DESC:TECHNICAL FIELD
[0001] The present disclosure relates generally to tracking a plurality of objects. The disclosure, more particularly, relates to tracking maneuvering objects.

BACKGROUND
[0002] Tracking of an object with high maneuvering capability is a challenging task. The tracking methodology must be evolved in a way to identify the kinematics of the object to track the object with an improved accuracy. Typically, tracking of an object involves estimating the velocity and/or acceleration of the object by implementing different filtering methods, also referred to as filter models such as Constant Velocity (CV) model, Constant Acceleration (CA) model, and additionally Constant Turn Rate (CTR) model. By considering only a Constant Velocity (CV) model, a linear object can be tracked precisely. With a Constant Acceleration (CA) model, an accelerating object can be tracked smoothly. A Constant Turn Rate (CTR) model helps in tracking a maneuvering object. These individual models work best independently. When an object moves on different paths, and with different kinematics, each of these models fails in tracking the object with desired accuracy.

[0003] US5214433A mentions tracking of maneuvering and non-maneuvering targets in the presence of stochastic acceleration. US5214433A utilizes a two-stage Kalman estimator. In a first stage, a bias-free filter providing target position and velocity estimates, and in a second stage, a bias filter providing estimates of target acceleration.

[0004] US5325098A mentions tracking of a maneuvering target, wherein a first filter estimates a partial system state at a time (k) in terms of target position measurements, and a plurality of second filters are each provided with an acceleration model hypothesis from a prior time (k-1) free of position and velocity constraints. Each second filter generates an acceleration estimate at time (k) and a likelihood at time (k) that the acceleration model hypothesis is correct. The likelihoods from the second filters are summed to generate a probability vector at time (k). A third interaction mixing filter generates the acceleration model hypotheses from prior time (k-1) using the probability vector from prior time (k-1) and the acceleration estimates from prior time (k-1).

[0005] US5999117A mentions detecting and tracking of turns of a maneuvering target by determining first and second radar information of the maneuvering target. The first and second radar information and a set of target speeds are used to determine a set of turn radii for the maneuvering target.

[0006] However, none of the prior arts mentioned in the aforesaid documents are capable of tracking high maneuvering objects. Further, existing tracking techniques as known in the art are also incapable of tracking of high maneuvering objects, as such techniques are restricted to implementation of any one of the aforesaid filter models.

[0007] Therefore, there is felt a need for developing a tracking technique capable of tracking high maneuvering objects accurately and effectively.

SUMMARY
[0008] This summary is provided to introduce concepts related to an object tracking system and method thereof. This summary is neither intended to identify essential features of the present disclosure nor is it intended for use in determining or limiting the scope of the present disclosure.

[0009] In accordance with the present disclosure, there is provided a tracking technique in the form of an object tracking method and system for tracking high maneuvering objects by combining a Constant Velocity (CV) model for tracking a linear object, at least one Constant Acceleration (CA) model for tracking an accelerating object, and at least one Constant Turn Rate (CTR) for tracking a maneuvering object, in an Interacting Multiple Model (IMM). Selecting the model combination in the Interacting Multiple Model (IMM) is one of the prime objectives. Specifically, the present disclosure discloses an implementation of an IMM comprising one CV, one CA, and three CTR’s for tracking the high maneuvering objects.

BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS
[0010] The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to reference like features and modules.

[0011] Figure 1 illustrates a system diagram for implementing the tracking technique for tracking a high maneuvering object according to the present disclosure.

[0012] Figure 2 illustrates an operational flow diagram depicting the tracking technique for tracking a high maneuvering object according to the present disclosure.

[0013] Figure 3 illustrates a gating diagram for predicting the position of the high maneuvering object being tracked according to the implementation of the tracking technique illustrated in Figure 2.

[0014] Figure 4 illustrates a graphical representation depicting tracking accuracy for an object with 4g maneuver, according to the implementation of the tracking technique illustrated in Figure 2.

