Browse Prior Art Database

ABSOLUTE HEADING SENSOR BLUNDER DETECTION USING RELATIVE HEADING SENSOR AND ROAD SEGMENT

IP.com Disclosure Number: IPCOM000007886D
Original Publication Date: 1996-Nov-01
Included in the Prior Art Database: 2002-May-02
Document File: 4 page(s) / 167K

Publishing Venue

Motorola

Related People

Vilin Zhao: AUTHOR [+3]

Abstract

The absolute heading sensor is a very important component of an in-vehicle navigation system. The least expensive one used nowadays is the magnetic compass. However, the problem with this kind of sensor is the sensitivity to magnetic anomalies. Whenever the vehicle is close to a bridge, an over- pass, an elevator track, a steel building or a con- struction site, the compass is affected by the abnor- mal magnetic field. Because of the randomness of these errors, it is very hard to use filters to correct the compass readings. The heading data received from Global Positioning System (GPS) has similar problem. When the satellite constellation is changed, the receiver may have abnormal heading output. In addition, when the GPS receiver is used in urban canyon areas, the multiple path signals may affect the accuracy of the GPS heading. 2. DESCRIPTION This method is based on the assumption that the relative heading sensor is quite accurate in the measurement of the heading variance most of the time. Our field tests demonstrated the correctness of this assumption. The method is based on a set of if-then-else rules to detect blunders and to make corrections.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 43% of the total text.

Page 1 of 4

Technical Developments

ABSOLUTE HEADING SENSOR BLUNDER DETECTION USING RELATIVE HEADING SENSOR AND ROAD SEGMENT

by Vilin Zhao, Leslie G. Seymour, and Elisha M. Kozikaro


1. PROBLEM STATEMENT

  The absolute heading sensor is a very important component of an in-vehicle navigation system. The least expensive one used nowadays is the magnetic compass. However, the problem with this kind of sensor is the sensitivity to magnetic anomalies. Whenever the vehicle is close to a bridge, an over- pass, an elevator track, a steel building or a con- struction site, the compass is affected by the abnor- mal magnetic field. Because of the randomness of these errors, it is very hard to use filters to correct the compass readings. The heading data received from Global Positioning System (GPS) has similar problem. When the satellite constellation is changed, the receiver may have abnormal heading output. In addition, when the GPS receiver is used in urban canyon areas, the multiple path signals may affect the accuracy of the GPS heading.

2. DESCRIPTION This method is based on the assumption that the relative heading sensor is quite accurate in the measurement of the heading variance most of the time. Our field tests demonstrated the correctness of this assumption. The method is based on a set of if-then-else rules to detect blunders and to make corrections.

then

l

very likely blunder occurs and use relative sen- sor variance to correct the system heading.

  The threshold values are determined based on the relative sensor used. For instance, in our imple- mentation, the Tl is 1.6 degrees for the differential odometer and the T2 is 5 degrees for the magnetic compass.

the cumulative heading variance in each con- secutive cycle is either more than 1.6 degrees positive or less than 1.6 degrees negative for more than 5 cycles,

l

then

l

very likely the vehicle is turning.

the vehicle is turning and absolute heading var- iance is 5 degrees more than the relative head- ing variance,

l

then

2.1 Without Map

  If there is no digital map available, the blunder detection will not have as high confidence level as with the map.

Detection Method:

l very likely blunder occurs and use relative head- ing variance to correct the system heading.

2.2 With Map

  If there is a digital map available, the blunder detection is easier than without the map. The basic algorithm for software execution flow is shown in Figure 1, where Hr stands for the heading of the relative sensor, Ha stands for the heading of the abso- lute sensor, Hs stands for the heading of the road segment. Tl and T2 represents for a relative head- ing variance threshold and an absolute heading var- iance threshold respectively

the relative heading variance is less than a threshold Tl, and

absolute sensor variance is more than thresh- old T2,

l

l

B Motaola. 1°C. 1996 104 November1996

[This page contains 14 pictures or other non-text objects]

Page 2 of 4

MO7VROLA Technical Developments

8

Obtain Heading of Relative Sen...