a smooth plain the crew found themselves running out of fuel over a “vast crater field and collections of truck-sized boulders”. Their landing computer, which no doubt was state-of-the-art for its time, was rendered useless as a result of being overwhelmed by too much data (Pyle, 2014), but of course we all know how that story ends. In a nutshell, Neil Armstrong used his judgment and gut instincts to achieve the climax of what’s arguably mankind’s greatest technological achievement to date. Had those men not been on that spacecraft we’d instead have all learned about the crash of Apollo 11.

I like machines as much as anybody, especially high tech ones in general and computers in particular, but they aren’t a substitute for people yet and won’t be in the foreseeable future, especially for complex and unpredictable missions like space flight.



Carberry, C., & Webster, J. (2017, January 17). Six Essential Reasons Why We Need to Send Humans to Mars. Fox News. Retrieved from http://www.foxnews.com/opinion/2017/01/17/six-essential-reasons-why-need-to-send-humans-to-mars.html


Hunt, K. (2017, February 15). India in Record Satellite Launch as Asia’s Space Race Heats Up. CNN. Retrieved from http://www.cnn.com/2017/02/13/asia/india-china-asia-space-race/index.html


Broad, W., & Sanger, D. (2007, January 19). Flexing Muscle, China Destroys Satellite in Test. The New York Times. Retrieved from http://www.nytimes.com/2007/01/19/world/asia/19china.html?mcubz=3


Roberts, E. (2016, October 4). Space Probe Finds Lost Philae Lander on Comet. CNN. Retrieved from http://www.cnn.com/2016/09/05/europe/philae-is-found/index.html


Knapton, S. (2017, September 2). How Our Immune Systems Could Stop Humans Reaching Mars. The Telegraph. Retrieved from http://www.telegraph.co.uk/science/2017/09/02/immune-systems-could-stop-humans-reaching-mars/


The Space Environment. (n.d.). Federal Aviation Administration. Retrieved from https://www.faa.gov/about/office_org/headquarters_offices/avs/offices/aam/cami/library/online_libraries/aerospace_medicine/tutorial/media/III.4.1.2_The_Space_Environment.pdf


Pyle, R. (2014, July 21). Apollo 11’s Scariest Moments: Perils of the 1st Manned Moon Landing. Space.com. Retrieved from https://www.space.com/26593-apollo-11-moon-landing-scariest-moments.html



Why Did the U.S. Army Ban DJI Drones?

Not long ago I saw some stories about a United States Army memo that prohibited the continued use by Army units of small unmanned aerial system (sUAS) models manufactured by Chinese company DJI. I went back and reviewed them and found the one by Engadget (Locklear, 2017) was my favorite, so I thought I’d share it and my thoughts on it with you.

The Army’s DJI drone ban was said to be due to “operational risks and vulnerabilities” identified by the Army Research Laboratory and United States Navy (Locklear, 2017) but no details of the risks and vulnerabilities were provided. There’s speculation that it was a result of an article by sUAS News claiming DJI is able “to collect audio, visual, and telemetry data on all flights across the Globe [sic]”, that “the details shared here are perhaps known to a limited number of worldwide owners and users of the DJI technology”, and that the information collected by DJI is accessible “using a simple Google search” (Pomaski, 2017).

It’s also speculated that instead of or in addition to the supposed exposé by sUAS News that the Army’s memo may have been a reaction to the widely revealed flaws in DJI software that allow its products’ owners to override things such as maximum height and no-fly zone (NFZ) restrictions (Locklear, 2017).

I agree that one or both of those might be the “risks and vulnerabilities” the Army is concerned with, but if that’s the case then the Army should reconsider its decision. With regard to the sUAS News story, even that story acknowledges that the information only gets sent to DJI when “a pilot is using the DJI GO 4 app and uploads a flight record to the DJI server” and that “legitimate reasons do exist for the manufacturer to review some details of a flight’s record” (Pomaski, 2017).