[0015] Figure 5 illustrates a graphical representation depicting tracking accuracy for an object with 6g maneuver, according to implementation of the tracking technique illustrated in Figure 2.

[0016] Figure 6 illustrates a graphical representation depicting an object drop when an Interacting Multiple Model (IMM) is with one Constant Velocity (CV) model, three Constant Acceleration (CA) models, and one Constant Turn Rate (CTR) model.

[0017] Figure 7 illustrates a graphical representation depicting smooth tracking when an Interacting Multiple Model (IMM) is with one Constant Velocity (CV) model, one Constant Acceleration (CA) model, and three Constant Turn Rate (CTR) models, according to the implementation of the tracking technique illustrated in Figure 2.

[0018] Figure 8 illustrates a flow chart depicting the steps involved for implementing the tracking technique for tracking a high maneuvering object as illustrated in Figure 1 and 2.

[0019] It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative methods embodying the principles of the present disclosure. Similarly, it will be appreciated that any flow charts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

DETAILED DESCRIPTION
[0020] The present disclosure describes about a tracking technique implemented through various embodiments of an object tracking system and method thereof, for tracking high maneuvering objects.

[0021] In the following description, for purpose of explanation, specific details are set forth in order to provide an understanding of the present disclosure. It will be apparent, however, to one skilled in the art that the present disclosure may be practiced without these details. One skilled in the art will recognize that embodiments of the present disclosure, some of which are described below, may be incorporated into a number of systems.

[0022] However, the systems and methods are not limited to the specific embodiments described herein. Further, structures and devices shown in the figures are illustrative of exemplary embodiments of the presently disclosure and are meant to avoid obscuring of the presently disclosure.

[0023] It should be noted that the description merely illustrates the principles of the present disclosure. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described herein, embody the principles of the present disclosure. Furthermore, all examples recited herein are principally intended expressly to be only for explanatory purposes to help the reader in understanding the principles of the disclosure and the concepts contributed by the inventor to furthering the art and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosure, as well as specific examples thereof, are intended to encompass equivalents thereof.

[0024] In accordance with an exemplary embodiment, a method for tracking one or more maneuvering objects comprises: validating at least one plot area having one or more tracks, correlating the validated plot area with said one or more tracks of the plot area, sensing first measurement data related to the objects, tracking linear position data from the first measurement data, and generating linear measurement data, tracking acceleration data from the first measurement data, and generating acceleration measurement data, tracking speed and heading angle data from the first measurement data, and generating constant turn rate measurement data. Further, the method includes combining the linear measurement data, the acceleration measurement data, and the constant turn rate measurement data, and generating second measurement data, wherein second measurement data includes an initial state vector and an initial covariance matrix of the objects, predicting a state vector and covariance vector of the objects, and generating a prediction value and computing difference between the predicted value and the second measurement data, by performing interacting multiple model filtering to determine a state estimation vector and an estimation covariance matrix of said objects.

[0025] In an aspect, the method includes computing weightages of the linear measurement data, the acceleration measurement data, and the constant turn rate measurement data.

[0026] In an aspect, the method includes identifying target position measurement data; and generating a transition probability matrix in each cartesian coordinate of the initial state vector and the initial covariance matrix.

[0027] In an aspect, the method includes includes filtering weightages of each the linear measurement data, the acceleration measurement data, and the constant turn rate measurement data, using the transition probability matrix.

[0028] In an aspect, the method includes generating a report for at least one target from the target position measurement data.

[0029] In an aspect, the method includes storing, in a database, the predicted value, the first measurement data, the acceleration measurement data, and the constant turn rate measurement data, the second measurement data, the target position measurement data, and the report.

[0030] In an aspect, the method includes measuring distance between the plot area and the one or more tracks, and identifying the least distance from the plot area and the one or more tracks.

[0031] In another aspect, the one or more tracks of the plot area are selected within a specified range of the plot area, the range is identified by using techniques including a Root Mean Square (RMS) range error technique, an RMS Azimuth error technique, and an RMS Elevation error technique.

