08-03-2011, 02:43 PM
PRESENTED BY
CHITRA L
ARTIFICIAL PASSENGER.ppt (Size: 741.5 KB / Downloads: 270)
Introduction
■ Driver fatigue according to NHTSA annually causes
1,00,000 crashes
1,500 fatalities
71,000 injuries
Majority of the road accidents were preceded by eye closures of 0.5 second as long as 2 to 3 seconds
A normal human blink last 0.2 to 0.3 seconds
Existing System
Miniature system installed in driver’s hat.
Use of simulation drinks (eg: coffee &tea).
Tablets to prevent sleeping
In order to overcome the disadvantages of existing method IBM introduces a new sleep prevention technology device.
ARTIFICIAL PASSENGER
Developed By
IBM (International Business Machines Corporation, NY) has developed a software that holds the conversation with driver to determine whether the driver can respond alertly enough, called “Artificial Passenger”.
Designed to make solo journey safer and more bearable.
Showing the dashboard of the car where the whole
AP system is attached
About Artificial Passenger
The AP is an Artificial Intelligence based companion that will be resident in software and chips embedded in the automobile dashboard.
Conversational planner: holds a profile of the driver, including details of his interests and profession.
A microphone picks up driver’s answer and breaks it down into separate words with speech-recognition software.
A camera built into the dashboard, tracks driver’s head and lip movements to improve the accuracy of speech recognition.
A voice analyzer then looks for signs of tiredness by checking to see if the answer matches your profile. Slow responses and lack of attention are signs of fatigue.
There is a mobile indicator device, with the help of which the driver can answer the call using touch sensitive wheel.
The touch sensitive wheels have sensors on it.
It helps the driver to control
the devices of the car
without interrupting the
dialog between dashboard
and the driver.
Components
The components which supports the working of the system:
Automatic speech recognizer (ASR)
Natural language processor (NLP)
Driver analyzer
Conversational planner (CP)
Alarm, Microphone
Camera and Eye tracker
Automatic speech recognition
Allows the computer to identify the words spoken on to a microphone.
Two ASRs used:
Speaker independent ASR – used for decoding voice signals of the driver.
Operates with voice car media and decodes tapes, audio book, telephone mails etc.
Natural language processor
NL Processing is a combination of CS and linguistics.
Major functions of NLP are:
Co – reference resolution
Machine translation
Natural language generation
Natural language understanding
Processes decoded signals of textual data provided from ASR.
Identifies related meanings from the contents of the decoded messages.
Produces variant of responses.
This output goes to the driver analyzer as an input.
Driver analyzer
Receives the textual data and voice data from NLP.
Measures the time response using a clock.
Depending on the time responses, conclusions about driver’s alertness will be given to the conversational planner.
Steps for detecting driver vigilance
Conversational planner
It is the heart of the system.
Instructs the language generator to produce the response.
If the driver is in perfect condition CP instructs the language generator to continue the conversation otherwise it is instructed to change the conversation.
For each question, it creates a set of possible answers.
Alarm
If the CP (conversational planner) receives information that the driver is about to fall asleep, an alarm (buzzer) system is activated.
Microphone
For picking up the words and separate them by some internally used software for conversation.
Camera
■ Analyze video sequences of the driver to determine whether the driver is paying adequate attention to the road.
■ Track the head and lip movements of the driver.
Used for the improvement of accuracy of the speech recognition system.
Calculations vary for day and night conditions
Night time is determined based on the following conditions:
If camera clock is between 18.00 hrs and 07.00 hrs
If the intensity of image is lower than the threshold value
If the night time driving
headlight is on
Eye Tracker
Eye-tracking software is used for this purpose. The camera will track the eyes.
System measures the corneal reflection of an infrared light emitting diode (LED), which illuminates and generates a reflection of the surface of the eye. Glint is used for calibration and tracking.
Working of eye tracker
Working
Conversational planner initiates the conversation
Camera tracks the head and lip movements and also eye movements and detect whether the driver is alert or not.
