Intelligent vehicles and road transportation systems (ITS) · Brainstorming session prior lecturing...

11-juin-15 D Gingras – ME470 IV course CalPoly Week 1 1

Intelligent vehiclesand road transportation systems (ITS)

Week 1 : Introduction, context and applications

ME470

Denis GingrasWinter 2015

11-juin-152

You are courageous taking this challenging course because:This course is brand new: the course is currently being built. This is version 1.0. So several bugs may happen.

The topic is still in R&D mode: there is no undergrad textbooks available, the topic is very multidisciplinary and fairly complex requiring a broad background, hence difficult to teach and to learn.

The lecturer (me) is new to CalPoly: I just arrived to SLO, never taught to CalPolybefore, not even in the US. I will be in a steep learning curve at the beginning.

The lecturer (me) is not English speaking native: I am French speaking native. So please indulge my accent and speak clearly in class. It will be appreciated.

I thank you in advance for your patience and understanding. We are all in the same learning boat!

Opening remarks

D Gingras – ME470 IV course CalPoly Week 1

Course objectives The main objective of this course is to provide a solid introduction to ITS in general and intelligent vehicles (VI) in particular.

Provide the student with an introductory overview on intelligent vehicle technologies and vehicular engineering.

After the course, the student has a fairly good knowledge of the overall tools used in intelligent vehicle engineering and how and where to use them.

The student will be able to analyze some basic IV problems, select models and tools to solve it and apply the solution to a particular situation.

The student will gain sufficient autonomy to pursue future in-depth or specialized studies in particular sub-topics.

11-juin-153 D Gingras – ME470 IV course CalPoly Week 1

The course is divided in a set of ten sub-topics covering the main aspectsof intelligent vehicles. The course is using a top-down approach allowingthe student to gain sufficient autonomy to pursue on their own, during orafter the course, in-depth studies in sub-topics they are interested in.

The course being strongly multidisciplinary in nature, the material iscovered using a toolbox approach, where each week focuses on a specificset of tools and knowledge blocs covering a particular sub-topics.

Being introductory in nature, simple problems-solving situations based onsimplified assumptions will be used. In a given sub-topic, emphasized isput on how to state typical problems and how to select a proper approachto solve it, having overviewed the set of tools available.

Strong and active student participation is emphasized. Understanding is favored over memorization. Examples in the course uses case-studies of real systems and applications. Brainstorming session prior lecturing will allow a more active participation

of the students and ease the learning process.

Approach emphasized


The course spans over 10 weeks.

Program for each week:

Part 1: Monday PM 4:10 to 5:30

Open questions and introductory discussion (brainstorming), 20 mins. (all)Lecture, 60 mins. (Prof)Wrap-up discussion and assignments for part two, 10 mins (all)

Part 2: Wednesday PM 4:10 to 5:30, local 022 - 0315

Summary of part one covered material, 10 mins (all)Seminar no 1, 20 mins (identified student)Discussion on seminar no 1, 10 mins (all)Seminar no 2, 20 mins (identified student)Discussion on seminar no 2, 10 mins (all)Wrap-up on the topic of the week, 10 mins (all)

Course structure


Logistics

Availability for consultation (room ….)

Monday afternoon: 1:30 to 3:30 PMWednesday afternoon 1:30 to 3:30 PM

To reach me:

Location: my office (Module 92M A-110 in the parking C7)Email: [email protected]

To know more about me:Université de Sherbrooke web site: www.usherbrooke.caAUTO21; www.auto21.caLIV (Laboratory on Intelligent Vehicles at UdeS) web site:www.gel.usherbrooke.ca/LIVPersonal web site: www.denis-gingras.com


Seminar (20 points).

Each student will prepare one 20 min seminar based on particular aspects related to the main topic of a given week. To build the seminar, the student will be provided with suggested readings. However the student is free to use any material he considered appropriate. Sources and references MUST be indicated. The seminar is followed by a 10 mins Q&A and discussion where the student must be prepared to answer the questions of his/her classmates. Seminars material must be build with Powerpoint. Demos, if any, should run on Matlab/Simulink. After the seminar, all material shall be transmitted to the lecturer in a zip file clearly identified with the student’s name.

One 20-min seminar 20%One 15-page essay 20%Semester project 30%Final exam 30%

Evaluation


One 15-page essay (20 points)

The student must complete one 15-page essay during the semester. The topic of the essay is selected from a list by the student at the beginning of the course and should cover one or several aspects correspond to one of the 10 subtopics of the course. Student copies due March 6 2015 must be clearly identified, clean and well written. Reference sources should be clearly identified and complete.

Evaluation

Final exam (30%)

2 hours individual exams. No books, no notes, unless indicated otherwise. Calculators are allowed. No web access, no handheld devices, no lap-tops.

Semester project (30%)

The students will team-up in group of three. The project is to complete a Matlabdemo in multi-sensor data fusion for vehicle positioning and navigation. Each student team will submit a 20-page report and their running demo, due March 6 2015. See the project guide on the course web page for details.


The weekly written quiz is used to help you reflect on the weekly topic prior the lectures test out your prior knowledge and vision on that particular topic.

The weekly written quiz need to be completed prior the first lecture of each week. Each week, at the beginning of the first lecture, we will brainstorm together in class on these

questions and compile your answer elements. Therefore I ask you to send me your answers prior the first lecture of each week.