[0032] In another aspect, the step of predicting the state vector and the covariance vector includes re-predicting the one or more tracks within the specified range.

[0033] In an aspect, the method includes updating the objects when one measurement is gated to one track.

[0034] In another aspect, the step of updating includes updating the objects when one plot area falls within a plurality of gates, and computing the distance between the plot and the track and finding the least distance.

[0035] In another aspect, the step of updating includes updating the objects when a plurality of plot areas falls within the track, and computing the distance between the plot and the track and finding the least distance.

[0036] In another aspect, the step of updating includes updating the objects when the plurality of objects falls within a plurality of tracks.

[0037] In another aspect, updating the objects when the plurality of objects falls within a plurality of tracks includes associating the updated objects with the plurality of tracks using a Global Nearest Neighbourhood (GNN) technique.

[0038] In another aspect, the predicted value is center region of a gate.

[0039] In an aspect, the method includes determining size of the gate based on parameters including measurement and prediction covariance, object maneuverability, and object revisit time.

[0040] In another aspect, the step of predicting the state vector and the covariance vector of the objects includes finding the delta time based on the initial state vector; predicting the state and covariance from each of the models; computing IMM predicted states by using filter weight and filter state of each model; and computing the predicted time and wait for the next measurement data.

[0041] In accordance with another exemplary embodiment, an object tracking system (100) for tracking one or more manoeuvring objects comprises a memory (110) configured to store pre-determined rules, a processor (112) configured to cooperate with the memory (110) to receive the pre-determined rules, the processor (112) configured to generate system processing commands. The system includes a validation module (114) configured to validate at least one plot area having one or more tracks, a correlation module (116) configured to cooperate with the validation module (114), the correlation module (116) configured to correlate the validated plot area with the one or more tracks of the plot area, a sensing module (118) configured to cooperate with the correlation module (116), the sensing module (118) configured sense first measurement data related to the objects from the correlated plot area, an Interacting Multiple Model (IMM) unit (120) configured to cooperate with the sensing module to receive the first measurement data. Further, the IMM Unit includes at least one constant velocity (CV) module (122) configured to track linear position data from the first measurement data, and generate linear measurement data, at least one constant acceleration (CA) module (124) configured to track acceleration data from the first measurement data, and generate acceleration measurement data, at least one constant turn rate (CTR) module (126) configured to track speed and heading angle data from the first measurement data, and generate constant turn rate measurement data, and a combiner (128) configured to cooperate with the CV module (122), the CA module (124), and the CTR module (126), the combiner (128) configured to combine the linear measurement data, the acceleration measurement data, and the constant turn rate measurement data, and generate second measurement data, wherein second measurement data includes an initial state vector and an initial covariance matrix of the objects. Further, the system includes a prediction module (130) configured to predict a state vector and covariance vector of the objects, and further configured to generate a predicted value; and a computation module (132) configured to compute difference between the predicted value, and the second measurement data, by performing interacting multiple model filtering to determine a state estimation vector and an estimation covariance matrix of the objects.

[0042] In an aspect, the sensing module (118) includes a plurality of sensors, including proximity sensors, image sensors, and motion detection sensors.

[0043] In an aspect, the computation module (132) is configured to compute weightages of the linear measurement data, the acceleration measurement data, and the constant turn rate measurement data, generated by the constant velocity (CV) module (122), constant acceleration (CA) module (124), and constant turn rate (CTR) module (126), respectively.

[0044] In an aspect, the IMM unit (120) includes a filtering module (134), the filtering module (134) configured to identify target position measurement data, and generate a transition probability matrix in each cartesian coordinate of the initial state vector and the initial covariance matrix, to allow the constant velocity (CV) module (122), the constant acceleration (CA) module (124), and the constant turn rate (CTR) module (126) probabilities in each coordinates of initial covariance matrix.

