TRACKING ALGORITHM

• US 20140139375A1
• Filed: 06/19/2012
• Published: 05/22/2014
• Est. Priority Date: 06/21/2011
• Status: Active Grant

• Current Assignee: BAE Systems Plc
• Original Assignee: BAE Systems Plc
• Inventors: Faragher, Ramsey Michael, Sarno, Giuseppe Carlos
• Granted Patent: US 9,568,587 B2
• Time in Patent Office: Days
• US Class Current: 342/451
• CPC Class Codes: G01S 5/02525 Gathering the radio frequency fingerprints
• CPC Class Codes: G01S 5/0278 involving statistical or probabilistic considerations G01S5/0252, G01S5/0294 take precedence
• CPC Class Codes: G01S 5/0294 Trajectory determination or predictive filtering, e.g. target tracking or Kalman filtering

Abstract

A method of tracking an entity by monitoring a signal, the signal tending to vary spatially and be generally time-invariant, the entity moving from a first location within an area to a second location within the area, the method being suitable for use when the location of the source of the signal is unknown, the method comprising providing a plurality of particles for use with a particle filter, each particle being associated with a first particle location, a first particle location being an estimate of the first location of the entity, providing an estimate of the motion of the entity between the first location and the second location, using the estimate of the motion and using the particle filter, for each particle, updating the first particle location for that particle thereby producing an updated particle location, the updated particle location being an estimate of the second location of the entity, for each updated particle, estimating at least one expected signal parameter at the updated particle location, measuring a signal parameter at the second location of the entity, assigning a weight to each updated particle depending on the expected signal parameter estimated for that particle and the measured signal parameter, estimating the second location of the entity by determining a function of the weighted updated particles, and inputting the estimated location and measured signal parameter, as a location/parameter data set, to a database.

Claims

1. A method of tracking an entity by monitoring at least one parameter of a signal, the signal tending to vary spatially and be generally time-invariant, the entity moving from a first location within an area to a second location within the area, the method being suitable for use when the location of the source of the signal is unknown, the method comprising:

providing a plurality of particles for use with a particle filter, each particle being associated with a first particle location, a first particle location being an estimate of the first location of the entity;

providing an estimate of the motion of the entity between the first location and the second location;

using the estimate of the motion and using the particle filter, for each particle, updating the first particle location for that particle thereby producing an updated particle location, the updated particle location being an estimate of the second location of the entity;

for each updated particle, estimating at least one expected signal parameter at the updated particle location;

measuring a signal parameter at the second location of the entity;

assigning a weight to each updated particle depending on the expected signal parameter estimated for that particle and the measured signal parameter;

estimating the second location of the entity by determining a function of the weighted updated particles; and

inputting the estimated location and measured signal parameter, as a location/parameter data set, to a database.

2. A method according to claim 1 comprising monitoring a first radio signal having a first frequency wherein at least one signal parameter is the signal strength of the first radio signal.

3. A method according to claim 1 comprising monitoring a second radio signal, the second radio signal having a second frequency distinct from the first frequency, and wherein at least one signal parameter is the signal strength of the second radio signal.

4. A method according to claim 1 comprising monitoring the local magnetic field and wherein at least one signal parameter is the magnetic field strength.

5. A method according to claim 1 comprising monitoring the local magnetic field and wherein at least one signal parameter is the vector of the magnetic field.

6. A method according to claim 1, wherein the step of providing an estimate of the motion of the entity between the previous location and the current location comprises implementing a zero-velocity update or ZUPT process.

7. A method according to claim 1, wherein:

the entity has mounted or coupled to it, or the entity comprises, an inertial measurement unit; and

the step of providing an estimate of the motion of the entity between the previous location and the current location is performed using data from the inertial measurement unit,

the method further comprising estimating one or more biases of the inertial measurement unit, wherein the step of, for each particle, updating the particle location for that particle is performed using the estimate of one or more biases of the inertial measurement unit.

8. A method according to claim 7, wherein the step of estimating one or more biases of the inertial measurement unit is performed using a Kalman filter.

9. A method according to claim 7, the method further comprising, using the estimated second location of the entity, updating the estimates of the one or more biases of the inertial measurement unit.

10. A method according to claim 1, wherein the step of, for each updated particle, estimating an expected signal parameter at the updated particle location comprises using previously collected location/signal data sets, each data set relating a location to a known signal parameter.

11. A method according to claim 1 wherein estimating an expected signal parameter at the updated particle location comprises implementing a Gaussian Process method given previously collected location/parameter data sets as recorded in the database.

12. A method according to claim 1, the method further comprising, using a global positioning system, measuring the second location of the entity, wherein the step of assigning a weight to each updated particle is performed depending on the measured second location of the entity.

