Transcript [.ppt]

ParkNet
Drive-by Sensing of Road-Side
Parking Statistics
Sutha Mathur, Tong Jin, Nikhil Kasturirangan,
Janani Chandrashekharan, Wenzhi Xue, Marco Gruteser,
Wade Trappe
Rutgers University
Michael Betancourt
UCF - EEL 6788
Dr. Turgut
Overview
1. Introduction
2. Design Goals and Requirements
3. Prototype Development
4. Parking Space Detection
5. Occupancy Map
6. Mobility Study
7. Improvements
8. Conclusion
Introduction - Problems
• Traffic congestion costs tons of money
o 4.2 billion lost hours
o 2.9 billion gallons of gasoline wasted
o Looking for parking contributes to these numbers
• Lack of information
o Hard to determine best prices for meters and where
they should be placed
o Current parking detection systems are costly
Introduction - ParkNet
• Drive-by Parking Monitoring
o Uses ultrasonic sensor
attached to the side of
cars
o Detects parked cars and
vacant spaces
• Attaches to vehicles that
comb through a city (taxi,
police, etc.)
• Location accuracy based on
GPS and environmental
fingerprinting
Introduction - Objectives
• Demonstrating the feasibility of the mobile sensing
approach including the design, implementation and
evaluation of the system
• Proposing and evaluating a method of environmental
fingerprinting to increase location accuracies
• Showing that if the mobility system were currently
attached to operating taxis, it would operate with enough
samples to determine parking availability
Design Goals - Real-time Information
• Improve traveler decisions with respect to mode of
transportation
• Suggesting parking spaces to users driving on the road
• Allow parking garages to adjust their prices dynamically
according to demmand
• Improve efficiency of parking enforcement in systems that
utilize single pay stations for multiple parking spots
Design Goals - Parking Information
• Space count
o Sufficient for most parking applications
• Occupancy Map
o Useful for parking enforcemen
Design Goals - Cost and Participation
• Low-cost Sensors
o Typical per spot parking management systems ranges
from $250 to $800 per spot
o Current systems are difficult to place in areas without
marked parking spots
• Low Vehicle Participation
o Be able to function without a lot of cars fitted
o Keep costs down
Prototype Development - Hardware
• Moxbotix WR1 rangefinder
o Waterproof
o Emits every 50ms
o 12-255 inches
• PS3 Eye webcam
o 20 fps
o Used for ground truth
o Not in production
• Garmin GPS
o Readings come at 5Hz
o Errors can be less than
3m
• On-board PC
o 1GHz CPU
o 512 MB Ram
o 20 GB HD
o PCI WiFi card
o 6 USB ports
Prototype Development - Deployment
• System was placed on 3
vehicles
• 3 specific areas were
marked off to be analyzed
• Data was collected over a 2
month period
• Drivers were oblivious to
the data collection
• All range sensor data is
tagged with:
Kernel-time, range,
latitude, longitude, speed
Prototype Development - Verification
• PS3 Eye
o Mounted just above the rangefinder
o Took pictures at 20fps that were time tagged
• Each picture was manually checked to see if there was a
car parked
• This was used to verify the data collected from the system
Parking Space Detection - Challenges
• Ultrasonic sensor does not have a perfectly narrow-width
• GPS Errors
• False alarms
o Other impeding objects: Trees, people, recycling bins
• Missed detections
o Parked vehicles classified to be something other than a
parked car
Parking Space Detection - Dips
• A "dip" is a change in the rangefinder readings which
usually occurs when there is an object in view
Two Cars Parked Together
Far
Close
Parking Space Detection - Algorithms
• Slotted Model
o Determines which dips are classified as cars
o Subtracts the total number of cars found with the total
number of spaces available in the area
• Unslotted Model
o Determines which dips are classified as cars
o Measures the distance between dips to see if it is large
enough to fit a car
• Training
o 20% of the data is used for training
o 80% of the data is used for evaluating performance
Parking Space Detection - Slotted
Slotted Model Accuracy
Parking Space Detection - Unslotted
Unslotted Model Accuracy
Occupancy Map - GPS Error
• Selected 8 objects and determined their absolute GPS
position using Google Maps
• Corresponded the GPS reading gathered from the trials to
the objects
• Used the reading from one object to correct the others
Occupancy Map - Environmental
Fingerprinting
• F
ixed objects in the environment
used to increase positional
accuracy
• Recognition Walkthrough
1.GPS coordinates indicate system
is near known object
2.Parses rangefinder readings
3.Determines what is not a parked
car
4.Tries match the pattern with the
known object
5.If object found, correct position
if within 100m
Mobility Study - Taxicab Routes
• Public dataset of 536 taxicabs GPS position every 60
seconds
• Routes were approximated by linear interpolation
• Found that taxicabs spend the most time in downtown
areas where parking is scarce
• Determined the mean time between cabs visiting a
particular street.
Mobility Study - Taxicab Mean Time
Greater San Francisco
Downtown San Francisco
Mobility Study - Cost Analysis
• Current Cost:
o Parknet: (~$400 per sensing vehicle) x (number of
vehicles needed to get desired rate of detection)
o Fixed Sensor: ($250-800 per space) x (number of spaces)
• Uses opportunistic WiFi connections to transfer data
• Easily managed due to the much smaller number of fixed
sensors
• Example
o 6000 parking spots
o Parknet: 300 cabs, 80% coverage every 25 minutes, $0.12
million
o Fixed Sensor: $1.5 million
Improvements
• Multilane Roads
o Moving cars could be determined by long dips
o Rangefinder would need to be longer
• Speed Limitations
o Sensors currently work best at speeds below 40mph
• Obtaining Parking Spot Maps
o Difficult for large areas
o Algorithms could determine location surroundings after
data collection has been started
• Using vehicles current proximity sensors
Conclusion
• Data collected
o 500 miles over 2 months
• Accuracy
o 95% accurate parking space counts
o 90% accurate parking occupancy maps
• Frequency and Coverage
o 536 vehicles equipped
o Covers 85% every 25 minutes of a downtown area
o Covers 80% every 10 minutes of a downtown area
• Cost Benefits
o Estimated factor of 10-15 times cheaper than current
systems
• Questions?
Links
Fixed Parking System (SFpark)
http://sfpark.org
http://vimeo.com/13867453