[0045] In an aspect, the filtering module (134) is configured to filter weightages of each the linear measurement data, the acceleration measurement data, and the constant turn rate measurement data, using the transition probability matrix in the IMM unit (120).

[0046] In an aspect, the filtering module (134) includes an IMM filter, configured to filter data from the constant velocity (CV) module, the constant acceleration (CA) module, and the constant turn rate (CTR) module to interact through mixing of state estimates to track an arbitrary object trajectory.

[0047] In another aspect, the system (100) includes a report generation module (136), the report generation module (136) configured to generate a report for at least one target from the target position measurement data.

[0048] In another aspect, the system (100) includes a database (140) configured to store the predicted value, the first measurement data, the acceleration measurement data, and the constant turn rate measurement data, the second measurement data, the target position measurement data, and the report.

[0049] In another aspect, the system (100) includes a distance measurement module (138) configured to measure distance between the plot area and the one or more tracks, and further configured to identify the least distance from the plot area and the one or more tracks.

[0050] In another aspect, the system (100) includes an updation module (142) configured to update the objects when one measurement is gated to one track.

[0051] In an aspect, the updation module (142) is configured to update the objects when one plot area falls within a plurality of gates, and compute the distance between the plot and the track and find the least distance.

[0052] In an aspect, the updation module (142) is configured to update the objects when a plurality of plot areas falls within the track, and compute the distance between the plot and the track and find the least distance.

[0053] In an aspect, the updation module (142) is configured to update the objects when the plurality of objects falls within a plurality of tracks.

[0054] In an aspect, the updation module (142) includes a data association module (143) configured to associate the updated objects with the plurality of tracks using a Global Nearest Neighbourhood (GNN) technique.

[0055] In another aspect, the system (100) includes a size determination module (144) configured to determine size of the gate based on parameters including measurement and prediction covariance, object maneuverability, and object revisit time.

[0056] In an exemplary implementation of the method and system as disclosed herein above, the object tracking is carried out by combining one Constant Velocity (CV) model, one Constant Acceleration (CA) model, and three Constant Turn Rate (CTR) models in an Interacting Multiple Model (IMM). The number of CTR models is more in the IMM to handle tracking of a high maneuvering object. Additionally, each of the three constant turn rate (CTR) models have different uncertainty factor values to handle different levels of maneuvers of the objects. These uncertainty values can be fine-tuned in the real field object tracking. Table 1 below illustrates accuracy of the filter models in terms of range error, azimuth error and elevation error resulting from the implementation of the models at bank angles of zero degree and sixty degrees each. The error values provided are root mean square (RMS) values of the range, azimuth and elevation errors.

Model Bank Angle
(Input in Degree) RMS Range Error (m) RMS Azimuth Error (milRad) RMS Elevation Error (milRad)
CV Model 0 0.85 2.53 2.62
60 143.3 15.13 35.70
CA Model 0 5.05 3.48 2.41
60 19.12 5.28 3.20
CTR Model 0 1.30 2.17 4.65
60 7.12 3.68 2.41
IMM 0 2.19 2.68 4.43
60 2.00 2.44 1.96

Table 1: Accuracy of the filter models

[0057] The implementation of the method and system implements the tracking technique according to the operational flow as illustrated in Figure 1, wherein the correlation operation is implemented to choose the best observation to track association.

[0058] A gating operation is implemented for eliminating unlikely observation-to-track pairing. Figure 2 illustrates a gating diagram for predicting the position of the high maneuvering object being tracked, by eliminating unlikely observation-to-track pairing. In gating operation, all the tracks within the specified values of Range, Azimuth and Elevation are picked, a track-to-current plot time retaining the gate size is re-predicted, and a measurement that falls within a gate of picked-up and re-predicted track is processed further. The gating operation gives rise to the following four conditions:
(a) A single measurement is gated to a single track – in this condition an assignment can be immediately made for updating the target;
(b) A single plot falls within multiple gates – in this condition the statistical distance between the plot and the track is calculated and the least distance is assigned for updating the target;
(c) Multiple plots are available for a single track – in this condition the statistical distance between the plot and the track is calculated and the least distance is assigned for updating the target, while the rest of the plots are used for initiating new tracks; and
(d) Multiple plots fall within multiple tracks are available – in this condition, conflict arises which is solved by data association.

