Friday, April 17, 2015

Share |



Guidelines for Safe On-Road Testing of SAE Level 3, 4, and 5 Prototype Automated Driving Systems (ADS)


March 26,2015 "SAE International has created the first-ever safety guidelines for on-road testing of prototype models of fully automated vehicles.


J3018 Safety Guidelines for the On-Road testing of Prototype Models of Fully Automated Vehicles is an important document as more fully automated vehicles with Advanced Driver Assistance Systems (ADAS) will be tested on public roads. Read more


New York City Taxicab Transportation Demand Modeling for the Analysis of Ridesharing and Autonomous Taxi Systems (Draft)

AJ Swaboda'15 April 2015, Using the NYC T&LC file of all taxi trips taken in NYC in 2013, the casual ride-sharing potential of NYC's taxi patrons is computed under various levels of service consisting of initial departure delays  of up to 5 minutes (waiting for ride-share opportunities) and the number and geographic scope of ultimate destinations.


For the 5 most active departure locations the casual ride share potential is high (45% reduction in loaded taxi Miles and a 54% reduction in Dispatched taxis throughout the year, if departures are delayed 5 minutes).  The spatial and temporal diffusivity of the rest of NYC's taxi trips is so large that a departure delay of 5 minutes enables the taxi miles saved to be only 23% and dispatched taxis are reduced by only 34%.


Note: while expanding the ridesharing analysis to all trips would substantially increase the ridesharing; however, since these trips are currently taken by walking or conventional bus or subway, there is no societal benefit associated with enticing them to share taxi rides.  They are sharing only with other existing conventional mass transit users.  Alain


Truly" Empty Vehicle Repositioning and Fleet-Sizing: Optimal Management of an Autonomous Taxi System in New Jersey on a Typical Weekday


Kyle Douglas'15 April 2015 "The thesis outlines a series of nai ve empty vehicle repositioning and fleet-sizing  policies before providing a formal and detailed mathematical model for the empty vehicle repositioning  problem speci fic to an autonomous taxi system. It then provides policies for empty vehicle repositioning and fleet-sizing based on the model. Read more


Self-driving cars? Pah! How about just meeting basic safety standards first?


By Ryan Bubear: Columnist on 16 April, 2015 "...Of more immediate concern is the shockingly low level of basic safety still displayed by so many of today’s cars on sale in low-income markets. Indeed, a recent Global NCAP report revealed just how worrying things are, with chairperson Max Mosley stating that “crash test standards introduced twenty years ago for cars sold in Europe are yet to be met by many new cars, and even brand new models, being sold today in leading middle-income countries in Africa, Asia, and Latin America”...." Read more  Absolutely!! Even worse in trucks and buses.  Alain


How Will Attorneys (or any of us) Adapt?


Judge David Langham 04/16/2015  "... If software decreases the need for attorneys, is there a marked effect on non-lawyers? If a hamburger churning machine replaces three employees at the local fast food joint, is there a significance to the consumer? ... " Read more Surprised that I included this one.   Attorneys Adapt??? :-)  Alain


Watch: Audi's self-driving car in action


Alex Robbin 18 Apr 2015  The self-driving version of the Audi A7 Sportback is only a prototype at the moment, and there are numerous legal issues to be overcome before it can go on sale. However, the technology feels remarkably close to being production ready.  In fact, the car is so competent at changing lanes and accelerating and braking as needed that it isn't long before I feel completely comfortable taking my eyes off the road. ".. Read more Just watch the video and the comments by the driver Alain


Planes without pilots? After Alps crash, some experts ponder autonomous jets


AP April 17, 2015 To improve airline safety, maybe we need to remove the pilots.


That radical idea is decades away, if it ever becomes a reality. But following the intentional crashing of Germanwings Flight 9525 by the co-pilot, a long-running debate over autonomous jets is resurfacing. At the very least, some have suggested allowing authorities on the ground to take control of a plane if there is a rogue pilot in the cockpit. ...

There was a time when riding an elevator without an operator seemed unimaginable. Today, we don't think twice about stepping into an empty elevator. Airports around the world have trams without drivers, as do some subways systems. Even cars are starting to take some of that control away from us: the latest models will automatically brake if there is a sudden hazard...."Read more  Yup!  Only difference is that the way our current system operates is mass transit-ish:  Big vehicles, each with many passengers that jointly pitch in to pay for the expensive pilots.  Which is why we all have to scramble to get to/from major airports and change planes in order to aggregate enough of us to be able to pay for the pilots.  (We really don't need the stewardesses.  We've now learned to bring our own brown- bag lunch).  However, if automation replaced the pilots, then the economics of smaller planes would change substantially leading to more frequent point-to-point services which actually match where we are coming from and going to, when we want to do it.  Think about it... Air travel breaking out of its conventional mass transit mode of operation and becoming more autonomousTaxi-like! I love it!  Maybe I should do an "Uber"  soliciting the participation of recreational pilots  in preparation for the aPlane market.  ;-)  Prof. K.


  Why Pilots Still Matter


Patrick Smith, April 10, 2015 "...The automation only does what we tell it to do. On the 767 that I fly, there are multiple ways to set up and command any routine climb, descent or change of course. Meanwhile, more than 99 percent of landings, and a full 100 percent of takeoffs, are performed manually..."  Read more


Airplane! (2/10) Movie CLIP - Automatic Pilot (1980) HD


The 4 Best Stocks to Invest in Autonomous Driving


Sam Mattera - April 16, 2015 "...For investors, the upside could be equally as impressive. Although the industry remains in its early days, several companies are poised to take advantage of the shift. Below are four stocks that offer exposure to the coming autonomous car revolution.... Read more   Hmmm... the no-brainers! Alain



Recompiled Old News :

Deep learning could help safely automate cars

March 27, 2015 "Speakers at the recent Nvidia GPU Technology Conference suggested that it would be more efficient to have machines teach themselves how to respond to the infinite changes that occur as autonomous vehicles drive from site to site. Deep learning has been used by companies such as Google and Netflix to help predict what consumers are looking for. Some observers feel various forms of machine learning will be a disruptive technology that changes many fields, possibly including automotive.