Please note that the weekly written quiz are not evaluated. No big stress involved! The weekly written quiz along with the brainstorming session makes the learning process

more dynamic to the student. The second purpose of those weekly written quiz is to give the lecturer an idea on your

knowledge background and better steer the course accordingly. It is therefore important that you “play along” to make the course more interesting and

enlightening to all and for your own benefit. A quiz by nature, is a relatively rapid and spontaneous test. Your answers should be based on

your actual knowledge and not the result of an extensive research. Avoid non informative trivial answers like “I don’t know” or “I have no idea”. There is always

a bit we know about something. Take the time to read carefully each question and think about it (put your brain to work!).

On the other hand, if you spend more than 10-15 minutes per question, or if you google to get your answer, this is no longer a quiz…

The weekly written quiz


11-juin-1510

Undergraduate course in probability and statistics Undergraduate course in linear algebra Undergraduate course in calculus Undergraduate course in linear systems Undergraduate course in measurement and instrumentation

The students must have basic knowledge in Matlab/Simulink and Newtonian mechanics.

Knowledge in control theory, signal processing, or data analysis, MEMs, dynamical systems, sensors and actuators would be an asset

Prerequisites

D Gingras – CalPoly ME470 IV course description

11-juin-1511

Course outline


Week 1 : Introduction to intelligent vehicles, context, applications and motivations

Week 2 : Vehicle dynamics and vehicle modelling Week 3: Positioning and navigation systems and sensors Week4: Vehicular perception and map building Week 5 : Multi-sensor data fusion techniques Week 6 : Object detection, recognition and tracking Week 7: ADAS systems and vehicular control Week 8 : VANETS and connected vehicles Week 9 : Multi-vehicular scenarios and collaborative architectures Week 10 : The future: toward autonomous vehicles and automated driving

(Final exam)

11-juin-1512

Welcome and opening remarks Course outline, methodology and other logistic issues

(see document on course description) Brainstorming: Open questions and introductory discussion History of road transportation and automotive context Motivations for intelligent vehicles and applications History of intelligent vehicles R&D Basic features and technologies of intelligent vehicles Basics on traffic modeling and analysis

Week 1 outline


11-juin-1513

Why have you taken this course?

BrainstormingOpen questions and introductory discussion


Brainstorming

11-juin-1514

Why is road transportation important to you and society?



Brainstorming

11-juin-1515

What do like and dislike about current cars ?



Brainstorming

11-juin-1516


What are the three main components of a road transportation system (automotive)?

Brainstorming


11-juin-1517

What is the main source of road accidents ? Give some examples.


Brainstorming


11-juin-1518

What are the main variables affecting or influencing safety ?


Brainstorming


11-juin-1519

Give some examples of current safety systems in cars.


Brainstorming


11-juin-1520

Why making vehicles more intelligent?



Brainstorming

11-juin-1521

Define “intelligence” from an engineering point of view?



Brainstorming

11-juin-1522

Define the following words:



ImaginingThinkingReasoningKnowledgeMeaningPerceptionPlanningCognitionAttentionAwarenessConsciousness

Brainstorming


What are the main capabilities/features of an “intelligent” vehicle?


Brainstorming

11-juin-1524

What are the main constraints/challenges to mass produce and sell intelligent vehicles ?



Brainstorming

4000 BC- 3500 BC – invention of wheels on carts Mesopotamia 2000 BC - horses domesticated Roads (Trails) to satisfy pedestrians and horses Big civilizations

Egypt: 3000 years B.C. Babylon, Greece, Creta: 1000 years B.C. Rome: 1000 years 600 BC to 400 AD.

Via Appia (312 BC) structure of 1 m to 1,50 m wide

1790 – bicycle 1862 – automobile 1867 – motorcycle 1908 – assembly line (Ford)

History of road transportation History

Roman Via AppiaD Gingras – ME470 IV course CalPoly Week 1 11-juin-1525

The automobile as we know it was not invented in a single day by a single inventor.

It is estimated that over 100,000 patents created the modern automobile.

First theoretical plans for a motor vehicle have been drawn up by both Leonardo da Vinci and Isaac Newton.



First automobile at 2mph !The horse had been humanity's primary form of transportation for more than two millennia…it seemed ridiculousthat anything would ever replace horses.



Etymology: origin of the word automobile"The new mechanical wagon with the awful name automobile has come to stay..."

New York Times (1897)

The credit for the name automobile goes to a 14th Century Italian painter and engineer named Martini. Martini did draw plans for a man-powered carriage with four wheels. Automobile comes from the Greek word, "auto" (meaning self) and the Latin word, "mobils" (meaning moving).

The word car is derived from Celtic word "carrus," (meaning cart or wagon).

Other names for motor vehicles used in patent applications: • Oliver Evans applied for a U.S. patent in Philadelphia in 1792 for a

"oruktor amphibolos" • George Selden received a patent for a "road machine" in 1879. • The Duryea brothers patented their "motor wagons" in 1895. • Henry Ford called his 1896 car a “quadricycle."

Other early media references to motor vehicles included names such as: autobaine, autokenetic, autometon, automotor horse, buggyaut, diamote, horseless carriage, mocole, motor carriage, motorig, motor-vique, and the oleo locomotive.



Old engraving depicting the 1771 crash of Nicolas Joseph Cugnot's steam-powered car into a stone wall (first motor vehicle accident ever).

2.5 mph on only three wheels. The vehicle had to stop every ten to fifteen minutes to build up steam power

Nicolas Joseph Cugnot: the first inventor



1680 - Dutch physicist, Christian Huygens, designed (but never built) an internal combustion engine that was to be fuelled with gunpowder.

