I need a function written in c++ using google Ceres, to solve an optimisation problem.
I have a multi-sensor tracking setup as below:
Three sensors, each streaming transform data as position xyz and rotation xyzw.
The sensors are locked together.
I know the extrinsic parameters of the sensors in relation to each other, as in the image below.
As the sensors stream data, one or another of them with give data that is incorrect, giving spikes or dropouts in the data streams.
What I need, is a function that takes in the three data streams, and uses the extrinsic parameters as a constraint, returning one single transform pose as xyz and xyzw that ignores the dropped or incorrect data spikes.
I would like the functionality to also pass in a 'weight' or confidence for each sensor.
I would like the function to be easily expandable to more than three sensors .
Please ONLY bid on this project if you have experience with Ceres / bundle adjustment / computer vision.
Please ask if i have not been clear.
I will supply test data, as well as extrinsics, as csv files.
10 фрилансеров(-а) в среднем готовы выполнить эту работу за $521
I hope to see you in chat. I am an experienced python machine learning developer with full-stack knowledge and career. I'm sure I can do this perfectly. Thanks for your kind attention.