State of the Art &

Near Term Future of

Smart Driving Carsby

Alain L. Kornhauser, Ph.D.Professor, Operations Research & Financial Engineering

Director, Program in Transportation Faculty Chair, PAVE (Princeton Autonomous Vehicle Engineering)

Princeton University

Presented at

NJ DoTTrenton, NJJune 18, 2013

Where Have We Been?

The Automobile’s 1st 125 Years

(1886-2011)

Benz patent 1886 1st Automobile Benz Circa 2011

Delivered: Enormous Personal Freedom & Mobility

But…Safe Operation Requires Continuous Vigilance

We Love the Freedom & MobilityBut…Continuous Vigilance is an unrealistic requirement for drivers

Txtng while driving is out of control…

TravelTainment Industry Wants Everyone’sAttention

In In 717 out of 723 accidents ((99%)

http://orfe.princeton.edu/~alaink/SmartDrivingCars/NHTSA_Hendricks2001_UnsafeDrivingActs.pdf

“In 717 out of 723 crashes (99%), a driver behavioral error caused or contributed to the crash”

Early Estimate of Motor Vehicle Traffic Fatalities in 2012

Response is Laudable

Kirkland, WA

But… Not Likely to be Effective

What About Automation?

Preliminary Statement of Policy Concerning Automated Vehicles

Extending its vehicle safety standards from Crash Mitigation to Crash Avoidance with Aim at Full Self-Driving Automation

Level 0 – No-Automation. The driver is in complete and sole control of the primary vehicle controls (brake, steering, throttle, and motive power) at all times, and is solely responsible for monitoring the roadway and for safe operation of all vehicle controls. Vehicles that have certain driver support/convenience systems but do not have control authority over steering, braking, or throttle would still be considered “level 0” vehicles. Examples include systems that provide only warnings

• Level 1 – Function-specific Automation: Automation at this level involves one or more specific control functions; if multiple functions are automated, they operate independently from each other. The driver has overall control, and is solely responsible for safe operation, the driver. The vehicle’s automated system may assist or augment the driver in operating one of the primary controls – either steering or braking/throttle controls (but not both).

• Level 2 - Combined Function Automation: Automation of at least two primary control functions designed to work in unison to relieve the driver of control of those functions. Vehicles at this level of automation can utilize shared authority when the driver cedes active primary control in certain limited driving situations. The driver is still responsible for monitoring the roadway and safe operation and is expected to be available for control at all times and on short notice.

• Level 3 - Limited Self-Driving Automation: Vehicles at this level of automation enable the driver to cede full control of all safety-critical functions under certain traffic or environmental conditions.

• Level 4 - Full Self-Driving Automation: The vehicle is designed to perform all safety-critical driving functions and monitor roadway conditions for an .

Preliminary Statement of Policy Concerning Automated Vehicles

The Aim to Full Self-Driving is Laudable;

But…, Ironically, it may beoverly ambitious and potentially counter productive

Eternal Vigilance should not be the price of Freedom derived from the Automobile

We Like to Drive and Can Be Vigilant

But… Preferably only when we want to; Otherwise, Let us “TXT”

By Focusing on Level 3 (Limited Self-Driving, aka SmartDrivingCar)

NHTSA Can Capture “All” Safety Benefits & Make the Car Even More Desirable

The Jump to Level 4 (Full Self-Driving) Delivers Broad Societal Benefits Equal Mobility for All (young, old, handicapped, disadvantaged), Elimination of Congestion, Halving of Energy, Pollution

These are NOT NHTSA’s Mission & Shouldn’t be Distracted by that Goal

Where Are We Now?R&D

“Level 3 Semi Self-Driving Automation”

http://www.youtube.com/watch?v=cdgQpa1pUUE

Has drive ~ 500,000 miles with “Level 3: Limited Self-Driving Automation”

But…Hardware too expensive and Reliance on 3D Google Maps is “non-elegant”

Where Are We Now?