“Deep learning is arguably as exciting as the creation of the Internet,” said Jen-Hsun Huang, Nvidia’s CEO. “Humans can’t code if-then-else statements for all the situations vehicles face. We want to augment today’s advanced driver-assistance systems with deep learning systems that will learn the behavior of drivers over time.”


Deep learning lets machines program themselves to classify objects. In automotive systems, deep learning can be used to create a database of images and actions to be taken given their position in relation to the vehicle.   Read more   Hmmm  ..Certainly one of the approaches we're researching.  See also: Towards Deep Neural Network Architectures Robust to Adversarial Examples Alain


Half-baked stuff that probably doesn't deserve your time:


Auto Analyst: The Remainder of My Career Will Be Focused on This One Chart


"...Morgan Stanley Analyst Adam Jonas says this chart is so important that you need to put it on your wall. .."

Read More  Hmmm... The Human-Autonomous vs. Owned-Shared chart is a good way to imagine the evolution of from where we are now to where we might end up; however, I disagree with Jonas' characterization of this 2-D  space.

1.  Today: certainly conventional cars are in the lower left hand quadrant; however, we are also in the upper right quadrant with APMs at airports, automated metros in cities  and elevators in all tall buildings. (I grant you that the scope of the autonomy is limited to their own track systems; however, their scope is much larger than the scope that  Uber, Lyft, .. will ever achieve in serving the lower right quadrant)

2.  The extent to which Tesla, ... will ever cover the higher portion (driverless) of the upper left hand quadrant is very questionable because, there is essentially zero consumer demand for driverless with substantial scope, I claim.

3.  To go beyond today in the upper right hand quadrant requires driverless with high scope; neither of which can evolve from the lower right or the upper-left quadrants any better than directly from the lower left.  Achievement of the upper right requires not only the technology (wide scope driverless) but also the economic advantage of elimination of the labor cost (properly stated in the Morgan-Stanley report but,  which Uber, etc. can't demonstrate, even if they prey on the most disadvantaged in our society).  Alain


The Connected Car Faces Its First Class Action Lawsuit


Doug Newcomb 3/31/15 "Connected car technology faces no shortages of potential roadblocks, including industry standardization, data management, consumer acceptance and government regulation, just to name a few. Central among these are legal issues confronting not only automakers currently building and selling connected cars, but all stakeholders – automotive suppliers, technology providers and also government officials – involved and that view significant benefits to connecting cars to each other and to traffic infrastructure to reduce traffic, fuel consumption, emissions and, most importantly, the number of lives lost in car accidents...."  Read more


  Researchers Hack Into Driverless Car System, Take Control of Vehicle


May 2015 Allyson Versprille "A group of Virginia-based researchers funded by the Defense Department found that it is relatively easy to remotely hack into a driverless car’s control system, but said hardening against such an attack can be inexpensive...."  Read more Hmmm... I can also break a display window at Tiffany's and grab the diamonds or drop a bowling ball from an overpass and hit a passing car below but it doesn't mean that I'm going to do it.  "The only thing we have to fear is fear itself." FDR. Anyway... at least it is suggested that "..,hardening against such an attack can be inexpensive." And what about the date of this post being "May 2015"...Credibility?? Alain


NASA made an autonomous car too, and it makes Google’s look dull


Apr 13, 2015 "Today the space agency decided to show the fruits of its labor, however, posting a video on its YouTube account of the finished product. It is called the Modular Robotic Vehicle, MRV for short, and it can -- among other things -- drive itself when needed...."  Read more  Hmmmm... well maybe? They've done the remote control actuator part, but have they done the environmental sensor and intelligence part?  Easy part done; hard part yet to tackle? Alain


  Driverless cars, robot surgeons drive Nokia-Alcatel merger


Matt Campbell 4/16/15 "...Nokia's $16.6-billion takeover of Alcatel-Lucent will create a leader in building networks for today's smartphones. The companies are betting they can also get an edge connecting millions of intelligent machines that haven't been invented yet...." Read more Hmmm... Is this another bad call by Nokia?  Driverless cars will allow the Nokia-Alcatel bandwidth to be consumed, but it will do precious little to enable driverless mobility.  Alain



C'mon Man!  (These folks didn't get/read the memo)


Hmmm... C'Mon Man!  Alain


Calendar of Upcoming Events





April 22, 15:00 GMT

What to plan for when planning for Automated Transit Networks

discussed will be the key aspects of assessing whether advanced transit systems such as personal rapid transit (PRT) and group rapid transit (GRT) are a fit with the application’s requirements and characteristics. Featured expert speakers represent ARUP, Lea+Elliott and PRT Consulting.


Call for Papers



November 4-9, 2015


Call for Papers









Prof. Alain L. Kornhauser

Princeton Autonomous Vehicle Engineering (PAVE)

Princeton University





Heathrow Operational PodCar


Only Steering, Inside

Audi Jam-Assist Demo

Pedestrian Recognition

Lane-Keeping w Brakes

Masdar PRT

Mining Application

Economist Cover Story