The suggestion by sUAS News that the transmission of this information represents something known by only a small number of people is an odd claim given that DJI openly discusses how to upload the flight records, indicating that those records will be “synced with our servers” when the user performs the function. DJI also clearly indicates that they keep a copy those records indefinitely, mentioning that the records are downloaded back to the user’s mobile device if the user deletes them from their device and then syncs with the DJI server (DJI Support, 2016).

In addition, research into DJI’s data collection by the National Oceanic and Atmospheric Administration (NOAA) late in 2016 found “no threat for data leakage” and that, “the majority of transactions to the DJI servers were to . . . check for software updates [which is] quite common for software of this type, and nothing unusual was detected” (Popper, 2017). Lastly, the claim by sUAS News that the uploaded data can be accessed by parties outside DJI using a “simple Google search” has never been substantiated or repeated by any other source.

With respect to the hacks that allow DJI aircraft owners to bypass height and NFZ restrictions, this also doesn’t represent a compelling reason for the Army to abandon DJI aircraft. If anything, they allow the drones become even more flexible and useful even in those places where DJI has defined a no-fly zone. These hacks could potentially also be useful to America’s enemies (Corfield, 2017), but that fact doesn’t strengthen the case against their use by the U.S. Army.

Having said all that, concerns about security breaches by foreign companies and governments aren’t necessarily irrational. For example, years after IBM sold its personal computer line to China-based Lenovo, a serious security vulnerability called “Superfish” (Fox-Brewster, 2015) appeared that may have been intentionally introduced and used by the Chinese government (Croke, 2015). With that in mind it’s appropriate for the Army to be cautious, but hopefully their memo on the use of DJI drones and the “increased awareness of cyber vulnerabilities” that led up to it are based on more than overly dramatic internet stories from poorly informed self-proclaimed internet “experts”.

In summary, I’m doubtful that the sUAS News story and the DJI drone hacking situation have much real relevance to the Army’s use of DJI drones. However, it’s clear that the potential threat of drones being compromised by a hostile government is real, and that such a compromise might have a negative impact on U.S. military and intelligence services. In my opinion, the real moral of the story is that the United States needs to have a trustworthy source of affordable, easily obtained, and full-featured sUAS models.



Locklear, M. (2017, August 4). US Army Reportedly Ceasing Use of All DJI Drone Products. Engadget. Retrieved from https://www.engadget.com/2017/08/04/us-army-ceasing-use-dji-drone/


Pomaski, K. (2017, May 16). A Global Information Gathering Network for UAS – DJI Data Collection. sUAS News. Retrieved from https://www.suasnews.com/2017/05/global-information-gathering-network-uas-dji-data-collection/


DJI Support. (2016, August 9). DJI Go App – Syncing DJI GO Flight Records [Video File]. Retrieved from https://www.youtube.com/watch?v=brKQZ-gTBXg


Popper, B. (2017, August 7). A Government Study Found DJI Drone, Banned By US Army, Kept Data Safe. The Verge. Retrieved from https://www.theverge.com/2017/8/7/16106810/dji-drone-banned-government-study-data-safety


Corfield, G. (2017, April 26). Drone Maker DJI Quietly Made Large Chunks of Iraq, Syria No-Fly Zones. The Register. Retrieved from https://www.theregister.co.uk/2017/04/26/dji_drone_geofencing_iraq_syria/


Fox-Brewster, T. (2015, February 19). Superfish: A History of Malware Complaints and International Surveillance. Forbes. Retrieved from https://www.forbes.com/sites/thomasbrewster/2015/02/19/superfish-history-of-malware-and-surveillance/#31aa5fe527b9


Croke, P. (2015, February 24). Lenovo’s PCs Spy for China. Baltimore Post-Examiner. Retrieved from http://baltimorepostexaminer.com/lenovos-pcs-spy-china/2015/02/24


Are Unmanned Commercial Ships Almost Here?

