This week I thought I might have jinxed the Mars Ingenuity helicopter. In my last column, I wrote about how Ingenuity, and robots in general, often overperformed in their roles as space explorers, meaning that we may not need to risk the lives of human astronauts anytime soon. Ingenuity was designed to accomplish a single proof-of-concept mission to show that flight on Mars was possible. At the time of my previous column, it had already accomplished five flights, complete with full video accompanying each one, and was getting ready to assist the Perseverance Rover with Mars exploration.
Then the helicopter almost crashes on its sixth flight. Even if Ingenuity had crashed, its mission would have still been successful at this point, but I’m glad that it made it back to the ground safely after a very wild ride.
NASA discovered what went wrong and I’m sure they are working to correct it. One key issue is that flying a drone on Mars is not like controlling one zipping around a park on Earth. Lack of atmosphere aside, the biggest hurdle is the fact that it takes between five and 20 minutes for a signal to reach Mars, and the same amount of time for Ingenuity to report back. So it’s not like watching your helicopter weave around trees on Earth and making course corrections as needed. Instead, NASA needs to program the flight ahead of time and then hope the onboard systems and their pre-flight instructions mesh to perform a successful mission.
A Big Trak Trek
That reminds me of a pretty cool toy I had growing up. The Big Trak was an extremely advanced programmable vehicle at the time. Created by Milton Bradly, it resembled a Mars rover in a way with six wheels and a space-like look, though I always imagined it as a tank of the future. Instead of controlling Big Trak directly, you instead used a complex-looking keyboard mounted on the back of the vehicle to stack commands into its memory that Big Trak would later follow. So a sample program might tell it to move forward five units (a unit was the size of the craft itself) and then rotate 45 degrees to the right, fire its forward phaser and then move forward again before turning around or backing up and returning to its point of origin.
There was no monitor, so you had to memorize your commands or write them down as you strung them together, and test out your programs live with Big Trak rolling along. You could also loop commands similar to when programming in the educational Logo computer language.
I believe the model that I had could remember up to 16 instructions. I used to build elaborate obstacle courses in my basement with ramps, army men and little walls. Then I would try to successfully navigate my Big Trak through them, which was a lot of fun but also really challenging. The Big Trak had rudimentary safety controls where if it got stuck pushing against a wall for a time, it would abort its program, though it was still vulnerable to dropping off ramps or other surfaces, so you had to be pretty careful.
Safety controls are what made the Ingenuity go a little haywire on its sixth flight. The helicopter’s navigation camera takes photos that help the aircraft to fly. Only this time, one of the images failed to load, which caused all the ones that followed to have inaccurate time stamps. Ingenuity began wildly correcting its flight path based on the faulty navigation images. Havard Grip, the chief pilot for Ingenuity at NASA’s Jet Propulsion Laboratory, said in a statement that the craft began “adjusting its velocity and tilting back and forth in an oscillating pattern” once it reached an altitude of 33 feet. And it kept doing that throughout the flight.
Unlike my Big Trak, other safety systems on Ingenuity kicked in, and the helicopter was able to land safely within 16 feet of its intended destination. But I bet the controllers back on Earth were sweating that flight as the data began to roll in.
Autonomous Drones Go to War
Back on Earth, we got a much more disturbing report about autonomous drones performing warfighting activities without human intervention. The incident in question actually happened last March during Libya’s civil war, but the United Nations report on it was just released. It involved a weaponized drone called an STM Kargu-2. According to the company that makes it, the drone uses embedded real-time image processing capabilities and machine learning algorithms to hunt down targets without human intervention.
I remember years ago at the old FOSE trade show in D.C., feeling rather small while looking up at an imposing stationary robot armed with machine guns and missiles that was designed to guard borders or other critical facilities. The robot was supposed to be able to identify threats and engage them without human intervention. But when I asked company officials about that, they told me that it would never happen because robots would never be given the authority to kill without human approval.
I guess Turkey, which is the main country using the Kargu-2 drones, doesn’t have the same misgivings about killer robots acting independently. Unlike the robot I met at FOSE, the drone in this incident does not fire weapons, but instead just slams into targets and explodes.
According to the UN Security Council’s Panel of Experts on Libya Report, enemy forces were hunted down and attacked by drones “without requiring data connectivity between the operator and the munition.” The drones identified targets and had the authority to attack without permission from their controllers. And they did that, possibly for the first time in human history.
Flying Cars Taking to the Skies
I didn’t want to end my column on a note about drones attacking people, because as much as I admire the technology, I have seriously mixed feelings about that. So on a happier note, military officials said this week that flying cars might be coming to a battlefield one day soon.
I originally wrote about this for Nextgov back in 2018, and although the military was intrigued by the idea, the technology really wasn’t quite ready for that role back then. The Air Force launched its Agility Prime initiative as part of its Accelerated, Affordable, and Agile Acquisition program to develop it, however, and now envisions flying cars taking soldiers to battle and evacuating wounded to safety.
The military now calls flying cars organic resupply buses (ORBs) and envisions using something like Lift Aircraft’s revolutionary Hexa aircraft to ferry troops into battle. Lt. Gen. S. Clinton Hinote, the Air Force’s deputy chief of staff for strategy, integration and requirements said in a recent interview that flying cars could certainly become a reality for military operations in the near future.
The flying part may be different, but using cars to get to combat isn’t a new concept. During the battle of the Marne in 1914, the French used a fleet of 600 taxi cabs to rush 3,000 troops to battle as emergency reinforcements. They helped to turn the tide and win that battle, even though they tallied up over 70,000 francs in cab fares. Presumably, the military’s new flying cars (or ORBs) would provide the rides for free for our troops.
John Breeden II is an award-winning journalist and reviewer with over 20 years of experience covering technology. He is the CEO of the Tech Writers Bureau, a group that creates technological thought leadership content for organizations of all sizes. Twitter: @LabGuys
source: NextGov
