Method of estimating location and orientation of magnetic dipoles using extended Kalman filtering and Schweppe likelihood ratio detection

• US 5,381,095 A
• Filed: 06/21/1993
• Issued: 01/10/1995
• Est. Priority Date: 06/21/1993
• Status: Expired due to Term

• Current Assignee: Conexant Systems Incorporated (Synaptics Incorporated)
• Original Assignee: Rockwell International Corporation
• Inventors: Andrews, Angus P.
• Primary Examiner(s): Wieder, Kenneth A.
• Assistant Examiner(s): PHILLIPS, ROGER
• Application Number:
  US08/080,907
• Time in Patent Office: 568 Days
• Field of Search: 340/995, 340/988, 364/424.02, 364/436, 324/207.24, 324/207.11, 324/228, 324/244, 324/326, 324/260
• US Class Current: 324/326
• CPC Class Codes: G01C 21/28 with correlation of data from several navigational instruments
• CPC Class Codes: G01V 3/081 the magnetic field is produced by the objects or geological structures characterised by the method of magnetic field measurement G01R33/00
• CPC Class Codes: G08G 1/042 using inductive or magnetic detectors
• CPC Class Codes: G08G 1/096716 where the received information does not generate an automatic action on the vehicle control
• CPC Class Codes: G08G 1/096758 where no selection takes place on the transmitted or the received information
• CPC Class Codes: G08G 1/096783 where the origin of the information is a roadside individual element
• CPC Class Codes: G08G 1/22 Platooning, i.e. convoy of communicating vehicles

Abstract

A method for real time processing of information received from a plurality of permanent magnets embedded in the pavement of an Intelligent Vehicle Highway System. A sequence of magnet orientations provides coded information to sensors mounted on automotive vehicles. The method uses a combination of extended Kalman filtering and Schweppe likelihood estimation to determine the relative location and orientation of the magnets. A separate location estimator (extended Kalman filter) is provided for each of the possible magnet orientations. A decision process identifies the most likely orientation of the magnet, and the Kalman filter selected by the process contains the best estimate of the location of the magnet. Schweppe likelihood estimation is extended to use the detector as a decoder of the information provided by the orientation sequence of the magnets. Failure of the decision process to arrive at a clearly defined decision (resulting, for example, from a missing or defective magnet) can be carried to a higher decoding level in which possible sequences of orientations and their relative likelihood can be used to select the most likely sequence.

Claims

1. A method of determining the location and orientation of a magnet having one of a known set of orientations, comprising the steps of:

providing a plurality of extended Kalman filters, each of said extended Kalman filters corresponding to an assumed one of the known set of possible magnet orientations;

measuring a magnetic field produced by the magnet;

providing sensor signals corresponding to said measured magnetic field to each of said plurality of extended Kalman filters;

each of said extended Kalman filters generating an estimate of the relative location of the magnet based on said assumed orientation and generating a likelihood that said assumed orientation corresponds to the orientation of the magnet;

determining the most likely one of said assumed orientations;

selecting the one of said Kalman filters corresponding to said most likely one of said assumed orientations; and

providing output signals corresponding to said most likely orientation and estimate of the relative location of the magnet generated by said selected Kalman filter.

2. The method of claim 1, wherein the step of measuring said magnetic field comprises making a sequence of measurements.

3. The method of claim 2, wherein the step of generating an estimate of the magnet location further comprises generating measurement innovations sequences from each of said extended Kalman filters.

4. The method of claim 3, wherein the step of determining the most likely one of said assumed orientations further comprises computing relative likelihoods of said assumed orientations from said measurement innovations sequences.

5. A method of determining the location and orientation of a magnet having one of a known set of orientations, comprising the steps of:

providing a plurality of extended Kalman filters, each of said extended Kalman filters based on an assumed orientation corresponding to one of the known set of orientations of the magnet;

making a sequence of measurements of a magnetic field produced by the magnet;

providing sensor signals corresponding to said magnetic field measurements to each of said plurality of extended Kalman filters;

each of said extended Kalman filters generating an estimate of the relative location of the magnet, generating measurement innovations sequences comprising differences between each of said measurements and a predicted value, and feeding back said innovations to correct said relative location estimate;

computing relative likelihoods of each of said assumed orientations from said innovations sequences using Schweppe likelihood ratio detection;

selecting the one of said extended Kalman filters corresponding to the most likely one of said assumed orientations; and

providing output signals from said selected Kalman filter corresponding to said most likely orientation and relative location estimate of the magnet.

6. The method of claim 5, further comprising the steps of:

measuring magnetic fields produced by a plurality of magnets in sequence, each of said magnets having one of said set of known orientations, said sequence of magnets providing coded data; and

determining orientation and location of each of said plurality of magnets; and

receiving said coded data from said sequence of magnets.

7. The method of claim 5, wherein the step of measuring magnetic fields comprises measuring magnetic fields of magnets embedded in a highway with sensors mounted on an automotive vehicle.

8. A method of determining the location and orientation of a magnet embedded in a highway with one of a known set of orientations, comprising the steps of:

providing an automotive vehicle with a magnetic sensor;

making a sequence of measurements of a magnetic field produced by the magnet as said vehicle passes over said magnet;

providing a plurality of extended Kalman filters, each of said extended Kalman filters based on an assumed orientation corresponding to one of the known set of orientations of the magnet;

providing signals corresponding to said magnetic field measurements from said sensor to each of said plurality of extended Kalman filters;

each of said extended Kalman filters generating an estimate of the relative location of the magnet, generating measurement innovations sequences comprising differences between each of said measurements and a predicted value, and feeding back said innovations to correct said relative location estimate;

computing relative likelihoods of each of said assumed orientations from said innovations sequences using Schweppe likelihood ratio detection;

selecting the one of said extended Kalman filters corresponding to the most likely one of said assumed orientations; and

providing output signals from said selected Kalman filter corresponding to said most likely orientation and relative location estimate of the magnet.

9. The method of claim 8, further comprising the steps of:

measuring magnetic fields produced by a plurality of magnets embedded in sequence in said highway, each of said magnets having one of said set of known orientations, said sequence of magnets providing coded data; and

determining orientation and location of each of said plurality of embedded magnets; and

receiving said coded data from said sequence of embedded magnets.

10. The method of claim 9, further comprising the step of using said orientation relative likelihoods for detection and correction of decoding errors from said sequence of embedded magnets.