A lot of what we’ve been reading about and discussing has to do with unmanned maritime vehicles (UMVs) intended for military use, but there’s also a lot of potential for use in the civilian sector. Something that that I enjoyed and learned a lot from on that topic was an article written by a vice president at Rolls-Royce, which apparently is very involved in automating ships, and he predicts that they’ll be here sooner rather than later (Levander, 2017).

In this article the author says that he believes that the first autonomous ship will probably be a harbor tug or ferry, and that autonomous ships will be common within 10 or 15 years. Early in the article he briefly discusses how industry and governments are working to make that happen and to determine how UMVs will fit into the bigger picture.

Next he talks about some of the advantages of a remotely operated or fully autonomous ship. For example, he claims they’ll increase safety and I’m inclined to think that he’s right because – as with ground transportation – humans appear to be the weakest link in the chain. According to the U.S. Coast Guard, up to 96% of all marine casualties are caused by human error (Minter, 2017), which is similar to the percentage cited for other human-operated vehicles such as cars (Schaper, 2016).

The author also points out that autonomous or remotely operated ships can increase efficiency and potentially reduce costs. This would be accomplished by facilitating ship designs with little or no room on board for human operators, resulting in something that allows more cargo to be carried with less wind resistance. The savings would come not just from the space actually occupied by the ship’s crew such as the deckhouse and crew quarters but also from the ability to eliminate things like ventilation, heating, and air conditioning. In fact, by one estimate 44% of a ship’s costs are associated with accommodating its human occupants (Minter, 2017), so it doesn’t seem too much of a stretch to believe that having no crew on board could save a lot of money.

Another advantage of unmanned ships discussed would be the elimination of piracy where crews are held for ransom. For example, the ongoing problems with pirates from Somalia such as the highly-publicized incident involving the Maersk Alabama (Axe, 2012) on which the Captain Phillips movie was based wouldn’t have occurred if ships sailing through that area were unmanned. And if a remotely controlled or autonomous ship were boarded by pirates, it could be made to stop or to travel in a circle until they had been dealt with.

Another problem the author suggests could be solved by unmanned ships is one that has also been mentioned in the context of the trucking industry, namely a shortage of labor (Wee, 2014). While the jobs have become more complex, the number of people willing to live on ships and spend extended periods of time away from their families has been declining, and autonomous or remotely-operated ships could mitigate this problem the same way it’s expected to help the trucking industry (Levander, 2017).

One potential new problem with unmanned ships is the issue of how maintenance would be handled. The author acknowledges this but is somewhat dismissive of the concern, suggesting that the ship’s systems will be closely monitored and that when a problem is detected, “preventative maintenance can be scheduled at the next port of call or, if need be, by dispatching people to make repairs while the ship is still at sea” (Levander, 2017).

While it is true that there are efforts underway to better automate and plan ship maintenance (Pozniak, 2017), the author’s explanation for how to address this issue seems simplistic to me. Cruise ships spend much of their time in littoral areas with calm seas and yet medical evacuations from them are notoriously expensive (Potter, 2015). Likewise, it would seem that delivering and retrieving a maintenance crew to and from a ship that might be in rough water in the middle of an ocean would be cost prohibitive for anyone attempting to use unmanned maritime vehicles. This is a problem that’s somewhat unique to marine vehicles because ground and air vehicles spend relatively little time in remote areas where reaching them for maintenance purposes is problematic. Combine that with the fact that ships by nature operate in environments hostile to machinery (corrosion, temperature extremes, etc.), and you have what I believe is a fairly compelling problem that the author doesn’t address very convincingly.

In summary, I agree with the author’s contention that there are major benefits that could be realized by transitioning to autonomous or remotely operated ships, specifically in terms of safety and efficiency. On the other hand, I’m doubtful that maintenance can be made predictable and routine enough for it to be economically viable to transport repair crews to and from commercial ships that travel far from land. Partly for that reason, I’m also skeptical that, “oceangoing cargo ships [will] be routinely plying the world’s seas in 10 or 15 years” (Levander, 2017). Overall I’m bullish on the future of unmanned marine vehicles, but I’m not as optimistic as the Rolls-Royce VP seems to be about where and how widely they’ll be used.



