Maneuvering target tracking method via modifying the interacting multiple model (IMM) and the interacting acceleration compensation (IAC) algorithms

• US 7,079,991 B2
• Filed: 04/22/2002
• Issued: 07/18/2006
• Est. Priority Date: 04/20/2001
• Status: Expired due to Fees

• Current Assignee: Singapore Technologies Aerospace Limited (Government of Singapore)
• Original Assignee: Singapore Technologies Aerospace Limited (Government of Singapore)
• Inventors: Lam, Chian Poh, Li, Luo Wen
• Primary Examiner(s): Rodriguez, Paul L.
• Assistant Examiner(s): GUILL, RUSSELL L
• Application Number:
  US10/127,170
• Publication Number:
  US 20030200065A1
• Time in Patent Office: 1,548 Days
• Field of Search: 703/2, 342/95, 342/96, 342/101, 342/195
• US Class Current: 703/2
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G06F 17/10 Complex mathematical operations function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544

Abstract

The present invention relates to the field of target tracking and more generally to a method employing improved algorithms, which achieve excellent tracking performance for a high-g maneuvering target. The two-model Interacting Multiple Model algorithm and the Interacting Acceleration Compensation algorithm will be modified by introducing adaptive factors through the detection of the normalized innovation squared which is chi-square probability distributed. The implementation results show that the modified algorithms can handle the target sudden maneuver better and are more accurate than the original algorithms.

Claims

1. An estimator for method tracking systems providing improved tracking of maneuvering targets comprising:

using a modified IMM algorithm in which the expressions for each of the P^J(k|k−

1), the state prediction covariance for model j, j=1 to n, is modified so that a factor, f^J, multiplies the process noise covariance, Q^J (k), for each j, j=1 to n, resulting in that the P^J(k|k−

1) are now expressed as

P^J(k|k−

1)=F^J(k)P^J(k−

1|k−

1) F^J(k)^T+f^JQ^J(k);

 

Setting the factors f^J=1 for those k such that V^J(k)S^J(k) V^J(k)^T<

C, where C is an initialized threshold, and f^J being equal to an initially chosen number greater than one for those k such that V^J(k)S^J(k) V^J(k)^T>

C, V^J(k) being the innovation and S^J(k) the innovation covariance for model j, j=1 to n,

Whereby an estimator method for tracking as an element of a tracking system, providing improved tracking of maneuvering targets.

3. The method of claim 1 wherein n=2.

4. The method of claim 1 wherein n=2, with model 1 being a constant velocity model and model 2 being a constant acceleration model.

2. An estimator method for tracking systems providing improved tracking of maneuvering targets comprising:

using a modified IAC algorithm in which the expression for P(k|k−

1), the bias-free model state prediction covariance, is modified so that a factor, f, multiplies the bias free process noise covariance, Q(k), resulting in a P(k|k−

1) that is now expressed as

P(k|k−

1)=F(k)P(k−

1|k−

1)F(k)^T+fQ(k),

And the expression for P(k|k−

1), the bias model innovation covariance, is modified so that a factor, g ,multiplies the bias model process noise covariance, Q^b(k), resulting in that P(k|k−

1) is now expressed as

P(k|k−

1)=P(k−

1|k−

1)+gG^b(k−

1)Q^b(k−

1)G^b(k−

1)^T

 

Setting the factor f=1 for those k such that V(k)S(k) V(k)^T<

C, where C is an initialized threshold, and f being equal to an initially chosen value greater than one for those k such that V(k)S(k)V(k)^T>

C, V(k) being the innovation and S(k) the innovation covariance for the bias free model,

Setting the factor g=1 for those k such that V^b(k) S^b(k) V^b(k)^T<

C, where C is an initialized threshold, and g being equal to an initially chosen value greater than one for those k such that V^b(k) S^b(k) V^b(k)^T>

C, V^b(k) being the innovation and S^b(k) the innovation covariance for the bias model,

Whereby an estimator method is used as an element of a tracking system providing improved tracking of maneuvering targets.

5. The method of claim 2 wherein said bias-free model is a constant velocity model.

6. The method of claim 2 wherein said bias model is a constant acceleration model.