1807 - Francois Isaac de Rivaz of Switzerland invented an internal combustion engine that used a mixture of hydrogen and oxygen for fuel. Rivaz designed a car for his engine -the first internal combustion powered automobile. However, his was a very unsuccessful design.

1824 - English engineer, Samuel Brown, adapted an old Newcomen steam engine to burn gas, and he used it to briefly power a vehicle up Shooter's Hill in London.

1858 - Belgian-born engineer, Jean Joseph Étienne Lenoir, invented and patented (1860) a double-acting, electric spark-ignition internal combustion engine fuelled by coal gas. In 1863, Lenoir attached an improved engine (using petroleum and a primitive carburetor) to a three-wheeled wagon that managed to complete an historic fifty-mile road trip.

1862 - Alphonse Beau de Rochas, a French civil engineer, patented but did not build a four-stroke engine (French patent #52,593, January 16, 1862).

1864 - Austrian engineer, Siegfried Marcus*, built a one-cylinder engine with a crude carburetor, and attached his engine to a cart for a rocky 500-foot drive. Several years later, Marcus designed a vehicle that briefly ran at 10 mph that a few historians have considered as the forerunner of the modern automobile by being the world's first gasoline-powered vehicle.

Early history of the combustion engineHistory of road transportation History


1866 - German engineers, Eugen Langen and Nikolaus August Otto, improved on Lenoir's and de Rochas' designs and invented a more efficient gas engine.

1873 - George Brayton, an American engineer, developed an unsuccessful two-stroke kerosene engine (it used two external pumping cylinders). However, it was considered the first safe and practical oil engine.

1876 - Nikolaus August Otto invented and later patented a successful four-stroke engine, known as the "Otto cycle".

1876 - The first successful two-stroke engine was invented by Sir Dougald Clerk. 1883 - French engineer, Edouard Delamare-Debouteville, built a single-cylinder four-stroke

engine that ran on stove gas. It is not certain if he did indeed build a car, however, Delamare-Debouteville's designs were very advanced for the time - ahead of both Daimler and Benz in some ways at least on paper.

1885 - Gottlieb Daimler invented what is often recognized as the prototype of the modern gas engine - with a vertical cylinder, and with gasoline injected through a carburetor (patented in 1887). Daimler first built a two-wheeled vehicle the "Reitwagen" (Riding Carriage) with this engine and a year later built the world's first four-wheeled motor vehicle.

1886 - On January 29, Karl Benz received the first patent (DRP No. 37435) for a gas-fuelled car. 1889 - Daimler built an improved four-stroke engine with mushroom-shaped valves and two V-slant

cylinders. 1890 - Wilhelm Maybach built the first four-cylinder, four-stroke engine.

Early history of the combustion engineHistory of road transportation History


America's first gasoline-powered commercial car manufacturers were Charlesand Frank Duryea. The brothers were bicycle makers who became interested in gasoline engines and automobiles and built their first motor vehicle in 1893, in Springfield, Ma.By 1896, the Duryea Motor Wagon Company had sold thirteen models of the Duryea, an expensive limousine, which remained in production into the 1920s.

Duryea: the first mass producers of cars - assembly line in parallel.



Henry Ford (1863-1947) invented an improved assembly line (production in series) and installed the first conveyor belt-based assembly line in his car factory in Ford's Highland Park Michigan plant, around 1913-14. The assembly line reduced production costs for cars by reducing assembly time. Ford's famous Model T was assembled in ninety-three minutes. Ford made his first car, called the “quadricycle“, in June 1896.

Success came after he founded the Ford Motor Company in 1903. This was the third car manufacturing company created to produce the cars he designed. He introduced the Model T in 1908 and it was a success. After installing the moving assembly lines in his factory in 1913, Ford became the world's biggest car manufacturer. By 1927, 15 million Model Ts had been manufactured.



1969: Apollo 11 computer 2.048 MHz CPU, Memory 74 kB, RAM 4kB

Exponential Growth in Computer Processing Power (Moore’s law) and Computer Aided Engineering (CAE) Capability

Year Model Size (No. of Elements)

1980 500 - 5000

1985 1,000 – 10,000

1990 3,000 – 20,000

1995 5,000 – 60,000

2000 10,000 – 500,000

Today Larger than 5,000,000

http://www.aprosys.com/img/06.pdf

The rise of computing powerHistory of road transportation History


Over 100 years of electricity/electronics in cars:

1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

High voltage ignition system

Low voltage ignition system

4 V battery

Spark plugs

1 million T-Ford sold in 1915

DC generator

AC generator

Selenium rectifier

First radio in car

3 phase current generator

Transistor controlled ignition coiled

Solid state radio

Closed-loop air-fuel ratio control

Cruise control

Electronic transmission control

ABS

12 V battery

Drive by wireTelematics

Microprocessor and microcontroller

ADAS

ICsTransistor

Airbags

Hybrid

Incr

ease

d co

mpl

exity

Source: BOSCH GmbH





The car (light blue line) is closely following the mobility long term trend.