Levander, O. (2017, January 28). Forget Autonomous Cars — Autonomous Ships are Almost Here. IEEE Spectrum. Retrieved from http://spectrum.ieee.org/transportation/marine/forget-autonomous-cars-autonomous-ships-are-almost-here


Minter, A. (2017, May 16). Autonomous Ships Will Be Great. Bloomberg. Retrieved from https://www.bloomberg.com/view/articles/2017-05-16/autonomous-ships-will-be-great


Schaper, D. (2016, October 20). Human Errors Drive Growing Death Toll in Auto Crashes. NPR. Retrieved from http://www.npr.org/2016/10/20/498406570/tech-human-errors-drive-growing-death-toll-in-auto-crashes


Axe, D. (2012, October 17). 8,000 Miles, 96 Hours, 3 Dead Pirates: Inside a Navy SEAL Rescue. Wired. Retrieved from https://www.wired.com/2012/10/navy-seals-pirates/


Wee, H. (2014, August 20. Keep on Truckin’? Inside the Shortage of US Truck Drivers. CNBC. Retrieved from https://www.cnbc.com/2014/08/19/ckin-inside-the-shortage-of-us-truck-drivers.html


Pozniak, H. (2017, January 3). SEA-CORES Software Could Transform Ship Maintenance. The Telegraph. Retrieved from http://www.telegraph.co.uk/education/stem-awards/defence-technology/sea-cores-ship-maintenance-software/


Potter, E. (2015, July 6). Five Myths About Medical Evacuations. USA Today. Retrieved from https://www.usatoday.com/story/travel/advice/2015/07/06/medical-evacuation/29766691/


The Impact of Self-Driving Trucks

I’ve noticed that discussions of unmanned systems often have a negative or at least skeptical tone, playing on fears of soulless machines that will kill without conscience or at a minimum will take your job, leaving you unemployed and living in a cardboard box or in a van down by the river. Or maybe they’re going to do both: take your job and then kill you.

In some ways I think people see self-driving ground vehicles as more of a threat than missile-equipped drones, which makes sense because autonomous vehicles represent something that can affect them or someone they know personally in the form of a lost job. But just how real and how imminent is the threat? I came across a good article on that topic titled, “Self-Driving Trucks” (Freedman, 2017) in the MIT Technology Review’s recent issue covering “10 breakthrough technologies” that “have staying power [to] affect the economy and our politics . . . or influence our culture”.

The article opens with a truck driver’s description of an accident he was involved in and asks, “Could a computer have done better at the wheel? Or would it have done worse?” It then goes on to talk about advances made in recent years in the area of self-driving vehicles, discussing the advantages of trucks that drive themselves. For example, saving fuel by “platooning” (drafting) to reduce wind resistance is mentioned, and by one estimate can result in an average fuel savings of ten percent (Vincent, 2016). Given that fuel is said to represent a third of the cost of operating a long-haul truck (Freedman, 2017), the benefit to individual drivers and the industry as a whole from this strategy alone would be substantial.

Probably the biggest potential impact of self-driving trucks is the same thing that’s often mentioned in relation to cars: that automation is expected to have a major effect — in a good way, of course — on the number of deaths and injuries from accidents. Given that an estimated 90% of all accidents are caused at least partly by human error (Smith, 2013), it seems pretty realistic that machines that won’t become tired or distracted can eventually be better drivers than us.

The article suggests that the economic impact of self-driving trucks is likely to be pretty large, but it also emphasizes that the impact will be gradual and long term. With regard to truck drivers losing their jobs it specifically mentions that one of the companies pursuing self-driving trucks, “insists it has no plans to introduce products intended to operate trucks without a driver in the cab” and includes a quote from a product manager who predicts that truly autonomous trucks are “at least a decade away”. Even that may be optimistic given that Goldman Sachs recently predicted that the peak impact of unmanned vehicles on truck driver employment is “several decades away” (Balakrishnan, 2017).

