During my ground tests, I found that the velocity measurement (in North-East-Down format) of UBlox GPS is not accurate as I expected. The experiment is like this:
1. fix a UBlox GPS unit on a trolley, which is initially stationery;
2. move the trolley forward with a relatively constant speed (about 1 m/s);
3. Stop and check data.
I found that the recorded velocity is liked a bounded but random signal, osillating in the range of -1~1 m/s (yes, I observed even the -1 value at some points). May I know that have you guys encounter a similar problem, and how did you solve it?
If possible, can someone tell me how about the accuracy of the MediaTek GPS's velocity measurement. I roughly remember that it only gives course velocity. If so, based on the measured Euler angles, can I have the North-East-Down velocity vectors?
Many thanks for the help and suggestions from all of you!!