Historic View on Driver Assistance

• Turn signal reset – eliminates manual reset

• Synchronized manual transmission – eliminates throttle application

• Servo for breaking and steering – reduced force required from the driver

• Centralized door locking (incl. remote control) – enables opening and locking without access to respective door

• Automatic transmission – eliminates manual gear shifting

• Oil level indicator – eliminates dirty hands

• Anti‐lock Breaking System (ABS) – increased stability during braking




• Park Assist– parking without bumps and scratches

• Electronic Stability Program (ESC, ESP, DSC, ...) – improved stability within physical limits

• Brake Assist – reduced braking distance in emergency in braking situations

• Navigational Systems – eliminates map in driver‘s hands

• Climate Control – controls interior temperature to comfortable level

• Night vision – enhances driver‘s perception range in darkness

• Cruise Control – maintains vehicle velocity




• General remarks on the previous list of DAs– none of the previous examples is required for mobility – nevertheless driver assistance is an indispensable part of

vehicle equipment – advanced driver assistance systems also take over primary

driving tasks

• Main success factor of DAs– Relief of inconvenient tasks – Increase of safety – Complement or supplement of human skills


First electrical vehicles in 1880sTaxi in New-York, 1898, with docking station for recharging batteries



1940s Milwaukee Machine Tools ad proclaiming that air-conditioning will be available for automobiles after WWII.

1940s Cannon Electric ad proclaiming the availability of in-car "pagers" after WWII.



Vision of future driving 1958


Source: Science Digest, Electronic Highway of the Future (Apr, 1958)

History


brake assisttraction controlchassis assistgood griphead up displaysgood visibility from driver's seatlow noise level in interiorlegibility of instrumentation, warning symbolsearly warning of severe braking aheadgood chassis balance and handlinganti-lock braking systemelectronic stability controlintelligent speed adaptationcollision warning/avoidance

Automobile safety is the study and practice of design, construction, equipment andregulation to minimize the occurrence and consequences of automobile accidents. Roadtraffic safety more broadly includes roadway design. Active safety is used to refer to technology assisting in the prevention of a crash. Passive safety is used to refer to technology to protect occupants during and/or after a crash.

seat beltsairbagspassenger safety celldeformation zonesloadspace barrier-netslaminated glasscorrectly positioned fuel tanksfuel pump kill switches(automatic) emergency callemergency medical services

Active safety examples Passive safety examples

Automotive context Context



Decomposing a imminent collision scenario

Typically, driver response time is in the order of a second or more, whereas active safety systems can be in the order of tens of ms.


Source: http://www.safecarguide.com/gui/new/neworused.htm, http://www1.eere.energy.gov/vehiclesandfuels/facts/printable_versions/2010_fotw613.html, http://www.sacbee.com/2012/10/09/4894505/average-fuel-economy-for-new-cars.html,http://www.nhtsa.gov/Laws+&+Regulations/CAFE+-+Fuel+Economy/2004+Automotive+Fuel+Economy+Program

Typical specs for a vehicle today

New purchase price: $30,000Engine output: 180 hpFuel economy: 23.3 mpg and 19.8 mpg (truck)Occupancy: 1.95 (car) – 2.35 (van)New-car buyer trade-in: after 4 years at 55,000 milesNew-car leases trade-in: after 3 years at 36,000 milesLife span: just over 13 yearsFinal mileage: 145,000 miles


Annual U.S. traffic volumes - 1982 & 2007

http://jagadees.files.wordpress.com/2008/08/dotapril2.png

***Per stats pulled from the government DOT



Annual U.S. hours delayed while traveling during peak hours

http://ops.fhwa.dot.gov/aboutus/one_pagers/perf_measurement.htm




An automobile needs to be: operational at -40 to +80 (120) deg C operational at 0 to 100% humidity operational at 0 to 3000m altitude shock resistant work without frequent adjustment, preferably self adjusting maintenance free (no pre-flight check…) can be sold at all markets, very different regions w/ varying regulation self explanatory, intuitive (no one reads handbook) 12 years of operation, no maintenance, wear needs to be "obvious" low cost mass producible recyclable non-toxic sell it and never see it again (especially no recall)



distance,speed, number

location, destination,speed, VIN

transmission speed(s),torque, gear, pressure,maintenance data

engine speed, torque,load, temperatures,pressures,mainten-ancedata

nearbyvehicles:

seats, restraint systems status,heating / ventilation, infotainment settings

body motion: yaw rate, lateral /longitudinal / vertical

acceleration

windows,doorsstatus

ambient,pressure,

temperature,lighting, rain

driver steering, brake, throttle,gear selection, turn signal, lights

brakes / steering /vehicle dynamics control / driver assistance

system statusMore than 4200 signals, 10M lines of code, 35% system costSource: Driving Cars Toward Complexity , I. Krueger (UC San Diego), NPR Interview, April 30, 2010

Vehicle complexityAutomotive context Context

Up to 80 controllers1

- Powertrain: ignition, injection, transmission, 4WD…- Safety: airbag, seatbelts, pre-tensioners…- Chassis: steering, brakes, dampers…- Driving Aid:parking aid, night

vision…- Entertainment: MP3,

CD, radio…

- HVAC: air conditioning…- Body: seats, doors, roof…

- Vision: lights, wipers, mirrors…- Information: displays, navigation…

Up to 20 Communications Networks2


- CAN: Powertrain, safety, chassis, driving aid -MOST: Entertainment and information- LIN: Body, vision, HVAC Source 2. “Software-Technologie in der Automobilindustrie”, K. Grimm, Daimler AG,2009

Source: 1. “Driving Cars Toward Complexity”, I. Krueger (UC San Diego), NPR Interview, April 30, 2010

Vehicle complexityAutomotive context Context


The roads are not used efficiently. IVs can improve

Traffic density Flexibility in road segments allocation

Motivation for intelligent vehicles MotivationsContext


Parking would be much more comfortable/safer for this driver if his car was equipped with a parking assist system, a rear-view camera and pedestrian detection. Would they work properly?