[0059] The operation of data association is implemented by taking the output of gating operation and making final measurement-to-track association. Particularly, in the condition where multiple plots fall within multiple gates of tracks, a global nearest neighborhood (GNN) technique is applied on the output of the gating operation. As can be seen from figure 2, the predicted values define the center of the gate region. The measurement O2 falls within the gates of all three tracks, and the GNN technique is then applied on the measurement O2. Typically, GNN technique deals with Munkre’s method which finds the least global cost and associates accordingly.

[0060] In an aspect, the size of a gate depends on the following factors: measurement and prediction covariance, target maneuverability and target revisit time. In initial phase, a gate is typically bigger than one, and thereafter the gate is updated more frequently.

[0061] The operations of updating the track, processing track filter and predicting the track of the object are implemented after a first measurement data is received from a sensor configured to detect the position of the high maneuvering object in a plot area. The first measurement data is treated as the filtered data which is same as the received data. From the next measurement as a result of confirmation, each filter model is initialized and hence the IMM is also initialized. The measurement data reception, track data updation, and track prediction steps are explained hereafter.

• Predict state and predict covariance
Firstly, the state and covariance of the object being tracked is predicted through different matrices.
Transition Matrix- A (9X9)
Identity Matrix – I (3x3)
1 0 0
0 1 0
0 0 1

Transition Matrix for CV Model
I T*I 0
0 I 0
0 0 0

Where, T is update rate
Transition Matrix for CA Model
I T*I T2/2*I
0 I T*I
0 0 I

Transition Matrix for CTR model
I sin (? *T)/? *I (1-cos (? *T))/?2*I
0 cos(? *T)*I sin (? *T)/? *I
0 -? *sin(? *T)*I cos(? *T)*I

? = sqrt((ax2 + ay2+az2)/(vx2+vy2+vz2))
Where, ? is Angular Velocity.

Plant Noise Covariance matrix - Q (9X9)
Plant Noise Covariance matrix for the CV Model
T3/3*I T2/2*I 0
T2/2*I T*I 0
0 0 0

Plant Noise Covariance matrix for the CA Model
T5/20*I T4/8*I T3/6*I
T4/8*I T3/3*I T2/2*I
T3/6*I T2/2*I T*I

Plant Noise Covariance matrix for the CTR Model
T5/20*I T4/8*I T3/6*I
T4/8*I T3/3*I T2/2*I
T3/6*I T2/2*I T*I

Predicted State of the object - Sp
Sp = A * Sf

Predicted Covariance of the object - Pp
Pp = A * Pf * AT + Q*Qn

Where, Qn is uncertainty factor associated with each model;
QCV = 5,
QCA = 10,
QCTR1 = 20,
QCTR2 = 40,
QCTR3 = 60.

• Mixed probability with respect to filter weightage
Next, a mixed probability matrix is computed using a transition probability matrix and filter weights of the individual models in the IMM.

Transition probability
TransitionProb(5x5) = diag(0.99), others(0.0025)
0.99 0.0025 0.0025 0.0025 0.0025
0.0025 0.99 0.0025 0.0025 0.0025
0.0025 0.0025 0.99 0.0025 0.0025
0.0025 0.0025 0.0025 0.99 0.0025
0.0025 0.0025 0.0025 0.0025 0.99

mixedProb (5x5) = diag(fWt)*transitionProb(5x5)
mixedProb(5x5) = mixedProb*diag(1/ones(1,5)*mixedProb)

• Mixed filter state and Mixed filter covariance
Thereafter, a mixed filter state and mixed filter covariance are calculated using filter state matrix and mixed probability matrix.

