Background: I'm using GPS + pressure/temperature/humidity sensor (for elevation, BME280) + 10 DoF IMU (BNO055) to collect data about small geographical region (few hundret-ish square meters) by walking along specific, preselected paths, sidewalks and roads.

I intend to take the walks along the same paths several times to have as much (redundant) data to work with later as possible. My goal is to fuse GPS data with pressure data and IMU data to create three-dimensional map of the region.

GPS data rate is 1 data point per second. Pressure/temperature/humidity data rate is 1 data point per second. IMU data rate is ~40 data points per seconds.

At this point all the data is stored in three separate text files (NMEA + CSV + CSV) with time stamps for each data point. The time stamps is what currently keeps these data sets "synced".

Question: Now that I have this set of three files (with new sets added every time I repeat the measurements), what would be the best method to organize the data? Is there any data format that is intended to combine different readings (with different data rates) into one, coherent stream of data? Is such format or method of organization dictated by work flow? Which work flow would you recommend for working with GPS + pressure + IMU data sets?

I don't want to discard any data points. I don't want to lose raw data's resolution. I want to keep original data's format.

I'm not asking about data fusion itself, only about organizing collected data.

Please let me know if this is not the best place to ask this question, perhaps I should have posted it to Robotics.

  • There is no simple answer as I'm sure you found. This problem takes more time and energy during the lifetime of a project than seems necessary. What did you end up doing? – Hamid Jan 8 '19 at 18:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.