Source Roland Berger Insights, Automotive Competence Center Client Magazine Issue 01.2013

Use case of ADAS: Harsh operating conditions: A Canadian example…;-)


Typical driving scene: Where is Charlie?


Source: D Langer, Volkswagen Electronics Research Laboratory, 2012


Typical driving scene analysis


This type of scene analysis must be done in real-time, typically every 50 to 100 ms. (blue: pedestrian, red: moving cars, yellow: non-moving cars, green: lanes)



Current invasive intelligenceBasic technologies of IVs

IVsbasics


To develop a vehicle « counciousness », we need to get information on:

Its own internal states Its immediate surrounding Its remote/extended surrounding Its level of risk in perception/decision/action making

Develop a vehicular « innate survival instinct »

Basic technologies of IVs IVsbasics


2000 2025

Complexityimplementation

difficulty

Systems

External modules Internalmodules Sensors

Driver IDfingerprint

Driver’s vigilence monitoring Oral man-machine dialog

Man- machine interfacesvoice command systems - haptic systems

Active anti-collisionsensors

Enhanced night visionand lanes/pedestrian detection

« Total awareness »

« Smart dust »MEMS, SOC Discrete sensors and actuators

Nano

Recognition – scene analysisEx. road signs

« Networked vehicle »5.9 GHz DSRC

Integrated information systems

Automatic driving systems

Intelligence penetration in cars

Bluetooth, 802.11« wireless »

Distributed network sensors

multivariable adaptivecontrol systems

hierarchical multilevelarchitectures

« Drive by-wire »

Learning faulty-tolerant autonomous

systems

Source: Siemens VDO

Keys, boardand buttons

Side mirrors

Cell phones

PID

IVsbasicsBasic technologies of IVs


2009: …and then came the Google car…

Automotive context ContextIVsbasics


Estimation position and heading of ego-vehicleEstimating and tracking position of other vehiclesDetecting, classifying and positioning obstaclesControl vehicle stability and dynamicsEstimating and controlling braking performanceNavigationCommunication with other vehicles and outside world

Some advanced technical functions in IVs


Infotainment: not safety related.



Main tasks of IVs


IVsbasics

Perception Motion Control

Cognition

Real WorldEnvironment

Localization

PathEnvironment ModelLocal Map

"Position" Global Map

Basic technologies of IVs


An intelligent vehicle consists basically of four fundamental technologies: environment perception and modeling, localization and map building, path planning and decision-making, and motion control.

Source: H. Cheng, Autonomous intelligent vehicles: theory, algorithms, and implementation, Springer, 2011

IVsbasics


Basic architecture of IVs IVsbasics

Source: Vermaas L L G et al., Intelligent Vehicle Survey and Applications, Advances in Technological Applications of Logical and Intelligent Systems, G. Lambert-Torres et al. (Eds.), IOS Press, 2009

VisionPerception

PositioningNavigation

CommunicationCollaboration

Artificial intelligence

ControlActuation




IVsbasics

Challenge: from the tons of data coming out of the sensors, how to extract and compress the useful information to insure reliable real-time reasoning.


Basic technologies in IVs

The AI part: Basic brain model in IVs

IVsbasics

Basic driver model


Source: E. DONGES, “A two-level model of driver steering behavior", Human Factors, vol. 20, No 6, 1978.

Basic technologies in IVs IVsbasics

A very basic vehicular motion control




Interactive road situation analysis framework







IVsbasics

Dead reckoning positioning system

IVs basic architecture



IVsbasics

Another example from VW

Toward integration of automotive control systems:

Software« Intelligence »

Sensors Controllers Actuators

MEMs

Smart sensor Smart actuator




Typical framework of a vehicle control system




Computer systemic view of IV main components

Source:Nan-Ning Zheng, “Toward Intelligent Driver-Assistance and Safety Warning Systems”, IEEE Intelligent systems magazine, 2004.



Source:Nan-Ning Zheng, “Toward Intelligent Driver-Assistance and Safety Warning Systems”, IEEE Intelligent systems magazine, 2004.


Architecture for a driver assistance and safety warning system

DOT’s vision of intelligent vehicles


Source: US DOT NHTSA ACAS Program, final report, 2000



IV apps areas can roughly be divided into three groups depending on the level of support to the driver (Note the DOT IHVS has set 5 levels):

Advisory systems: These systems provide an advisory/warning to the driver. No action is taken by the vehicle. Exemples are collision warning systems, animal warning at night, side object warning (blind spot), and driver impairment monitoring.

Semi-autonomous systems: These systems take partial control of the vehicle, either for driver assistance or for an emergency intervention to prevent a collision These systems often use haptic measures, i.e. based on the sense of touch, to assist the driver. Semi-autonomous systems include functions such as CMBS (Collision Mitigation Braking System), lane-keeping, Adaptive Cruise Control (ACC), parking assist and precise docking.

Fully autonomous systems. This kind of systems take full control of the vehicle Current examples are low speed automated driving (for congested traffic) or platooning on highways.

Levels of automation in intelligent vehicle applications Basic technologies of IVs IVs

basics

Smart sensors and actuators into IVs Actual state-of-the-art: transducers linked together with a

microprocessors through a common bus. Smart sensors are now made of multiple chips. Micromachining and VLSI circuitry is merging. Smart sensors do more than just picking up and sending a

signal, they have embedded algorithms that process and interpret data, communicate and self-calibrate over time.