It detects the frequency of yawning of the driver and generates an alarm signal if the above frequencies exceeds a selected threshold value.
It allows the driver to open and close the door from a remote location.
Communicates with drivers of other vehicles is the driver has a heart attack or is sick
Using a touch sensitive wheel the driver can open the window, control the volume of the ipod, attend phone calls etc.
Give information about the chances of vehicle failure and maintains records for accident investigation case.
Aspects And Features
There are five aspects of Artificial Passenger
Situation Controller
Camera system
Temperature Indicator
Odor Detector
Communication System
Workload manager
Situation controller
The situation controller consists of:
Processing Device and an image monitor coupled to processing device
Communication Device for communicating message
Controller for controlling functions of vehicle
Camera system
The camera system is used :
To monitor the environment of a vehicle.
Provides alarm indication to the owner.
Identifies position of items & notifies the driver about its position.
Temperature Indicator
To analyze temperature of vehicle.
In combination with camera, the AP determines whether a child or pet is left in a vehicle that is beginning to get very hot or cold.
It sends a message to the owner/driver and takes the correct actions
The AP is able to analyze the situation & executes a corrective action
Opens window or door to allow temperature to moderate
Allow child or pet to leave the vehicle after informing the driver /authorities.
Ordor Sensor/ Detector
AP uses Odor Detector as well as camera to detect whether groceries were left in the vehicle.
AP informs the driver/owner about this.
Periodically sprinkles sweet air inside the vehicle.
Communicaiton system
Interacts with owner of vehicle from remote location.
It utilizes GPS to locate & communicate with vehicle owner.
Through communication system the owner can open a vehicle door remotely & let the person out who has been locked
Workload Manager
Determine a moment – to – moment analysis of the user's cognitive workload.
Done by collecting data about user conditions, monitoring local and remote events.
Read the profiles and mood of drivers in other vehicles.
It must be designed in such a way that it can integrate sensor information and rules on when and if distracting information is delivered.
The workload manager is connected to the safety driver manager (SDM).
The goal of SDM is to evaluate the potential risk of a traffic accident by producing measurements related to stresses on the driver or vehicle.
Features
The main features of the AP are:
Conversational Telematics
Analyzing the data
Sharing data
Retrieving the data on demand
Conversational Telematics
Intelligent enough to anticipate the user needs
Provides uniform access to devices and network
services inside and outside the car
Reports the car conditions
Makes you stay awake
Analyzing the data
Automated analysis initiative is data management software for identifying failure trends and predicting specific vehicle failures before they happen.
System consists of capturing, storing, retrieving and analyzing vehicle data.
Evaluates the data and takes the corrective measures.
Internet based diagnostics server reads the car data to determine the root cause of vehicle failure
Sharing data
Collecting dynamic and event – driven data is a problem.
Ensuring security and integrity while sharing data.
Retrieving data on demand
Resource manager must manage broad range of data that changes rapidly.
Server must give service providers the ability to declare what data they need, without knowing location of data.
Applications
Prevents the driver, falling asleep during long and solo trip.
If any problem it alerts the vehicles near by this, so the driver there becomes alert.
Opens and closes the doors and windows of the car automatically.
It is also used for the entertainment (playing verbal games, listening to music etc)
Advantages
Makes automobile driving more convenient.
Checks for signs of sleepiness, if detected alert the driver.
Maintains a database for “accident investigation use”.
Sensors in the front and rear of the automobile controls the distance between cars and automatically apply breaks if something gets in front of the car.
Disadvantages
High cost.
Sensors in the front and rear of automobile doesn’t help if danger comes from sides.
NLP component should be downside to run on local computers.
Remote connections to severs are not available everywhere, can have delays, and are not robust.
Some users will produce some phrases that are not represented in collected data nor in grammar that are developed from this data.
Future Enhancements
Provide us with a shortest-time routing based on road conditions changing because of weather and traffic
Information about the cars on the route
Provides distributive user interface between cars
Conclusion
Successful implementation of artificial passenger would allow use of various service in car like reading emails, navigation, downloading music files, voice games without compromising a driver safety.