13. A method of tracking an entity by monitoring at least one parameter of a signal, the signal tending to vary spatially and be generally time-invariant, the entity moving from a first location within an area to a second location within the area, the method being suitable for use when the location of the source of the signal is unknown, the method comprising:

providing a plurality of particles for use with a particle filter, each particle being associated with a first particle location, a first particle location being an estimate of the first location of the entity;

providing an estimate of the motion of the entity between the first location and the second location;

using the estimate of the motion and using the particle filter, for each particle, updating the first particle location for that particle thereby producing an updated particle location, the updated particle location being an estimate of the second location of the entity;

for each updated particle, estimating at least one expected signal parameter at the updated particle location;

measuring a signal parameter at the second location of the entity;

assigning a weight to each updated particle depending on the expected signal parameter estimated for that particle and the measured signal parameter;

estimating the second location of the entity by determining a function of the weighted updated particles; and

inputting the estimated location and measured signal parameter, as a location/parameter data set, to a database, wherein the method comprises implementing a Distributed Particle Simultaneous Localisation and Mapping algorithm.

14. A method of generating a signal parameter map of an area, the method comprising:

providing an entity moving within the area;

performing at least one iteration of a method of tracking an entity by monitoring at least one parameter of a signal, the signal tending to vary spatially and be generally time-invariant, the entity moving from a first location within an area to a second location within the area, the method being suitable for use when the location of the source of the signal is unknown, the method of tracking an entity comprising;

providing a plurality of particles for use with a particle filter, each particle being associated with a first particle location, a first particle location being an estimate of the first location of the entity;

providing an estimate of the motion of the entity between the first location and the second location;

using the estimate of the motion and using the particle filter, for each particle, updating the first particle location for that particle thereby producing an updated particle location, the updated particle location being an estimate of the second location of the entity;

for each updated particle, estimating at least one expected signal parameter at the updated particle location;

measuring a signal parameter at the second location of the entity;

assigning a weight to each updated particle depending on the expected signal parameter estimated for that particle and the measured signal parameter;

estimating the second location of the entity by determining a function of the weighted updated particles; and

inputting the estimated location and measured signal parameter, as a location/parameter data set, to a database, and thereby producing at least one measured signal parameter and an estimated location of the entity corresponding to each of the at least one measured radio signal strength; and

using the each measured signal parameter and each corresponding estimated location of the entity, determining a signal parameter map of the area.

15. A method of determining an estimation of a location within an area of a further entity, the method comprising:

measuring a signal parameter at the location of the further entity; and

using a signal parameter map of the area and the measured signal parameter at the location of the further entity, estimating the location within the area of the further entity;

wherein

the signal parameter map of the area is generated using a method of generating a signal parameter map of an area, the method of generating a signal parameter map of an area comprising;

providing an entity moving within the area;

performing at least one iteration of a method of tracking an entity by monitoring at least one parameter of a signal, the signal tending to vary spatially and be generally time-invariant, the entity moving from a first location within an area to a second location within the area, the method being suitable for use when the location of the source of the signal is unknown, the method of tracking an entity comprising;

providing a plurality of particles for use with a particle filter, each particle being associated with a first particle location, a first particle location being an estimate of the first location of the entity;

providing an estimate of the motion of the entity between the first location and the second location;

using the estimate of the motion and using the particle filter, for each particle, updating the first particle location for that particle thereby producing an updated particle location, the updated particle location being an estimate of the second location of the entity;

for each updated particle, estimating at least one expected signal parameter at the updated particle location;

measuring a signal parameter at the second location of the entity;

assigning a weight to each updated particle depending on the expected signal parameter estimated for that particle and the measured signal parameter;

estimating the second location of the entity by determining a function of the weighted updated particles; and

inputting the estimated location and measured signal parameter, as a location/parameter data set, to a database,

and thereby producing at least one measured signal parameter and an estimated location of the entity corresponding to each of the at least one measured radio signal strength; and

using the each measured signal parameter and each corresponding estimated location of the entity, determining a signal parameter map of the area.

16. Apparatus for tracking an entity, the entity moving from a previous location within an area to a current location within the area, the apparatus comprising:

a motion estimation unit;

a particle filter;

a radio scanner;

a magnetometer device; and

one or more processors;

wherein

the motion estimation unit is arranged to provide an estimate of the motion of the entity between the previous location and the current location;

the particle filter comprises a plurality of particles wherein, each particle is associated to a particle location;

the particle filter is arranged to, using the estimate of the motion, for each particle, update the particle location for that particle from a previous particle location to an updated particle location;

the previous particle location is an estimate of the previous location of the entity;

the updated particle location is an estimate of the current location of the entity;

the radio scanner is arranged to measure a radio signal strength at the current location of the entity;

the magnetometer device is arranged to measure a magnetic field parameter at the current location of the entity; and

the one or more processors are arranged to;

for each updated particle, estimate an expected radio signal strength at the updated particle location, and estimate an expected magnetic field parameter at the updated particle location;

assign a weight to each updated particle depending on the expected radio signal strength estimated for that particle and the measured radio signal strength, and depending on the expected magnetic field parameter estimated for that particle and the measured magnetic field parameter; and

estimate the current location of the entity by determining a function of the weighted updated particles.