search

TakahashiJinxu / ublox2rinex

an ROS package to convert the GNSS raw measurements from topic "/ublox_gps_node/rxmraw" in ROS bag to RINEX file. tested on California part of the UrbanLoco dataset.
20 stars 7 forks source link

question about the timestamp #2

Open huanhexiao opened 3 years ago

huanhexiao commented 3 years ago

I have converted the ublox of CA-20190828151211 to rinex format, the first epoch is :

2019 08 28 22 16 05.249000 0 12

i think the file name stands for the starting time, so the rinex time stamp is very different from this. My question is whether the time stamp in the rinex file is the real GNSS time, or the converted time from ros, as the topic "/ublox_gps_node/fix"?

TakahashiJinxu commented 3 years ago

Hello! In my opinion, since the data were collected in California in summer, the timestamp in the name of the bag is Pacific Summer Time, which is 7 hours later than UTC. And the timestamp in the rinex file is real GPS time, which is very close to UTC. Hence the timestamps are different by about 7 hours.

I have converted the ublox of CA-20190828151211 to rinex format, the first epoch is :

2019 08 28 22 16 05.249000 0 12

i think the file name stands for the starting time, so the rinex time stamp is very different from this. My question is whether the time stamp in the rinex file is the real GNSS time, or the converted time from ros, as the topic "/ublox_gps_node/fix"?

huanhexiao commented 3 years ago

Hello! In my opinion, since the data were collected in California in summer, the timestamp in the name of the bag is Pacific Summer Time, which is 7 hours later than UTC. And the timestamp in the rinex file is real GPS time, which is very close to UTC. Hence the timestamps are different by about 7 hours.

I have converted the ublox of CA-20190828151211 to rinex format, the first epoch is :

2019 08 28 22 16 05.249000 0 12

i think the file name stands for the starting time, so the rinex time stamp is very different from this. My question is whether the time stamp in the rinex file is the real GNSS time, or the converted time from ros, as the topic "/ublox_gps_node/fix"?

I think you are right, it is reasonable.