CHITRA L
ARTIFICIAL PASSENGER.ppt (Size: 741.5 KB / Downloads: 270)
Introduction
■ Driver fatigue according to NHTSA annually causes
1,00,000 crashes
1,500 fatalities
71,000 injuries
Majority of the road accidents were preceded by eye closures of 0.5 second as long as 2 to 3 seconds
A normal human blink last 0.2 to 0.3 seconds
Existing System
Miniature system installed in driver’s hat.
Use of simulation drinks (eg: coffee &tea).
Tablets to prevent sleeping
In order to overcome the disadvantages of existing method IBM introduces a new sleep prevention technology device.
ARTIFICIAL PASSENGER
Developed By
IBM (International Business Machines Corporation, NY) has developed a software that holds the conversation with driver to determine whether the driver can respond alertly enough, called “Artificial Passenger”.
Designed to make solo journey safer and more bearable.
Showing the dashboard of the car where the whole
AP system is attached
About Artificial Passenger
The AP is an Artificial Intelligence based companion that will be resident in software and chips embedded in the automobile dashboard.
Conversational planner: holds a profile of the driver, including details of his interests and profession.
A microphone picks up driver’s answer and breaks it down into separate words with speech-recognition software.
A camera built into the dashboard, tracks driver’s head and lip movements to improve the accuracy of speech recognition.
A voice analyzer then looks for signs of tiredness by checking to see if the answer matches your profile. Slow responses and lack of attention are signs of fatigue.
There is a mobile indicator device, with the help of which the driver can answer the call using touch sensitive wheel.
The touch sensitive wheels have sensors on it.
It helps the driver to control
the devices of the car
without interrupting the
dialog between dashboard
and the driver.
Components
The components which supports the working of the system:
Automatic speech recognizer (ASR)
Natural language processor (NLP)
Driver analyzer
Conversational planner (CP)
Alarm, Microphone
Camera and Eye tracker
Automatic speech recognition
Allows the computer to identify the words spoken on to a microphone.
Two ASRs used:
Speaker independent ASR – used for decoding voice signals of the driver.
Operates with voice car media and decodes tapes, audio book, telephone mails etc.
Natural language processor
NL Processing is a combination of CS and linguistics.
Major functions of NLP are:
Co – reference resolution
Machine translation
Natural language generation
Natural language understanding
Processes decoded signals of textual data provided from ASR.
Identifies related meanings from the contents of the decoded messages.
Produces variant of responses.
This output goes to the driver analyzer as an input.
Driver analyzer
Receives the textual data and voice data from NLP.
Measures the time response using a clock.
Depending on the time responses, conclusions about driver’s alertness will be given to the conversational planner.
Steps for detecting driver vigilance
Conversational planner
It is the heart of the system.
Instructs the language generator to produce the response.
If the driver is in perfect condition CP instructs the language generator to continue the conversation otherwise it is instructed to change the conversation.
For each question, it creates a set of possible answers.
Alarm
If the CP (conversational planner) receives information that the driver is about to fall asleep, an alarm (buzzer) system is activated.
Microphone
For picking up the words and separate them by some internally used software for conversation.
Camera
■ Analyze video sequences of the driver to determine whether the driver is paying adequate attention to the road.
■ Track the head and lip movements of the driver.
Used for the improvement of accuracy of the speech recognition system.
Calculations vary for day and night conditions
Night time is determined based on the following conditions:
If camera clock is between 18.00 hrs and 07.00 hrs
If the intensity of image is lower than the threshold value
If the night time driving
headlight is on
Eye Tracker
Eye-tracking software is used for this purpose. The camera will track the eyes.
System measures the corneal reflection of an infrared light emitting diode (LED), which illuminates and generates a reflection of the surface of the eye. Glint is used for calibration and tracking.
Working of eye tracker
Working
Conversational planner initiates the conversation
Camera tracks the head and lip movements and also eye movements and detect whether the driver is alert or not.
It detects the frequency of yawning of the driver and generates an alarm signal if the above frequencies exceeds a selected threshold value.
