Boeing’s Plan For Autonomous Flight – OMG! Do they NOT learn anything?!
To Whomever it concerns:
• Boeing had the MB-47 pilot-less 6 engine swept wing jet in 1953. My father had a relationship to it.
• Boeing had the QB-47 pilot-less 6 engine swept wing jet in 1961. My father had a relationship to it.
• Boeing had 6 droned jets on 9/11; myself and Kristine Marcy had a relationship to it.
• MH370 was droned on 8 March 2014 and Malaysia and MAS called me to Malaysia to explain. I went.
• MH17 was also a droned B777 registration 9M-MR0
When Boeing says they are ‘testing’ a pilot-less transport category aircraft….they are not truthful (READ BELOW).
Perhaps those truly interested should review Civil Case 1:08-1600 (RMC) which exposed the pilot-less Boeings used on 9/11.
I am certain that Suzanne Kalfus and James Johnson of ALPA could testify as could FAA Huerta, Babbitt and ( redacted ). The last 4 presidents of ALPA were aware and suppressed also: Randy Babbitt, Duane Woerth,
Jon Prater and Lee Moak. Misprision of Felony wrongful death anyone?
Both Team Trump and Russian Defense Attache have received this communique earlier.
Boeing’s Plan For Autonomous Flight [We’ve already got it. It’s called BUAP!]
Jun 9, 2017 Guy Norris | Aviation Week & Space Technology
Boeing plans to flight-test an autonomous civil aircraft in the next two years, as part of a broad study aimed at proving whether or not aircraft with reduced crew, single or even no pilot could be operated for passenger and freight missions with the same levels of safely and integrity as current manned aircraft.
The study, which Boeing is disclosing for the first time, is already underway at Moses Lake, Washington, with ground tests of a modified vehicle designed to test autonomous taxiing, to be followed soon after by tests of autonomous flight algorithms in a simulator. Initial flight tests of the same artificial intelligence-based system will take place in 2018 using a pair of modified Cessna Caravans believed to have been obtained by Boeing for the program.
Further testing of autonomous ground operations, including engine start, pushback, taxiing, maneuvering and takeoff roll, may also be tested in 2019 using a 787 as part of Boeing’s extended EcoDemonstrator technology testbed program.
STEPS TO AUTONOMY
Boeing autonomous study tackles safety and looming pilot shortage
Ground tests of artificial intelligence algorithm for auto-taxi underway
Flight tests in light aircraft planned for 2018
Auto ground ops to be tested in 787 EcoDemonstrator in 2019
Boeing plan leverages unmanned aerial systems and underwater expertise
Boeing’s initiative comes amid a flurry of autonomous vehicle developments in the air, sea and ground, some of which the company hopes to leverage. While self-driving cars have grabbed much of the headlines for ground vehicle autonomy, Boeing has developed relevant expertise through a series of fully autonomous submarines produced by its Phantom Works unit, the latest of which is the Echo Voyager unmanned undersea vehicle. The company’s autonomous experience in the air has been built through unmanned air vehicle developments such as the company’s InSitu ScanEagle.
Elsewhere, DARPA’s Aircrew Labor In-Cockpit Automation System (ALIAS) program is showing how autonomy could be used to reduce workload and improve safety in existing aircraft, while at the same time building trust in automation to pave the way toward reduced crew sizes on certain missions. Following work with Aurora Flight Sciences andSikorsky, ALIAS has moved into a third phase with Sikorsky improving, expanding and certifying the capabilities developed through a tailorable autonomy kit for installation in both fixed- and rotary-wing aircraft.
“This isn’t a quest to take pilots out of the cockpit. This is a quest to ensure that the autonomous system maintains the same level of safety and integrity as we have today,” says Mike Sinnett, Boeing Commercial Airplanes vice president for product development. The primary driver for the autonomy project is a concern that the annual 4.8% growth in air traffic will outstrip the ability of the current training infrastructure to supply the estimated 1.5 million pilots that will be needed over the next 20 years.
“It begs the question of where all the experienced pilots are going to come from. We come from an environment where the military has provided those stick-and-rudder pilots with a lot of reserve aeronautical experience. That is not as true today as it was 10 or 20 years ago, and it will become increasingly less true as time goes on. This does present a concern for us,” says Sinnett. The issue will become even more acute if the explosion in urban air mobility, personal air vehicles and new small, regional air transport occurs as predicted, he adds.
“Will we really have all the pilots available to operate all these smaller aircraft? There has got to be a transition away from the requirement to have a skilled aviator operating the aircraft tactically to having a system that operates the vehicle autonomously. The question is, can we do that with the same level of safety and integrity as we have today? And that is a huge ‘if,’'” says Sinnett.
Boeing’s approach is, therefore, driven by rigid safety requirements that Sinnett says are “vastly different” from the scenario faced by the automobile industry. In the U.S. alone, there were 40,000 fatalities from automobile accidents in 2016, representing a 14% increase over 2014. “The auto industry has a different bar to us,” Sinnett says. “They have got to get below 40,000 fatalities. We have got to be as good as zero. It drives a very different way of thinking about the problem.”
Boeing’s autonomy test plan will use a Cessna Caravan, similar to the one used in the DARPA/Sikorsky ALIAS autonomy program. Credit: Sikorsky
Boeing says many of the individual technologies to enable autonomous operations already exist, but “not the overall aviation infrastructure architectures,” he says. “They are not yet sufficient to accommodate the widespread levels of autonomous operations at the same levels of safety and integrity. So our job is to understand where the gaps are, and then put together research plans that help us fill those gaps. That is what we are in the middle of doing right now.”
The study is also assessing if the path to autonomous operations can be tackled in a series of steps—or in more dramatic but much swifter leaps. Potential steps could include reducing the number of supernumerary crews who currently fly on longer-range missions. “Another step along the way might be to go from two pilots during cruise to one, with another pilot onboard getting meaningful rest,” says Sinnett, who cites the precedent of current single-pilot operations on some smaller passenger aircraft.
“We as a society are prepared to accept single-pilot operations,” he says. “So if it is OK for a single pilot to fly 10 passengers in a certain aircraft type, why would it not be OK for a single pilot to fly a freighter? So what are the steps it would take to allow that?”
Top technology challenges to be tackled for autonomous ground operations include pushback and engine start, taxi and takeoff. For flight operations, key challenges include autonomous sense and avoid, flightpath optimization and contingency planning. Infrastructure and environment topics include integration of an autonomous aircraft into airspace as well as the management of conflicts, cooperative trajectories and aircraft health management.
“Decisions will need to be made on the aircraft that are made by humans today,” says Sinnett. Because it is impossible to pre-program a deterministic system to anticipate the myriad of potential scenarios and options that pilots handle today, the program is evaluating a system that would rely on a non-deterministic decision-making protocol. A non-deterministic algorithm is an algorithm that, even for the same input, can exhibit different behaviors on different runs, as opposed to a deterministic algorithm that will repeatedly produce the same output from the same input.
“So we are doing early exploration with machine learning and artificial intelligence, to start understanding what that might mean for an aircraft. Is the requirement to have non-deterministic behavior in decision-making a deal-breaker in autonomous operations? Is that the one thing that might prevent us from doing this, or is [that] something we can address differently from how we certify aircraft today?” questions Sinnett.
Testing of the autonomous flight algorithm will involve repeating several identical scenarios, each with the same set of identical inputs, and recording the response. “We will qualify the improvements in the decision-making process of the artificial intelligence algorithm and we will learn from it. We will continue to expand our understanding through additional test cases and simulations. Then we will sit down with regulators and start talking through it,” says Sinnett.