search

praveen-palanisamy / multiple-object-tracking-lidar

C++ implementation to Detect, track and classify multiple objects using LIDAR scans or point cloud
MIT License
796 stars 229 forks source link

Retrieval of covariance matrices for data association #19

Closed Shivarama123 closed 4 years ago

Shivarama123 commented 4 years ago

Can we get the covariance matrices for data association from your algorithm ?? .... If yes can you please let me know how to retrieve the covariance matrices.

Thanks in advance!

praveen-palanisamy commented 4 years ago

Hi @Shivarama123 ,

The covariance matrices are not currently published but, you could easily publish them to a new topic as they can be accessed from the corresponding Kalman Filter tracker object (e.g. KF0) using the following reference table:

Data Attribute to get the value
Measurement Noise Covariance KF0.measurementNoiseCov
Process Noise Covariance KF0.processNoiseCov
Posterior Error Estimate Covariance KF0.errorCovPost

Hope that helps.

Shivarama123 commented 4 years ago

Thank you so much for your reply.

And regarding data association, can you please let me know which of the covariance matrices should be required?

Only the diagonal elements of Process noise covariance will suffice or do I have to consider some other as well?

It will be a great help. Thank you so much for your time.

praveen-palanisamy commented 4 years ago

It depends on the data association algorithm/process that you are trying to implement/use. You can start with the process or measurement noise covariance.

praveen-palanisamy commented 4 years ago

Closing this as the questions were answered. Re-open if you have any related follow-up Qs or open a new item.