Integrating bus interface circuitry is also coming , starting at the module level and spreading to the chip level.



MEMS Micro-Electro-Mechanical Systems

Miniature systems for sensing and actuating

Batch fabrication approach Utilizes microelectronics manufacturing

base Common technology for sensors,

actuators, and systems

MEMS manufacturing of automotive sensors began in 1981 with pressure sensors for engine control

Continued in the early 1990s with accelerometers to detect crash events for air bag safety systems

In recent years has further developed with angular-rate inertial sensors for vehicle-stability chassis systems and navigation.




Applications for MEMs in cars


Why using MEMS

Utilizes the economy of batch processing, together with miniaturization and integration of on-chip electronic intelligence

MEMS makes high-performance sensors available for automotive applications, at the same cost as the traditional types of limited-function sensors they replace.

In other words, sensors would have to be several times more expensive than MEMS if they were still made by traditional electromechanical/discrete electronics approaches.



Advanced Driver Assistance Systems

Advanced Driver Assistance Systems, or ADAS, are systems to help the driver in the driving process.

When designed with a safe Human-Machine Interface it should increase car safety and more generally road safety.

It takes over some of the primary driving tasks from the driver.

Examples: Stop-n-go ACC for congested traffic (~2003) Lane keeping on freeways (~2004) Automated throttle, brakes, steering in tedious stop-n-go traffic




Primary Driving Tasks – are required to get from current position to destination – Navigation – Maneuver (e.g. lane change) – Trajectory (e.g. velocity, steering, stabilization)


See e.g.: Wuhong Wang, Fuguo Hou, Huachun Tan & Heiner Bubb (2010): A Framework for Function Allocations in Intelligent Driver Interface Design for Comfort and Safety, Int. Journal of Computational Intelligence Systems, 3:5, 531-‐541



Primary Driving Tasks are required to get from current position to destination Navigation Maneuver (e.g. lane change) Trajectory (e.g. velocity, steering, stabilization)

Secondary Driving Tasks – control operation point of vehicle

(throttle, brake, gears) – turn signal , wiper, light, …

Tertiary Driving Tasks– control ambience – radio, phone




ValueJudgment

Sensor World BehaviorProcessing Model Generation

Sensors Structure Actuators

WorldSensors Structure Actuators

World

SensorProcessing

BehaviorGeneration

Planning

WorldModel

Behavior Coordination


Driving automation requires adetailed understanding of howhuman drivers operate.

Vehicle Communication Ad-hoc Networks: VANETS


Source: R Berger, Automotive insight, Automotive Competence Center Client Magazine, Issue 01.2013


Vehicle Communication Ad-hoc Networks: VANETS




U.S. Federal Communications Commission (1999):“ … services are expected to improve traveler safety, decreasetraffic congestion, facilitate the reduction of air pollution, …”

DSRC:Dedicated Short Range Communications



DSRC:Dedicated Short Range CommunicationsTwo basic types for safety:

Vehicle-to-Vehicle communication (V2V): Information is transmitted between vehicles. It enables vehicles to

know where the vehicles in its vicinity are and what they are doing.Applications include:

Forward Collision Warning Emergency Electronic Brake Light Blind Spot/Lane Change Warning Intersection Movement Assist Do Not Pass Warning Control Loss Warning

Vehicle-to-Infrastructure communication (V2I)Applications include:

Automatic tolling Traffic jam/construction site ahead warning

Systems use absolute positioning and relative positioning. Maps are sent from the infrastructure to the vehicle Positioning based on GPS and dead reckoning



Controller Area Network (CAN)

Bus connecting microcontrollers and devices

Powertrain, chassis, safety, driving aid

Media Oriented Systems Transport (MOST)

Low cost fiber optics for transport of high data volumes

Entertainment and information

Local Interconnect Network (LIN)

Small, slow, and cheap solution to integrate intelligent sensors

Vision, Body, HVAC


Embedded vehicle networks



Embedded vehicle networks

A high level comparison of the characteristics of LIN, CAN, and FlexRay/TTP.Source: P. E. Lanigan et al., Diagnosis in Automotive Systems: A Survey, Carnegie Mellon University, 2011


Powertrain ignition, injection,

transmission, 4WD Safety

airbag, seatbelts, pre-tensioners

Vision lights, wipers, mirrors

Chassis steering, brakes, suspension

Driving aids parking aid, night vision

Entertainment Body

Seats, doors, roof HVAC

air conditioning


Electronic Controller Unit (ECU)

generic automotive computing and communication topology

Source: P. E. Lanigan et al., Diagnosis in Automotive Systems: A Survey, Carnegie Mellon University, 2011


OBD stands for “On-Board Diagnostics.” It is a computer-based system originally designed to reduce emissions by monitoring the performance of major engine components.

On-Board Diagnostics: OBD-II:Basic technologies in IVs IVs

basics

A basic OBD system consists of an ECU (Electronic Control Unit), which uses input from various sensors (e.g., oxygen sensors) to control the actuators (e.g., fuel injectors) to get the desired performance. The “Check Engine” light, also known as the MIL (Malfunction Indicator Light), provides an early warning of malfunctions to the vehicle owner. A modern vehicle can support hundreds of parameters, which can be accessed via the DLC (Diagnostic Link Connector) using a device called a scan tool.