“Level 4 Full Self-Driving Automation”Operational in Exclusive Environments

Now exist in essentially every Major Airport

APMAutomated People Movers

Milan

Beijing

Paris

and a growing number of Driverless Metros

Personal Rapid Transit: Morgantown (1975 - )

Remains a critical mobility system & expansion being plannedToday…

> 25M Driverless VMT Zero serious accidents

And Today…• Masdar & Heathrow are operational

Video

> 1M Driverless VMT Zero accidents

• Autonomous Buses at La Rochelle (CyberCars/Cybus/INRIA) http://www.youtube.com/watch?v=72-PlSFwP5Y

– Simple virtual non-exclusive roadway • Virtual vehicle-based longitudinal (collision avoidance) and

lateral (lane keeping) systems

This is actually “Level 4 Full Self-Driving Automation”

Very Slow Speed (~ 10 mph); “Limited Pedestrian Environment”

– Driverless Trucks in Australian & Chilean Mines

http://www.youtube.com/watch?v=s0RCSX95QmE

Where Are We Now?Operational Bus Transit

“Level 0+ No Automation, Driver Assistance ”

Bus 2.0: Initial Demonstration of Transit-based Driver Assistance (Level 0+ , no control, only warning)

Fleet of 10 Gillig low floor busesMorning and evening express services22 mile (one-way) travel distanceReliable travel times in all weather and traffic conditions

Bus 2.0: Transit-based Driver AssistanceHow Do They Do It?

Bus 2.0: Transit-based Driver AssistanceHow Do They Do It?

It Just Works!

Where Are We Now?Available in ShowRooms for Consumers

“Level 2- Combined Automation wit Constant Vig1lance ”

Intelligent Drive (active steering )BAS-Plus Active Lane-Keeping Assist (braking not steering )

Volvo Truck Emergency braking

DOT HS 810 767 Pre-Crash Scenario Typology for Crash Avoidance Research

More on Google: Levandowski Presentation

With Mercedes the Market Leader in “Level 2-” and an incremental price tag that can be absorbed by a Price Leading Insurance Company, then other automakers will be enticed to follow which should lead to:

• Viral adoption by the car buying public

• “Moore’s Law type of price/performance improvement

• Market-driven Transition to “Level 2” and “Level 3” at same or even lower price structure

• Adoption and enhancement rates that are comparable to that enjoyed by airbags (With likely a comparable hick-up)

What’s Near-term for Transit?With Mercedes the Market Leader in “Level 2-” and an incremental price tag that can be absorbed by a Price Leading Insurance Company, then other automakers will be enticed to follow which should lead to:

• Fact: For over 40 years New Jersey has had the World’s Best “Bus Rapid Transit” System!

• It Consists of:• Efficient Boarding/Alighting @ Port Authority Bus Terminal

– 223 Departure Gates– Readily Accommodates 700 Buses/hr

The World’s Best Bus Rapid Transit System

• Fact: For over 40 years New Jersey has had the World’s Best “Bus Rapid Transit” System!

• It Consists of:• Efficient Boarding/Alighting @ PA Bus Terminal• Direct Access/Egress to Exclusive Lanes in the Lincoln

Tunnel• 3+ HOV Lanes on the NJ Turnpike that are, by default, essentially bus-only • Many Strategically Located Park&Ride Lots

The World’s Best Bus Rapid Transit System

• Pieces are Connected by:• “495-viaduct” Counter-flow Exclusive Bus Lane

(XBL)– Lane Segregation is by Removable Plastic Peg– Yet exceedingly Safe

» 3 (?) accidents in 41 years, no fatalities.

The World’s Best Bus Rapid Transit System

A Perfect Storm Opportunity• PABT in desperate need of “rehabilitation”• XBL at capacity • “Helix” due for “rehabilitation”• Desperately need: Increased late afternoon in-

bound capacity for busses• New bus procurement cycle begins in 2 years• Test facility available @ Ft. Monmouth

• add Intelligent Cruise Control with Lane Assist to the 3,000 buses…

• e.g. Daimler Benz Distronic Plus with Traffic Jam Assist• even at an incremental $75,000/bus this is just $200M

• Could achieve sustained 3.0 second headways– Increases practical throughput by 50%

• from 700 -> 1,000 buses/hr; 35,000 -> 50,000 pax/hr

– Increased passenger capacity comparable to what would have been provided by $10B ARC rail tunnel

• Institutionally manageable:– All Express Buses are acquired according to NJT Specs.– Facilities (XBL, LT, PABT) are controlled by PANY&NJT

• Ideal test facility available: – Ft. Monmouth

Improving The World’s Best Bus Rapid Transit System

• Concept Not New:

• Concept Makes Even More Sense Now!

Improving The World’s Best Bus Rapid Transit System

Near-term Opportunity for a Substantive Extension of Autonomous Transit

• Specific: General Mobility for Fort Monmouth Redevelopment– Currently: Decommissioned Ft. Monmouth is vacant .