• Predicted state and Predicted covariance
Finally, a predicted state and predicted covariance in the IMM are computed using individual model’s predicted state and predicted covariance.

[0062] Thus, the operations of updating the track, processing track filter and predicting the track of the object, are implemented by:
determining the delta time based on the track state;
predicting the state and covariance from each of the models;
computing IMM predicted states by using filter weight and filter state of each model; and
computing the predicted time and waiting for the next measurement from the radar.

[0063] Figure 3 illustrates a graphical representation depicting tracking accuracy for an object with 4g maneuver achieved by the implementation of the tracking technique with the IMM comprising one CV, one CA, three CTR filter models in accordance with the method and system as disclosed herein above. In Figure 1, g-turn = 75.53 degree for 4g turn maneuver.

[0064] Figure 4 illustrates a graphical representation depicting tracking the accuracy for an object with 6g maneuver, achieved by the implementation of the tracking technique with the IMM comprising one CV, one CA, three CTR filter models in accordance with the method and system as disclosed herein above. In Figure 2, g-turn= 83.0 degree, for 8g turn maneuver.

[0065] Figure 5 illustrates a graphical representation depicting an object drop when an IMM is with one CV model, three CA models, and one CTR model.

[0066] Figure 6 illustrates a graphical representation depicting smooth tracking when the IMM comprises one CV model, one CA model, and three CTR models, according to the implementation of the method and system as disclosed herein above.

[0067] In an exemplary embodiment, the object tracking system is configured to perform smooth tracking up to 8g. The object tracking system improves position accuracy of the object.

[0068] Table 2 below illustrates tracking percentages when the IMM is with one CV, three CA, one CTR, filter models; when the IMM comprises one CV, one CA, three CTR filter models.

Scenario Tracking percentage (CV-CA-CA-CA-CTR) Tracking percentage (CV-CA-CTR-CTR-CTR)
8g Maneuver by Object (Negative) at 25 Km on startup of trajectory 33 97
8g Maneuver by Object (Positive) at 25 Km on startup of trajectory 50 50
8g Maneuver by Object (Negative) at 25 Km on startup of trajectory 50 95
8g Maneuver by Object (Positive) at 25 Km on startup of trajectory 95 95
8g Maneuver by Object (Positive) at 25 Km on startup of trajectory 50 96
Inward from 25 Km and Object of 8g at 23 Km 33 95
Inward from 25 Km and Object of 8g at 15 Km 25 96
Inward from 25 Km and Object of 6g at 20 Km and Linear up to 15 Km and maneuver of 8g at 15 Km 25 94
Inward from 25 Km and Maneuver of 8g at 10 Km 33 98
Inward from 15 Km and Maneuver of 8g at 5 Km 25 98
Overall % 41.9 91.4

Table 2: Tracking percentage of IMM (1 CV, 3 CA, 1 CTR) and IMM (1 CV, 1 CA, 3 CTR)

[0069] It can be clearly seen from Table 2 that object tracking carried out by implementing the tracking technique with the Interacting Multiple Model (IMM) comprising one Constant Velocity (CV) model, one Constant Acceleration (CA) model, and three Constant Turn Rate (CTR) models, in accordance with the method and system as disclosed herein above, results in significantly accurate and efficient tracking of high maneuvering objects.

[0070] The foregoing description of the disclosure has been set merely to illustrate the disclosure and is not intended to be limiting. Since modifications of the disclosed embodiments incorporating the substance of the disclosure may occur to person skilled in the art, the disclosure should be construed to include everything within the scope of the invention.

,CLAIMS:

1. A method for tracking one or more maneuvering objects, the method comprising:
validating at least one plot area having one or more tracks;
correlating said validated plot area with said one or more tracks of said plot area;
sensing first measurement data related to said objects;
tracking linear position data from said first measurement data, and generating linear measurement data;
tracking acceleration data from said first measurement data, and generating acceleration measurement data;
tracking speed and heading angle data from said first measurement data, and generating constant turn rate measurement data;
combining said linear measurement data, said acceleration measurement data, and said constant turn rate measurement data, and generating second measurement data, wherein second measurement data includes an initial state vector and an initial covariance matrix of said objects;
predicting a state vector and covariance vector of said objects, and generating a prediction value; and
computing difference between said predicted value and said second measurement data, by performing interacting multiple model filtering to determine a state estimation vector and an estimation covariance matrix of said objects.