Electronic Stability Control

1. ESP-hydraulic unit with integrated ECU2. Wheel speed sensors3. Steering angle sensor

4. Yaw rate and acceleration sensor5. ECU for engine management

Traffic modeling and analysis


Let us consider vehicles crossing a certain point on a road. The time that a vehicle n reaches that point is denoted by and the time that the vehicle has completely passed the measurement point, is denoted by . The time headway of vehicle n is calculated by :

0nt 1

nt

0 11 /H n n nT t t s v nv = speed of vehicle n

Traffic modelingand analysis


If N vehicles are counted that cross that point on the road during a time interval ΔT, then the traffic flow is defined as:

NqT

1

1 N

nn

v vN

The arithmetic average speed v is given by,

The traffic density expressed in number of vehicles per km is given by, /q v

Thus to increase traffic flow, we can either increase the vehicle density or increase the average speed. This is not easy to achieve due to safety reasons.

Traffic modeling and analysisTraffic modeling

and analysis


The previous formulas does not display the peculiar feature of traffic flow that the aggregate traffic speed decreases with increasing traffic density. Two functional relations between the traffic flow q, the average speed v, and the vehicle density are illustrated below. This common used flow-density relation is the fundamental diagram. The vehicle speed is maximum at low densities and almost zero at high densities

Source: D. Helbing. Traffic and related self-driven many-particle systems. Reviews of Modern Physics, 73(4):1067–1141, December 2001.


and analysis


The fundamental diagram is usually expressed mathematically by:

1( ) expa

freecritical

q V va

Where is the free flow speed and is the critical density. And a is a parameter model depending on the road infrastructure (ex. number of lanes). NB: The formula above is not unique. Several other models have been proposed to describe traffic behavior, but this model is a common one.

The slope of the fundamental diagram (last slide) starts almost linearly and corresponds to the free-flow speed. In this region the density can increase while the average speed stays the same, thus increasing the traffic flow. With increasing density, the traffic flow increases up to a maximum, i.e. the capacity flow, which is referred to as a critical point. The corresponding density and vehicle speed are the critical density and the critical speed. At higher densities than the critical density, the average vehicle speed is significantly lower than in free-flow traffic.

freev critical


and analysis


Free flow: drivers can drive their desired speed, when traffic is unobstructed. In free-flow, drivers can maintain their desired speed, because their time headway is very large. Car-following mode: when the traffic density increases, drivers will adjust their speed such that they can follow the vehicle directly in front, while maintaining a safe time headway. Congested traffic: the vehicle density has passed a certain critical density such that the traffic flow and vehicle speed decrease significantly. If the vehicle speed drops to almost zero, the density has become too large and a traffic jam will occur.


and analysis

Traffic jams typically move upstream, thus against traffic direction. If more vehicles leave the traffic jam at the downstream front than vehicles entering the traffic jam at the upstream front, then the traffic jam will reduce in length. The outflow of a traffic jam is more or less a fixed quantity. If the fixed quantity is denoted by qL, the traffic jam will thus reduce in width if qinflow < qL, where qinflow is the traffic flow that enters the traffic jam at the upstream front.

Traffic jams:


We have mainly two usual approaches:

1) Reducing traffic inflow: If the inflow of a traffic jam is controlled by speed limitsupstream of the jam, a low-density wave is created that moves downstream. The high-density wave (the traffic jam) merges with the low-density wave created by the speed limits. The high-density and low-density wave can then compensate each other, thus eliminating the traffic jam.

2) Prevent traffic breakdown by keeping the traffic density smaller than the critical density. The traffic inflow must thus be controlled before the capacity reaches its maximum.

Strategies to reduce traffic jams:

Source: K. Nagel, P. Wagner, and R. Woesler. Still flowing: Approaches to traffic flow and traffic jam modeling. Operations Research, 51(5):681–710, 2003.

We will see during the course that collaborative intelligent vehicles can help in controlling traffic flow and avoiding traffic jams by optimizing the distance between vehicles and average vehicle speed using for example a Model Predictive Control (MPC) strategy. MPC is a generic model-based control method that computes control traffic light signals in order to optimize future process behavior .


and analysis


Model Predictive Control

Source: C.E. Garca, D.M. Prett, and M. Morari. Model predictive control: Theory and practice–a survey. Automatica, 25(3):335–348, 1989.

MPC is a generic model-based control method that computes control signals in order to optimize traffic behavior .


and analysis


Road safety situation is assessed by combining traffic rules, vehicle dynamics, and environment prediction. Since the safety distance varies with the speed of a host vehicle, preview time, rather than safety distance response, is often adopted as the measurement of safety. Hence, the safety response time is given by

where dr is the distance required to respond to the nearest object due to driver response delay, dv is the distance to slow down, ds is the safety distance between the host vehicle and obstacles, and v is the velocity of the host vehicle.

r v ss

d d dTv

Traffic modeling and analysis Traffic modelingand analysis

Safety response time


Traffic models

Source: S.P. Hoogendoorn and P.H.L. Bovy. State-of-the-art of vehicular traffic flow modelling. Proceedings of the Institution of Mechanical Engineers, Part I-Journal of Systems and Control Engineering, 215(I4):283–303, 2001.

Traffic models are classified according to the level of detail, i.e. microscopic, mesoscopic, or macroscopic.

Microscopic models describe the characteristics of individual vehicles as well as their interactions. From the individual vehicle characteristics and driver behavior, the acceleration, speed and position can be calculated for each vehicle.

Macroscopic models operate on a more aggregate level and describe traffic without distinguishing individual vehicles. Macroscopic models deal with traffic flow in terms of average densities, average speeds, and average flows.