• Ft. Monmouth Economic Revitalization Authority (FMERA) is redeveloping the 3 sq. mile “city”• Focus is on attracting high-tech industry• The “Fort” needs a mobility system.• FMEDA is receptive to incorporating an innovative mobility system• Because it is being redeveloped as a “new town” it can accommodate itself to be an ideal site for testing

more advanced driverless systems.

Where might We End Up?

Most every day…

• Almost 9 Million NJ residents • 0.25 Million of out of state commuters• Make 30+ Million trips • Throughout the 8,700 sq miles of NJ• Where/when do they start?• Where do they go? • Does anyone know???

– I certainly don’t• Not to sufficient precision for credible analysis

• I’ve harvested one of the largest troves of GPS tracks – Literally billions of individual trips, – Unfortunately, they are spread throughout the western

world, throughout the last decade. – Consequently, I have only a very small ad hoc sample of

what happens in NJ on a typical day.

I’ve Tried…

ORF 467 Fall 2012 41

• Motivation – Publicly available data do not contain:– Spatial precision

• Where are people leaving from?• Where are people going?

– Temporal precision• At what time are they travelling?

Trip Synthesizer

Project Overview

Why do I want to know every trip?

• Academic Curiosity• If offered an alternative, which ones would likely “buy

it” and what are the implications.• More specifically:

– If an alternative transport system were available, which trips would be diverted to it and what operational requirements would those trip impose on the new system?

• In the end… – a transport system serves individual decision makers. It’s

patronage is an ensemble of individuals, – I would prefer analyzing each individual trip patronage

opportunity.

Synthesize from publically available data:

• “every” NJ Traveler on a typical day NJ_Resident file

– Containing appropriate demographic and spatial characteristics that reflect trip making

• “every” trip that each Traveler is likely to make on a typical day. NJ_PersonTrip file

– Containing appropriate spatial and temporal characteristics for each trip

Creating the NJ_Resident file

for “every” NJ Traveler on a typical dayNJ_Resident file

Start with Publically available data:

2010 Population census @Block Level– 8,791,894 individuals distributed 118,654 Blocks.

County Population Census Blocks Median Pop/ Block Average Pop/BlockATL 274,549 5,941 26 46BER 905,116 11,171 58 81BUR 448,734 7,097 41 63CAM 513,657 7,707 47 67CAP 97,265 3,610 15 27CUM 156,898 2,733 34 57ESS 783,969 6,820 77 115GLO 288,288 4,567 40 63HUD 634,266 3,031 176 209HUN 128,349 2,277 31 56MER 366,513 4,611 51 79MID 809,858 9,845 50 82MON 630,380 10,067 39 63MOR 492,276 6,543 45 75OCE 576,567 10,457 31 55PAS 501,226 4,966 65 101SAL 66,083 1,665 26 40SOM 323,444 3,836 51 84SUS 149,265 2,998 28 50UNI 536,499 6,139 61 87WAR 108,692 2,573 23 42Total 8,791,894 118,654 74.1

Publically available data:

• Distributions of Demographic Characteristics– Age– Gender– Household size– Name (Last, First)

Ages (varying linearly over interval): input: output:[0,49] 67.5% 67.5%[50,64] 18.0% 17.9%[65,79] 12.0% 12.1%[80,100] 2.5% 2.5%

Gender: Input: Output:female 51.3% 51.3%

Household: Size: Probability: cdf: Expectation:couple 2 0.30 0.300 0.6couple + 1 3 0.08 0.380 0.24couple + 2 4 0.06 0.440 0.24couple + 3 5 0.04 0.480 0.2couple + 4 6 0.04 0.520 0.24couple + grandparent: 3 0.01 0.525 0.015single woman 1 0.16 0.685 0.16single mom + 1 2 0.07 0.755 0.14single mom + 2 3 0.05 0.805 0.15single mom + 3 4 0.03 0.835 0.12single mom + 4 5 0.03 0.865 0.15single man 1 0.12 0.985 0.12single dad + 1 2 0.01 0.990 0.01single dad + 2 3 0.005 0.995 0.015single dad + 3 4 0.005 1.000 0.02 2.42

Final NJ_Resident fileHome CountyPerson IndexHousehold IndexFull NameAgeGenderWorker Type IndexWorker Type StringHome lat, lonWork or School lat,lonWork CountyWork or School IndexNAICS codeWork or School start/end time