The reasons for wanting to keep humans in trucks aren’t all humanitarian, though. There are also practical barriers, many of which are the same obstacles that complicate the use of autonomous cars. For example, a self-driving truck has to be able to deal with situations such as construction, debris in the road, and accidents like the one described at the beginning of the MIT article. I also read a different article which pointed out that truck drivers regularly deal with issues related to their cargo before, during, or after a trip, and that’s something not even the most advanced self-driving vehicle can handle (Sherman, 2017).

Another theme of the MIT article is that the impact of self-driving trucks will be less a matter of eliminating jobs and more about changing them. It suggests that trucking jobs will become less demanding and more appealing, which in turn could help address the well-documented (Wee, 2014; Wolff-Mann, 2015) shortage of truck drivers in America that Michael mentioned in his Topic 1.4 discussion on “Unmanned Systems… Do We Need Them”.

The article does mention that there may sometimes be disadvantages associated with self-driving trucks. For example, lower transportation costs combined with an increased supply of truck drivers could result in a reduction in income for those drivers. This has been discussed in other places and sometimes includes mention of a possible “trickle-down” effect on those who earn a significant part of their income from truck drivers. For example, owners of and workers in restaurants, hotels, and truck stops could all be impacted by a transition to unmanned trucks (Zhang, 2015).

To summarize, though, I agree with each of the key points made in the MIT article:

  • Self-driving trucks will become a reality: The benefits of this technology — namely in lives and money saved — are compelling and will overcome any opposition. Computer-driven cars have demonstrated their safety during testing and real-world use, and trucks will too.
  • The transition to self-driving trucks will be gradual: Technological, political, and economic limitations will slow but not stop the adoption of self-driving vehicles, including trucks. Technological advances often face initial fear and opposition but eventually always move forward and find their place.
  • The net effect on the economy will be positive: Jobs will be lost and salaries will decline in some areas, but new and better-paying jobs will appear in others. This sometimes seems cruel and unfair on a micro (individual) level, but it’s just the nature of capitalism and that kind of “creative destruction” has consistently shown to be beneficial to the economy and society as a whole.



Freedman, D. (2017). Self-Driving Trucks. MIT Technology Review. 120(2), 62-71. Retrieved from https://www.technologyreview.com/s/603493/10-breakthrough-technologies-2017-self-driving-trucks/


Vincent, J. (2016, April 7). Self-Driving Truck Convoy Completes Its First Major Journey Across Europe. The Verge. Retrieved from https://www.theverge.com/2016/4/7/11383392/self-driving-truck-platooning-europe


Smith, B. (2013, December 18). Human Error as a Cause of Vehicle Crashes. The Stanford Law School Center for Internet Society. Retrieved from http://cyberlaw.stanford.edu/blog/2013/12/human-error-cause-vehicle-crashes


Balakrishnan, A. (2017, May 22). Self-driving cars could cost America’s Professional Drives Up To 25,000 Jobs a Month, Goldman Sachs Says. CNBC. Retrieved from https://www.cnbc.com/2017/05/22/goldman-sachs-analysis-of-autonomous-vehicle-job-loss.html


Sherman, E. (2017, June 14). Why Self-Driving Trucks Will Still Need Drivers. Trucks.com. Retrieved from https://www.trucks.com/2017/06/14/self-driving-trucks-need-drivers/


Wee, H. (2014, August 20). Keep on Truckin’? Inside the Shortage of US Truck Drivers. CNBC. Retrieved from https://www.cnbc.com/2014/08/19/ckin-inside-the-shortage-of-us-truck-drivers.html


Wolff-Mann, E. (2015, October 12). Need a Job? America Has a Giant Shortage of Truck Drivers. Money. Retrieved from http://time.com/money/4070028/american-truck-driver-shortage/


Zhang, S. (2015, May 21). Self-Driving Trucks Are Going to Kill Jobs, and Not Just for Drivers. Gizmodo. Retrieved from http://gizmodo.com/self-driving-trucks-are-going-to-kill-jobs-and-not-jus-1705921308