Data Fusion of Four ABS Sensors and GPS for an Enhanced Localization of Car-like Vehicles - PowerPoint PPT Presentation

About This Presentation
Title:

Data Fusion of Four ABS Sensors and GPS for an Enhanced Localization of Car-like Vehicles

Description:

Data Fusion of Four ABS Sensors and GPS for an Enhanced Localization of Car-like ... Each time a GPS measurement is performed. ... – PowerPoint PPT presentation

Number of Views:98
Avg rating:3.0/5.0
Slides: 13
Provided by: ics9
Learn more at: https://ics.uci.edu
Category:

less

Transcript and Presenter's Notes

Title: Data Fusion of Four ABS Sensors and GPS for an Enhanced Localization of Car-like Vehicles


1
Data Fusion of Four ABS Sensors and GPS for an
Enhanced Localization of Car-like Vehicles
by Philippe Bonnifait, Pascal Bouron, Paul
Crubille, and Dominique Meizel
  • Wiwat Ruengmee
  • ICS 280 Special Topic in Ubiquitous Computing
  • Ubiquitous Computing for Post-Crisis Logistics
  • 01/26/2006

2
Motivations / Problems
  • Wide availability of GPS and ABS systems in
    modern cars.
  • GPS suffers from satellite masks.
  • Classical solutions to satellite masks problem of
    GPS are an odometer, a gyro, and a magnetic
    compass.
  • Implement and test the localization system using
    GPS, ABS sensors, and a driving wheel encoder.
  • Improve the precision due to the redundancy of
    measurements.
  • Odometry versus Dead Reckoning.
  • This paper presents a new odometric technique and
    the data fusion using Extended Kalman Filter
    (EKF) to estimate the localization.

3
Approach (1)
  • Odometric technique using data from 4 ABS sensors
    together with a driving wheel encoder.
  • Basics
  • Displacement between two samples
  • How to estimate ? and ??
  • using the steering wheel ?and the distances
    travelled by each wheel

4
Approach (2)
5
Approach (3)
  • Differential odometry estimation and their
    estimation
  • Non-linearity leads to use of Extended Kalman
    Filter (EKF).
  • Each (5) estimation of (?, ?) to provide an
    odometric location is called odometric EKF.

6
Approach (4)
  • Sampling is performed when
  • Each time a GPS measurement is performed.
  • Each time the car has travelled one meter and the
    steering angle has changed more than 0.5 degree.
  • Sampling frequency is not constant and higher
    than 1 Hz

7
Results of the Odometric EKF
8
Architecture of localization system
GPS
1PPS signal
(x, y)
(x, y, F) P
ABS Computer Steering Angle
Odometric EKF
Localiser EKF
? S
v
? An estimation S Covariance matrix
9
Results of Sensor Fusion Of GPS and Odemetry
  • Experiments
  • 9.6 km long course
  • Three characteristics
  • normal operation
  • seven small GPS masks of ten seconds
  • Large masks of 5 minutes
  • Use of all the ABS sensors increases the
    precision of the positioning system.

10
Results of Sensor Fusion Of GPS and Odometry
11
Results of Sensor Fusion Of GPS and Odometry
12
Question?
Write a Comment
User Comments (0)
About PowerShow.com