Electricity: measuring and testing – Magnetic – Displacement
Reexamination Certificate
2001-03-14
2003-04-15
Lefkowitz, Edward (Department: 2862)
Electricity: measuring and testing
Magnetic
Displacement
C324S207220, C324S207230
Reexamination Certificate
active
06549004
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to positioning systems. More particularly, it relates to a system and methods for estimating the position and attitude (orientation) of a remote object through the measurement of magnetic fields generated by beacons.
BACKGROUND OF THE INVENTION
1. Motivation
Accurate sensing of the position and attitude of an object is a fundamental requirement in many applications, but is a challenging problem in the cluttered and unstructured environment of the real world. As an example application, mobile robots require position and attitude information to perform redundant or dangerous tasks, such as warehouse automation, floor sweeping, delivery of parts in factories, courier service in offices, or even inspection around the International Space Station. Other applications requiring a positioning system include personnel tracking, such as in military or police training exercises, object tracking, virtual reality, or motion capture for movie effects or animation.
Numerous sensing technologies have been developed to provide the position information required in these applications. Many existing positioning systems are limited in workspace and robustness because they require clear lines of sight or do not provide absolute (drift-free) measurements. For example, overhead cameras have been used to track infrared LEDs on mobile robots. The result is millimeter-level position accuracy, but clear lines of sight must be maintained between cameras and robots. Landmark vision techniques, building-fixed acoustic systems, and RF-based systems also generally require unobstructed lines of sight. Systems based on inertial sensors or wheel encoders avoid line of sight issues, but their measurements are not absolute and are subject to error accumulation. Many on-board vision and laser rangefinder techniques also accumulate error as the robot travels. These restrictions limit the workspace and robustness of a mobile robot in the real factory or warehouse environment.
Positioning systems based on magnetic fields offer an important alternative by providing absolute position and attitude information with the potential for no ‘line of sight’ restrictions.
2. Background Art
Magnetic field positioning systems use electrical current running through wires (referred to herein as ‘beacons’) to create magnetic fields. Magnetic fields obey superposition and therefore simply add as vectors. A remote sensor unit measures the sum magnetic field at its location. This sum magnetic field is a vector and has 3 independent components. For the sensor to determine its position and attitude (6 quantities), it must have more information than just the sum field. The sensor must have some way of determining the magnetic field due to an individual beacon or combination of beacons. In other words, the sensor must be able to distinguish the portion of the sum field that is produced by an individual beacon or combination of beacons. For example, if the sensor can determine the magnetic field produced by one beacon (3 measured quantities) and can also determine the magnetic field due to a second beacon (another 3 quantities), it may be possible to solve for position and orientation. Methods for distinguishing signals are known as ‘multiple access’ methods.
Existing magnetic field positioning systems use signal structures that fall into two categories, ‘AC’ fields or ‘pulsed DC’ fields.
In ‘AC’ systems, described in U.S. Pat. No. 3,868,565 (Kuipers) and U.S. Pat. No. 4,054,881 (Raab), the magnetic fields are sinusoidal in nature. Beacons produce fields at different frequencies, and frequency filtering is used by the remote sensor to distinguish the fields from individual beacons. Example waveforms for 3 beacons are shown in FIG.
1
. This multiple access technique is known as Frequency Division Multiple Access (FDMA).
In ‘pulsed DC’ systems, described in U.S. Pat. No. 4,849,692 (Blood), U.S. Pat. No. 4,945,305 (Blood), and U.S. Pat. No. 5,453,686 (Anderson), the magnetic fields are generated in pulses. Example waveforms for 3 beacons are shown in FIG.
2
. In this example, discrimination of the field due to a particular beacon is straightforward—only one beacon is producing a field at a given time. This multiple access method, in which a beacon produces fields during some ‘time slots’ and not during others, is known as Time Division Multiple Access (TDMA).
Further refinements to these magnetic field positioning systems are described in U.S. Pat. No. 5,640,170 (Anderson), where magnetic fields are created from a special anti-distortion source configuration, and U.S Pat. No. 5,600,330 (Blood), which can function with non-dipole fields. Eddy currents are a common error source in magnetic field systems, and methods to reduce this error are described in U.S Pat. No. 5,767,669 (Hansen et al) for ‘pulsed DC’ systems and U.S. Pat. No. 6,172,499 (Ashe) for ‘AC’ systems. Methods typically involve complicated schemes for detecting and removing effects of eddy currents on the computed signal.
3. Limitations of the Current State of Technology
Existing magnetic field positioning systems use signal structures and multiple access methods that are sub-optimal, especially when applied to a system with a large number of beacons. Improvements in accuracy, uniformity of coverage, sensor complexity, and magnetic field levels are still desirable for many applications.
Further, existing magnetic field positioning systems are generally based on fields created in one location, typically by 3 concentric beacons, giving ‘room-sized’ coverage areas. A much larger ‘building-sized’ coverage area could be provided if many beacons, perhaps hundreds or thousands, could be distributed at numerous locations throughout an entire area. A challenge that arises, however, is the processing method to estimate position and attitude given magnetic fields produced by beacons in multiple locations. The equations used to solve for position and attitude are nonlinear, and do not converge to a solution using standard techniques unless a good estimate of position and attitude is already known.
SUMMARY OF THE INVENTION
The present invention provides an apparatus and methods for a magnetic field positioning system to use a fundamentally different, and advantageous, signal structure and multiple access method, known as Code Division Multiple Access (CDMA). The apparatus contains a plurality of beacons at known positions and orientations, each generating a magnetic field that varies according to a pseudorandom code. A magnetic sensor produces magnetic field measurements that are analyzed by a processor to discriminate the fields produced by individual beacons and determine the position and attitude of the sensor. The beacons can also have their own sensors and processors to determine their relative positions. The magnetic sensor is typically fixed to an object whose position is being monitored.
The present invention also provides a method for estimating the position and attitude of a magnetic sensor, including the following steps: using a set of beacons, generating a plurality of magnetic fields at known beacon positions and orientations; measuring a sum magnetic field at a sensor position; and analyzing the magnetic field measurements to estimate a position and attitude of the sensor. The generated magnetic fields vary according to mutually orthogonal pseudorandom codes. Analysis includes correlating the magnetic field measurements with the codes and correcting for cross-correlation among codes.
This CDMA signal architecture, when combined with processing methods, leads to numerous advantages over the existing technologies, especially when applied to a system with a large number of magnetic field generators (beacons). Advantages include:
concentration of signal energy at lower frequencies, reducing eddy current noise and increasing accuracy;
CDMA ‘processing gain’ increases signal to noise ratio, increasing accuracy;
uniform coverage;
sensor unit less complex and easier to tune;
lower usable signal levels.
Further, the present invention
Aurora Reena
Lefkowitz Edward
Lumen Intellectual Property Services Inc.
The Board of Trustees of the Leland Stanford Junior University
LandOfFree
Distributed magnetic field positioning system using code... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Distributed magnetic field positioning system using code..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Distributed magnetic field positioning system using code... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3008980