Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
64 views

Final

The document describes Google's driverless car project led by Sebastian Thrun. It discusses how the car works by integrating Google Maps with sensors like LIDAR and radar, and artificial intelligence software. The car can steer, accelerate, brake and navigate itself to destinations safely using these systems. Over 140,000 miles have been completed so far with occasional human intervention, demonstrating the technology's promise to revolutionize transportation. Legal and social challenges still need addressing before widespread adoption.

Uploaded by

Anna
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
64 views

Final

The document describes Google's driverless car project led by Sebastian Thrun. It discusses how the car works by integrating Google Maps with sensors like LIDAR and radar, and artificial intelligence software. The car can steer, accelerate, brake and navigate itself to destinations safely using these systems. Over 140,000 miles have been completed so far with occasional human intervention, demonstrating the technology's promise to revolutionize transportation. Legal and social challenges still need addressing before widespread adoption.

Uploaded by

Anna
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 20

GOOGLE DRIVERLESS CAR

Page 1

What Is It?
It is a project led by Sebastian Thrun,Co-inventor of Google Street View and Director of the Stanford Artificial Intelligence Laboratory The Google Driverless Car is like any car, but: It can steer itself while looking out for obstacles It can accelerate itself to the correct speed limit It can stop and go itself based on any traffic condition It can take its passengers anywhere it wants to go safely, legally, and comfortably.

Page 2

How It Works?
General Idea Integrates Google Maps with various hardware sensors and artificial intelligence software
Google Maps Provides the car with road information Hardware Sensors Provides the car with real time environment conditions

Artificial Intelligence Provides the car with real time decisions


Page 3

Google Maps
Google Maps interacts with GPS and acts like a database

Speed Limits
Upcoming intersections Traffic Report Nearby collisions Directions
Page 4

Hardware Sensors
Gives real time environmental properties Environment is dynamic so need real time results Attempt to create fully observable environment

Page 5

LIDAR 'Eyes' of Google Car


The Velodyne LIDAR HDL-64E use an array of either 64 or 32 lasers to electronically "see" the environment, The module is set inside a rotating drum. Its lasers complement Google's mapping software and GPS data, which help orient the car on the road.

The LIDAR provides additional positional data, but also identifies other cars, bicycles, pedestrians, and road hazards.
Page 6

Parts Of LIDAR
Laser: 1550 nm lasers , eye-safe at much higher power .Better target resolution is achieved with shorter pulses.

Scanner and optics Time taken for image development is affected by the speed at which they are scanned.
Photo detector and receiver electronics Two main photo detector technologies are used in LIDAR: solid state photo detectors, such as silicon avalanche photodiodes, or photomultipliers. Sensitivity of the receiver is a parameter that has to be balanced in a LIDAR design.

Position and navigation systems LIDAR require instrumentation to determine the absolute position and orientation of the sensor. So Global Positioning System receiver and an Inertial Measurement Unit (IMU)is added Page 7

How LIDAR Works?


LIDAR Uses optical remote sensing technology Measure the distance and other properties of a target by illuminating the target with light, often using pulses from a laser. Can be used with a wide range of targets. A narrow laser beam can be used to map physical features with very high resolution

Page 8

Video Camera
Mounted near the rearview mirror to detect road signs and traffic lights ,speed limits . It also has computers on board the vehicle to recognize obstacles such as pedestrians and cyclists etc.

Page 9

RADAR
RADAR (Radio Detection And Ranging) A way to detect and study far off targets by transmitting a radio pulse in the direction of the target and observing the reflection of the wave. It is basically radio echo Radar observables: Target range Target angles (azimuth & elevation) Target size (radar cross section) Target speed (Doppler) Target features (imaging)
Page 10

Position Estimator
Position estimation consist of the following process : A priori map of its environment .
A combination of odometery and optical range sensing. An algorithm for matching sensing data to map . An algorithm to estimate precision.

Position Estimator is mounted on left rear wheel

Page 11

Artificial Intelligence
Google Maps and the hardware sensors data are sent to the AI
AI then determines: how fast to accelerate when to slow down/stop when to steer the wheel

which personality to incorporate Cautious Personality Aggressive Personality

Page 12

Relation to AI
Machine Learning The agent learns of its environment based on sensor data and its database Goals

The agent's goal is to take the driver to its desired destination safely and legally
Reasoning

Autonomy
The agent can solve tasks independently and does not need any human interaction

Car starts off with a cautious personality but can determine when to go into an aggressive personality Page 13

As of 2010: 6 Google Driverless Cars 5 Toyota Prius 1 Audi TT Total miles 1,000 miles without human intervention 140,000 miles with occasional human intervention 2 accidents Both human error Nevada 1st state to allow driverless vehicles can be legally operated on public roads

Currently...

Page 14

Advantages
Safer roads with fewer traffic collisions and therefore reduction in road injuries. Reduced journey time as fewer traffic jam. Reduction of physical road signage. Elimination of driving licenses. Alleviation of parking scarcity
Page 15

Disadvantages
Legal issues.

Wont fully eliminate car accidents , but will reduce them. Currently expensive.
Trained professional is needed to fix problems and broken part of robotic cars.
Page 16

Social Impact and Ethics


The job opportunity of a paid driver will be at stake . Machine caused accidents

Page 17

Conclusion
Driverless vehicles will make a great impact on the society. Motor Vehicle laws on every country should be reconstructed. These vehicles will be a great solution to avoid traffic congestion , parking problem etc. It will save time and life.

Researches should be made it 100% accuratePage 18

Reference
(afp), Charlotte Raab. "AFP: Google Brain Drives Cars in Quest for next Auto Revolution." Google. Web. 21 Nov. 2011. <http://www.google.com/hostednews/afp/article/ALeqM5hVGLbxyIGl0hBE0omPclrPNrwHg?docId=CNG.f44dc0982a11a52246a9f2b994a36 455.a1>. "Google Robot Car: Self Driving Artificial Intelligence Cars Created By Google." Viking305 on HubPages. Web. 21 Nov. 2011. <http://viking305.hubpages.com/hub/Google-testing-robot-cars-driveon-their-own-news-KITT-new-Knight-Rider>. "How Google's Self-Driving Car Works - IEEE Spectrum." IEEE Spectrum: Technology, Engineering, and Science News. Web. 21 Nov. 2011. <http://spectrum.ieee.org/automaton/robotics/artificialintelligence/how-google-self-driving-car-works>. Markoff, John. "Google Cars Drive Themselves, in Traffic NYTimes.com." The New York Times - Breaking News, World News & Multimedia. 09 Oct. 2010. Web. 21 Nov. 2011. <http://www.nytimes.com/2010/10/10/science/10google.html?pagewan Page 19 ted=1>.

THANK YOU
Page 20

You might also like