TARGET TRACKING

• US 20130006576A1
• Filed: 03/09/2011
• Published: 01/03/2013
• Est. Priority Date: 03/15/2010
• Status: Active Grant

• Current Assignee: BAE Systems Plc
• Original Assignee: BAE Systems Plc
• Inventors: Mathews, George Morgan, Nicholson, David
• Granted Patent: US 9,305,244 B2
• Time in Patent Office: Days
• US Class Current: 702/150
• CPC Class Codes: G01S 15/66 Sonar tracking systems
• CPC Class Codes: G01S 5/16 using electromagnetic waves other than radio waves
• CPC Class Codes: G01S 5/18 using ultrasonic, sonic, or infrasonic waves
• CPC Class Codes: G06F 18/253 of extracted features
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters

Abstract

An apparatus and a method are disclosed for tracking a plurality of targets (e.g. land-based vehicles) The method can include: for a first time-step, estimating a state of each target; at a second time-step, measuring values for a state of a target; for the second time-step, estimating a state of each target using the estimated target states for the first time-step; updating the estimated target states for the second time-step using the measured values; and performing an identity management process to estimate a probability that a particular state measurement corresponds to a particular target by providing a mixing matrix, wherein an element of the mixing matrix is based on an overlap between the updated estimated target states of targets in an underlying state space of the estimated target states.

Claims

1. A method of tracking a plurality of targets, the method comprising:

for a first time-step, estimating a state of each target;

at a second time-step, measuring values for a state of a target;

for the second time-step, estimating a state of each target using the estimated target states for the first time-step;

updating the estimated target states for the second time-step using the measured values; and

performing an identity management process to estimate a probability that a particular state measurement corresponds to a particular target, wherein the identity management process comprises;

providing a mixing matrix, an element of the mixing matrix being based on an overlap between the updated estimated target states of targets in an underlying state space of the estimated target states.

2. A method according claim 1, wherein the providing of the mixing matrix comprises:

determining values for elements {circumflex over (m)}_qr, where;

{circumflex over (m)}_qr=∫

√

{square root over (p_q(x)p_r(x))}{square root over (p_q(x)p_r(x))}dx;

p_i(x)=p(x_i=x|Y) for i=q,r;

x_i is a state of an ith target at the second time-step; and

Y is a set of all state values and identity parameter values measured at time steps up to and including the second time-step; and

forming the elements into matrix {circumflex over (M)}=[{circumflex over (m)}_qr].

3. A method according to claim 2, wherein determining the mixing matrix comprises:

normalising the matrix {circumflex over (M)}.

4. A method according to claim 1, wherein the updating of the estimated target states for the second time-step using the measured values is performed using one or more of:

a multi-hypothesis tracking process, a maximum likelihood data association process, and a joint probabilistic data association process.

5. A method according to claim 1, wherein the updating of the estimated target states for the second time-step using the measured values is implemented using one of a particle filter and a Kalman filter.

6. A method according to claim 1, wherein motion of the targets to be tracked is constrained.

7. A method according to claim 6, wherein the targets are land-based vehicles.

8. A method according to claim 1, wherein a path of one of the targets is proximate to a path of at least one different target.

9. A method according to claim 8, wherein the path of one of the targets overlaps a path of at least one different target.

10. A method according to claim 1 wherein estimating a state of each target is performed with a particle filter.

11. A method according to claim 1, wherein a state of a target comprises a value for a position of the target and/or a value for a velocity of the target.

14. At least one program stored in a non-transitory medium such that when executed by a computer system having one or more processors will cause the computer system to operate in accordance with the method of claim 1.

15. A non-transitory machine readable storage medium storing the at least one program according to claim 14.

12. Apparatus for tracking a plurality of targets, the apparatus comprising:

one or more processors arranged for a first time-step, to estimate a state of each target; and

one or more sensors arranged, at a second time-step, to measure values for a state of a target, wherein the one or more processors are configured;

for the second time-step, to estimate a state of each target using the estimated target states for the first time-step;

to update the estimated target states for the second time-step using the measured values; and

to perform an identity management process to estimate a probability that a particular state measurement corresponds to a particular target;

wherein the identity management process includes providing a mixing matrix, an element of the mixing matrix being based on an overlap between the updated estimated target states of targets in an underlying state space of the estimated target states.

13. Apparatus according to claim 12, wherein elements {circumflex over (m)}_qr of the mixing matrix have the following values:

{circumflex over (m)}_qr=∫

√

{square root over (p_q(x)p_r(x))}{square root over (p_q(x)p_r(x))}dx;

where;

p_i(x)=p(x_i=x|Y) for i=q,r,

x_i is a state of an ith target at the second time-step; and

Y is a set of all state values and identity parameter values measured at time steps up to and including the second time-step; and

wherein the elements are formed into matrix {circumflex over (M)}=[{circumflex over (m)}_qr].