2. The method as claimed in claim 1, wherein said method includes computing weightages of said linear measurement data, said acceleration measurement data, and said constant turn rate measurement data.

3. The method as claimed in claim 1, wherein said method includes:
identifying target position measurement data; and
generating a transition probability matrix in each cartesian coordinate of said initial state vector and said initial covariance matrix.
4. The method as claimed in claims 2 and 3, wherein said method includes filtering weightages of each said linear measurement data, said acceleration measurement data, and said constant turn rate measurement data, using said transition probability matrix.

5. The method as claimed in claims 1 and 3, wherein said method includes generating a report for at least one target from said target position measurement data.

6. The method as claimed in claim 1, wherein said method includes storing, in a database, said predicted value, said first measurement data, said acceleration measurement data, and said constant turn rate measurement data, said second measurement data, said target position measurement data, and said report.

7. The method as claimed in claim 1, wherein said method includes measuring distance between said plot area and said one or more tracks, and identifying the least distance from said plot area and said one or more tracks.

8. The method as claimed in claim 1, wherein said one or more tracks of said plot area are selected within a specified range of said plot area, said range is identified by using techniques including a Root Mean Square (RMS) range error technique, an RMS Azimuth error technique, and an RMS Elevation error technique.

9. The method as claimed in claims 1 and 8, wherein said step of predicting the state vector and the covariance vector includes re-predicting said one or more tracks within said specified range.

10. The method as claimed in claim 1, wherein said method includes updating said objects when one measurement is gated to one track.

11. The method as claimed in claims 1 and 10, wherein said step of updating includes updating said objects when one plot area falls within a plurality of gates, and computing the distance between the plot and the track and finding the least distance.
12. The method as claimed in claims 1 and 10, wherein said step of updating includes updating said objects when a plurality of plot areas falls within said track, and computing the distance between the plot and the track and finding the least distance.

13. The method as claimed in claims 1 and 10, wherein said step of updating includes updating said objects when said plurality of objects falls within a plurality of tracks.

14. The method as claimed in claim 13, wherein updating said objects when said plurality of objects falls within a plurality of tracks includes associating said updated objects with said plurality of tracks using a Global Nearest Neighbourhood (GNN) technique.

15. The method as claimed in claims 1 and 11, wherein said predicted value is centre region of a gate.

16. The method as claimed in claim 1, wherein said method includes determining size of said gate based on parameters including measurement and prediction covariance, object manoeuvrability, and object revisit time.

17. The method as claimed in claim 1, wherein the step of predicting said state vector and said covariance vector of said objects includes:
finding the delta time based on said initial state vector;
predicting the state and covariance from each of said models;
computing IMM predicted states by using filter weight and filter state of each model; and
computing the predicted time and wait for the next measurement data.