Mesoscopic models describe traffic flow in medium detail level, and can be situated between microscopic and macroscopic models. In mesoscopic models, individual vehicles are not distinguished as in microscopic models, but the behavior is specified in individual terms. Some types of mesoscopic models are based on gas-kinetic theories. The advantage of gas-kinetic models is that the behavior of individual vehicles can be described, without the need to describe their individual time-space behavior.

Traffic modeling and analysis Traffic modelingand analysis

The very essence of mobility adds degrees of difficulty in implementing intelligence in vehicles

This added intelligence will change radically the end-users behavior and the way the various automotive sectors will evolve (ex. aftermarket sector)

One of the biggest challenge will be to bring this intelligence at a very low cost with a high level of reliability.

Automotive components require the ruggedness of military parts at the price of consumer products.

Concluding remarks Intelligent vehicles to move increasingly and rapidly into

commercial vehicle markets -- “business-centered systems” where the intelligence enhances the bottom line

Public is becoming increasingly comfortable with driver aids and demand more relief from the tedium of driving …creating a strong market for ADAS in passenger cars.


Safety and cost will always be the two top priorities Beware: a careless integration of intelligence or new technologies

in vehicles may lead to lethal consequences ! The added-value of a car will be more and more in its embedded

intelligence at all levels materials-powertrain-sensors-systems. The building of tomorrow’s cars requires strong transdisciplinary

R&D and the convergence of several economic, environmentaland social factors.


Concluding remarks

Bishop R, Intelligent vehicles technology and trends, Artech House, Norwood, 2005

Cheng H., Autonomous intelligent vehicles: theory, algorithms, and implementation, Springer, 2011

Eskandarian A. et al (Ed.), Handbook of Intelligent Vehicles, preface and Chapter 1, pp2 to 12., Springer, 2012.

Ozguener U et al., Autonomous Ground vehicles, Artech House, 2011 Li Li & Dr. Fei-Yue Wang (Ed.), Advanced Motion Control and Sensing for

Intelligent Vehicles, Chap. 1, Springer 2007. Nunes U. et al., «Guest Editorial: Introducing Perception, Planning, and

Navigation for Intelligent Vehicles.” IEEE Transactions on Intelligent Transportation Systems, vol. 10, no. 3, pp. 375-379, September 2009

Siciliano B. et al. (Ed.), Handbook of robotics, Chapter 51, “Intelligent vehicles”, Springer, pp. 1175-1198, 2008

US DOT NHTSA ACAS Program, final report, 2000. Vermaas L L G et al., Intelligent Vehicle Survey and Applications, Advances in

Technological Applications of Logical and Intelligent Systems, G. Lambert-Torres et al. (Eds.), IOS Press, 2009

References


Bureau of Statistics, Quarterly Publications, http://www.bts.gov/publications/national_transportation_statistics/

Catling I (ed) (1993) Advanced technology for road transport: IVHS and ATT. ArtechHouse, Boston

DOT HS (2010) Traffic safety facts research note: summary of statistical findings. Highlights of 2009 motor vehicle crashes, DOT HS 811 363, Washington, DC. http://www-nrd.nhtsa.dot.gov/cats/index.aspx

DOT HS, US DOT NHTSA (2009) Traffic safety facts a compilation of motor vehicle crash data from the fatality analysis reporting system and the general estimates system. DOT HS 811 402, US DOT NHTSA, Washington, DC. http://www-nrd.nhtsa.dot.gov/CATS/

EnergyIndependence. http://www.americanenergyindependence.com/fuels.aspx http://www.merriam-webster.com/dictionary/intelligent Institute, College Station, http://ntl.bts.gov/lib/ jpodocs/repts_te/9063.pdf Little C., The Intelligent Vehicle Initiative: advancing 'Human-Centered' smart vehicles,

Federal Highway Administration, Public Roads, U.S. Department of Transportation, vol. 61, no. 2, pp. 18, 1997.

Intelligent Transportation Society of IEEE, http://ewh.ieee.org/tc/its/ ITS JPO DoT, Intelligent Transportation Systems (ITS) program overview.

http://www.its.dot.gov/its_program/about_its.htm

Additionnal references


McCall J., O. Achler, M. M. Trivedi, P. Fastrez, D. Forster, J. B. Haue, J. Hollan, and E. Boer, “A collaborative approach for human-centered driver assistance systems,” in Proc. IEEE Intell. Trans. Syst. Conf., Oct. 2004, pp. 663–667.

Michon JA (ed) (1993), Generic intelligent driver support. Taylor and Francis, London

National Transportation Statistics, US DOT Research and Innovative Technology Administration,

Swedish Government Health and Welfare Statistical Database (2006) http://192.137.163.40/epcfs/index.asp?kod=engelska

Texas Transportation Institute, Proceedings of a national workshop on IVHS, sponsored by Mobility 2000, Dallas, Texas Transportation.

US Energy Information administration, Use of energy in the United States explained. http://www.eia.gov/energyexplained/index.cfm?page=us_energy_use

Vlacic L. et al., Intelligent vehicle technologies, Society of Automotive Engineers (SAE) international. Butterworth-Heinemann, Boston, 2001

World Health Organization (2009) Global status report on road safety: time for action. World Health Organization, Geneva. www.who.int/violence_injury_prevention/road_safety_status/2009.

Additionnal references


QUESTIONS?


Intelligent vehicles and road transportation systems (ITS) · Brainstorming session prior lecturing...

Documents

Transcript of Intelligent vehicles and road transportation systems (ITS) · Brainstorming session prior lecturing...