Flash Player 9 (or above) is needed to view presentations.
We have detected that you do not have it on your computer. To install it, go here.

Like this presentation? Why not share!

# Seminar 5520 (Li Li)

## by 黎 李 on Nov 22, 2010

• 549 views

### Views

Total Views
549
Views on SlideShare
549
Embed Views
0

Likes
0
16
0

No embeds

### Categories

Uploaded via SlideShare as Microsoft PowerPoint

## Seminar 5520 (Li Li)Presentation Transcript

• Mobile Phone Based Drunk Driving Detection
Jiangpeng Dai, Jin Teng, XiaoleBai,
ZhaohuiShen, and Dong Xuan
Presented by
Li Li
• Outline
Problem Definition
Acceleration-Based Drunk Driving Cues
System Design & Implementation
Evaluation
Related Work
Discussion
• Problem definition
What is drunk driving?
Why do we need to use smart phone to detect it?
Requirements of drunk driving monitoring system
• Acceleration-Based Drunk Driving Cues
Lateral Acceleration and Lane Position Maintenance
• Acceleration-Based Drunk Driving Cues (cont’d)
Longitudinal Acceleration and Speed Control in Driving
Abrupt acceleration or deceleration
Erratic braking
Jerky stop
• System Design & Implementation
System Overview
• Design of Algorithm
Reading accelerations and angles by using accelerometer and orientation sensor
• Design of Algorithm (cont’d)
Lateral acceleration and longitudinal acceleration detection
• Lateral Acceleration Pattern Matching
The pattern matching is to check the variation between the maximum value and the minimum value of Alat within a pattern checking time window WINlat.
• Longitudinal Acceleration Pattern Matching
When the vehicle acts abnormally in either accelerating or decelerating direction, result in a large absolute value of Alon, making a salient convex or concave shape in its graph of curves.
Set different thresholds for positive Alon and negative Alon.
• Multiple Round Pattern Matching
Multiple round means that the matching process continues round after round, and the trigger condition is satisfied when several numbers of pattern are recognized.
Multiple round pattern matching will increase the accuracy of drunk driving detection.
• Implementation
Drunk driving detection system on Android G1 phone.
Java, with Eclipse and Android 1.6 SDK
Five major components:
User interface
System configuration
Monitoring daemon
Data processing
• EVALUATION
Data Collection
• EVALUATION (cont’d)
Detection Performance
False Negative (FN)
False Positive (FP)
Performance Description
• EVALUATION (cont’d)
Energy Efficiency
• Related Work
GPS
Expensive
Localization error
Energy consuming
Camera
High position requirements
Complicated
Energy consuming for image processing
• Discussion
Create another threshold
Normal