18. An object tracking system (100) for tracking one or more maneuvering objects, said system (100) comprising:
a memory (110) configured to store pre-determined rules;
a processor (112) configured to cooperate with said memory to receive said pre-determined rules, said processor configured to generate system processing commands;
a validation module (114) configured to validate at least one plot area having one or more tracks;
a correlation module (116) configured to cooperate with said validation module (114), said correlation module (116) configured to correlate said validated plot area with said one or more tracks of said plot area;
a sensing module (118) configured to cooperate with said correlation module (116), said sensing module (118) configured sense first measurement data related to said objects from said correlated plot area;
an Interacting Multiple Model (IMM) unit (120) configured to cooperate with said sensing module (118) to receive said first measurement data, said IMM unit (120) includes:
at least one constant velocity (CV) module (122) configured to track linear position data from said first measurement data, and generate linear measurement data;
at least one constant acceleration (CA) module (124) configured to track acceleration data from said first measurement data, and generate acceleration measurement data;
at least one constant turn rate (CTR) module (126) configured to track speed and heading angle data from said first measurement data, and generate constant turn rate measurement data; and
a combiner (128) configured to cooperate with said CV module (122), said CA module (124), and said CTR module (126), said combiner (128) configured to combine said linear measurement data, said acceleration measurement data, and said constant turn rate measurement data, and generate second measurement data, wherein second measurement data includes an initial state vector and an initial covariance matrix of said objects;
a prediction module (130) configured to predict a state vector and covariance vector of said objects, and further configured to generate a predicted value; and
a computation module (132) configured to compute difference between said predicted value, and said second measurement data, by performing interacting multiple model filtering to determine a state estimation vector and an estimation covariance matrix of said objects.

19. The system (100) as claimed in claim 18, wherein said sensing module (118) includes a plurality of sensors, including proximity sensors, image sensors, and motion detection sensors.

20. The system (100) as claimed in claim 18, wherein said computation module (132) is configured to compute weightages of said linear measurement data, said acceleration measurement data, and said constant turn rate measurement data, generated by said constant velocity (CV) module, constant acceleration (CA) module, and constant turn rate (CTR) module, respectively.

21. The system (100) as claimed in claim 18, wherein said IMM unit (120) includes a filtering module (134), said filtering module (134) configured to identify target position measurement data, and generate a transition probability matrix in each cartesian coordinate of said initial state vector and said initial covariance matrix, to allow said constant velocity (CV) module (122), said constant acceleration (CA) module (124), and said constant turn rate (CTR) module (126) probabilities in each coordinates of initial covariance matrix.

22. The system (100) as claimed in claims 20 and 21, wherein said filtering module (134) is configured to filter weightages of each said linear measurement data, said acceleration measurement data, and said constant turn rate measurement data, using said transition probability matrix in said IMM unit (120).

23. The system (100) as claimed in claim 21, wherein said filtering module (134) includes an IMM filter configured to filter data from said constant velocity (CV) module, said constant acceleration (CA) module, and said constant turn rate (CTR) module to interact through mixing of state estimates to track an arbitrary object trajectory.

24. The system (100) as claimed in claim 18, wherein said system includes a report generation module (136), said report generation module (136) configured to generate a report for at least one target from said target position measurement data.

25. The system (100) as claimed in claim 18, wherein said system includes a database (140) configured to store said predicted value, said first measurement data, said acceleration measurement data, and said constant turn rate measurement data, said second measurement data, said target position measurement data, and said report.

26. The system (100) as claimed in claim 18, wherein said system includes a distance measurement module (138) configured to measure distance between said plot area and said one or more tracks, and further configured to identify the least distance from said plot area and said one or more tracks.

27. The system (100) as claimed in claim 18, wherein said system includes an updation module (142) configured to update said objects when one measurement is gated to one track.

28. The system (100) as claimed in claims 18 and 27, wherein said updation module (142) is configured to update said objects when one plot area falls within a plurality of gates, and compute the distance between the plot and the track and find the least distance.

29. The system (100) as claimed in claims 18 and 27, wherein said updation module (142) is configured to update said objects when a plurality of plot areas falls within said track, and compute the distance between the plot and the track and find the least distance.

30. The system (100) as claimed in claims 18 and 27, wherein said updation module (142) is configured to update said objects when said plurality of objects falls within a plurality of tracks.

31. The system (100) as claimed in claim 30, said updation module (142) includes a data association module (143) configured to associate said updated objects with said plurality of tracks using a Global Nearest Neighbourhood (GNN) technique.

32. The system (100) as claimed in claim 18, wherein system includes a size determination module (144) configured to determine size of said gate based on parameters including measurement and prediction covariance, object maneuverability, and object revisit time.