It allows the driver to open and close the door from a remote location.
Communicates with drivers of other vehicles is the driver has a heart attack or is sick
Using a touch sensitive wheel the driver can open the window, control the volume of the ipod, attend phone calls etc.
Give information about the chances of vehicle failure and maintains records for accident investigation case.
Aspects And Features
There are five aspects of Artificial Passenger
Situation Controller
Camera system
Temperature Indicator
Odor Detector
Communication System
Workload manager
Situation controller
The situation controller consists of:
Processing Device and an image monitor coupled to processing device
Communication Device for communicating message
Controller for controlling functions of vehicle
Camera system
The camera system is used :
To monitor the environment of a vehicle.
Provides alarm indication to the owner.
Identifies position of items & notifies the driver about its position.
Temperature Indicator
To analyze temperature of vehicle.
In combination with camera, the AP determines whether a child or pet is left in a vehicle that is beginning to get very hot or cold.
It sends a message to the owner/driver and takes the correct actions
The AP is able to analyze the situation & executes a corrective action
Opens window or door to allow temperature to moderate
Allow child or pet to leave the vehicle after informing the driver /authorities.
Ordor Sensor/ Detector
AP uses Odor Detector as well as camera to detect whether groceries were left in the vehicle.
AP informs the driver/owner about this.
Periodically sprinkles sweet air inside the vehicle.
Communicaiton system
Interacts with owner of vehicle from remote location.
It utilizes GPS to locate & communicate with vehicle owner.
Through communication system the owner can open a vehicle door remotely & let the person out who has been locked
Workload Manager
Determine a moment – to – moment analysis of the user's cognitive workload.
Done by collecting data about user conditions, monitoring local and remote events.
Read the profiles and mood of drivers in other vehicles.
It must be designed in such a way that it can integrate sensor information and rules on when and if distracting information is delivered.
The workload manager is connected to the safety driver manager (SDM).
The goal of SDM is to evaluate the potential risk of a traffic accident by producing measurements related to stresses on the driver or vehicle.
Features
The main features of the AP are:
Conversational Telematics
Analyzing the data
Sharing data
Retrieving the data on demand
Conversational Telematics
Intelligent enough to anticipate the user needs
Provides uniform access to devices and network
services inside and outside the car
Reports the car conditions
Makes you stay awake
Analyzing the data
Automated analysis initiative is data management software for identifying failure trends and predicting specific vehicle failures before they happen.
System consists of capturing, storing, retrieving and analyzing vehicle data.
Evaluates the data and takes the corrective measures.
Internet based diagnostics server reads the car data to determine the root cause of vehicle failure
Sharing data
Collecting dynamic and event – driven data is a problem.
Ensuring security and integrity while sharing data.
Retrieving data on demand
Resource manager must manage broad range of data that changes rapidly.
Server must give service providers the ability to declare what data they need, without knowing location of data.
Applications
Prevents the driver, falling asleep during long and solo trip.
If any problem it alerts the vehicles near by this, so the driver there becomes alert.
Opens and closes the doors and windows of the car automatically.
It is also used for the entertainment (playing verbal games, listening to music etc)
Advantages
Makes automobile driving more convenient.
Checks for signs of sleepiness, if detected alert the driver.
Maintains a database for “accident investigation use”.
Sensors in the front and rear of the automobile controls the distance between cars and automatically apply breaks if something gets in front of the car.
Disadvantages
High cost.
Sensors in the front and rear of automobile doesn’t help if danger comes from sides.
NLP component should be downside to run on local computers.
Remote connections to severs are not available everywhere, can have delays, and are not robust.
Some users will produce some phrases that are not represented in collected data nor in grammar that are developed from this data.
Future Enhancements
Provide us with a shortest-time routing based on road conditions changing because of weather and traffic
Information about the cars on the route
Provides distributive user interface between cars
Conclusion
Successful implementation of artificial passenger would allow use of various service in car like reading emails, navigation, downloading music files, voice games without compromising a driver safety.