ATL 274,549BER 905,116BUR 448,734

CAM 513,657CAP 97,265

CUM 156,898ESS 783,969

GLO 288,288HUD 634,266HUN 128,349MER 366,513MID 809,858

MON 630,380MOR 492,276OCE 576,567PAS 501,226SAL 66,083

SOM 323,444SUS 149,265UNI 536,499

WAR 108,692NYC 86,418PHL 18,586BUC 99,865SOU 13,772NOR 5,046WES 6,531ROC 32,737

Total: 9,054,849

Assigning a Daily Activity (Trip) Tour to Each Person

NJ_PersonTrip file

• 9,054,849 records– One for each person in

NJ_Resident file• Specifying 32,862,668 Daily

Person Trips– Each characterized by a precise

• Origination, Destination and Departure Time

All TripsHome County

Trips TripMiles AverageTM# Miles Miles

ATL 936,585 27,723,931 29.6BER 3,075,434 40,006,145 13.0BUC 250,006 9,725,080 38.9BUR 1,525,713 37,274,682 24.4CAM 1,746,906 27,523,679 15.8CAP 333,690 11,026,874 33.0CUM 532,897 18,766,986 35.2ESS 2,663,517 29,307,439 11.0GLO 980,302 23,790,798 24.3HUD 2,153,677 18,580,585 8.6HUN 437,598 13,044,440 29.8MER 1,248,183 22,410,297 18.0MID 2,753,142 47,579,551 17.3MON 2,144,477 50,862,651 23.7MOR 1,677,161 33,746,360 20.1NOR 12,534 900,434 71.8NYC 215,915 4,131,764 19.1OCE 1,964,014 63,174,466 32.2PAS 1,704,184 22,641,201 13.3PHL 46,468 1,367,405 29.4ROC 81,740 2,163,311 26.5SAL 225,725 8,239,593 36.5SOM 1,099,927 21,799,647 19.8SOU 34,493 2,468,016 71.6SUS 508,674 16,572,792 32.6UNI 1,824,093 21,860,031 12.0WAR 371,169 13,012,489 35.1WES 16,304 477,950 29.3

Total 32,862,668 590,178,597 19.3

ORF 467 Fall 2012 50

Overview of Data Production

1. Generate population2. Assign work places3. Assign schools4. Assign tours / activity patterns5. Assign other trips6. Assign arrival / departure times

Project Overview

“Pixelated” New Jersey(“1/2 mile square; 0.25mi2)

aTaxi Concept• During peak demand aTaxis

operate between aTaxiStands• Autonomous vehicles wait

for walk-up customers • Located in “center” of each

pixel (max ¼ mile walk)• Departure is Delayed to

facilitate ride-sharing• Focus of Analysis:

• what is the ride-share potential?

• Ridesharing delivers:• Congestion relief• Energy savings• Reduced costs/passenger• Environmental sustainability

Warren CountyPopulation: 108,692

New Jersey Summary Data Item Value

Area (mi2) 8,061

# of Pixels Generating at Least One O_Trip 21,643

Area of Pixels (mi2) 5,411

% of Open Space 32.9%

# of Pixels Generating 95% of O_Trips 9,519

# of Pixels Generating 50% of O_Trips 1,310

# of Intra-Pixel Trips 447,102

# of O_Walk Trips 1,943,803

# of All O_Trips 32,862,668

Avg. All O_TripLength (miles) 19.6

# of O_aTaxi Trips 30,471,763

Avg. O_aTaxiTripLength (miles) 20.7

Median O_aTaxiTripLength (miles) 12.5

95% O_aTaxiTripLength (miles) 38.0

Thank Youalaink@princeton.edu

www.SmartDrivingCar.com

Discussion!

http://www.youtube.com/watch?v=MZ3s_cdk_yE&feature=player_embedded

http://www.youtube.com/watch?v=0D0ZN2tPihQ&feature=player_embedded

http://orfe.princeton.edu/~alaink/SmartDrivingCars/Videos/VolvoPlatooningConcept.wmv

http://www.youtube.com/watch?v=ridS396W2BY&feature=player_detailpage

Assorted Videos of Self-Driving Carshttp://orfe.princeton.edu/~alaink/SmartDrivingCars/Videos/1_FrozenLakeVID_onlySteeringWoIndividualWheelBraking.mp4

http://orfe.princeton.edu/~alaink/SmartDrivingCars/Videos/2_FrozenLakeVID_onlySteeringWoIndividualWheelBraking.mp4

http://orfe.princeton.edu/~alaink/SmartDrivingCars/Videos/3_FrozenLakeVID_onlySteeringWoIndividualWheelBraking.mp4

http://orfe.princeton.edu/~alaink/SmartDrivingCars/Videos/4_FrozenLakeVID_onlySteeringWoIndividualWheelBraking.mp4