MATLAB: The target state vector is generated in spherical coordinates but how do you reconstruct in cartesian

Sensor Fusion and Tracking Toolbox

In the following example:

https://www.mathworks.com/help/releases/R2019a/fusion/examples/multi-platform-radar-detection-fusion.html

The target state vector is generated in spherical coordinates but how do you reconstruct in cartesian?

Best Answer

We define the input to the trackers as an objectDetection. The objectDetection itself is agnostic to the measurement space you’re using (spherical or rectangular, how you define the measurements), the same way that the trackers are agnostic to your state space model.
The objectDetection contains a property called MeasurementParameters, which can be used to help the filter initialization function and the measurement function to convert from measurement space to state space and vice versa, respectively.
The MeasurementParameters can be put in an array so that you can move from sensor spherical coordinates to body rectangular coordinates and then to global (north-east-down) coordinates.
These MeasurementParameters are also used by the StaticDetectionFuser to calculate the combinations of angle-only detections to fuse into fully observable detections to the tracker.

Related Solutions

MATLAB: How do we know how well the gnn reconstructs the target state vector from the truth vector captured by the sensors

There is a metrics object to calculate the tracking accuracy, which is a counterpart of the track assignment metrics object. The name of that object is: trackErrorMetrics.
You can see an example of the outcome of the track error metrices in this example. If you open the example, you can click on the helperRunTracker function to see it in use.
You can see another example of it being used in this example. Look for the lines of code in the section Track Using an EKF in Cartesian Coordinates

MATLAB: How to display a polar coordinates image in Cartesian

For a simple example, if rho is the following 2D array:

 rho =
         0    0.2000    0.4000    0.6000    0.8000    1.0000
         0    0.2000    0.4000    0.6000    0.8000    1.0000
         0    0.2000    0.4000    0.6000    0.8000    1.0000
         0    0.2000    0.4000    0.6000    0.8000    1.0000
         0    0.2000    0.4000    0.6000    0.8000    1.0000
         0    0.2000    0.4000    0.6000    0.8000    1.0000
         0    0.2000    0.4000    0.6000    0.8000    1.0000

and theta is the following 2D array:

 theta =
     0     0     0     0     0     0
    15    15    15    15    15    15
    30    30    30    30    30    30
    45    45    45    45    45    45
    60    60    60    60    60    60
    75    75    75    75    75    75
    90    90    90    90    90    90

Here is the example code:

[rho, theta] = meshgrid(0:0.2:1.0,0:15:90);
[X Y] = pol2cart(theta*pi/180,rho);
S = surf(X,Y,ones(size(X))); 
Picture = imread('autumn.tif');
set(S,'FaceColor','Texturemap','CData',Picture);
view(2);

Type

doc cart2pol

 doc function_name

in the MATLAB command window for more information of the commands used in the example code.

For more information on the surface properties, please look at the following URL:

https://www.mathworks.com/help/releases/R2017a/matlab/ref/primitivesurface-properties.html

For a transition from Cartesian coordinates to polar or cylindrical, review the following function:

http://www.mathworks.com/access/helpdesk/help/techdoc/ref/cart2pol.shtml

Best Answer

Related Solutions

MATLAB: How do we know how well the gnn reconstructs the target state vector from the truth vector captured by the sensors

MATLAB: How to display a polar coordinates image in Cartesian

Related Question