Closed thomassedlmayer closed 3 months ago
The only remark I have is that this time could be different in different sensors that are running on the same car, so putting these fields in each DetectionData could provide the flexibility to reflect such scenarios. I mean, when there are multiple lidars witch more than 1 LidarDetectionData, each could have different real world time stamps, right? Let's further discuss in the meeting :-)
That would be an additional change, if that is wanted; right now the system time is the time of the source of the sensordata, so for a single sensor it is its time, after e.g. sensor fusion it is the system time of the sensor fusion unit, etc. If someone wants to keep track of all of this across, and need to funnel this into OSI rather than measuring at multiple points, and they need something different than the detection time stamps, then another field could be added later on. A lot of ifs though.
CCB 2024-03-25: Merge as-is after conflicts resolved.
Reviewed for v3.7.0
This PR adds a system/component-specific time to OSI SensorData. This could be used to simulate discrepancies between simulation time frame and component time frame. It could also be used to preserve the component's internal timestamps when using OSI for real sensor measurement data while having the simulation timestamp's zero time set to simulation/trace start (as it is recommended by OSI).
Outcome from WP 19: Real sensor data