-
Notifications
You must be signed in to change notification settings - Fork 126
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
About perception image data id #234
Comments
Agent comment from Leon in Zendesk ticket #126459: °°° |
@dji-dev How can i get the log for the aircraft's status? |
Agent comment from Leon in Zendesk ticket #126459: °°° |
@dji-dev |
Agent comment from Leon in Zendesk ticket #126459: °°° |
Just normal offlice indoor scenario, the aircraft is put on the table, and is only power up but not take off, the probability of this issue is almost 100%. We record the image id inside callback function, and analysis the numbers afterwards. It is easy to find there is lots of repeated and lost ids. For down direction stereo is almost 20%. The CPU usage of our payload board is only 10%. |
Agent comment from Leon in Zendesk ticket #126459: °°° |
The log starts from 13:55, our recording starts from 14:00 to 14:05. i cannot remeber clearly but should be in this range. |
Agent comment from Leon in Zendesk ticket #126459: °°° |
"the auxiliary impact function" in chinese is "视觉避障功能“ ? |
Agent comment from Leon in Zendesk ticket #126459: °°° |
We close the auxiliary impact function, the lost rate drops to around 0.5%. But during the aircraft taking off, the lost rate increases to around 10% per channel. This shows the lost is getting worse when the aircraft is taking off. |
Agent comment from Leon in Zendesk ticket #126459: °°° |
When running the sdk demo, it prints a lot of this warning:
Is this means something wrong in our configuration? Will this affect the data transfer performance? |
Agent comment from Leon in Zendesk ticket #126459: °°° |
After changing to a USBtoTTL device, there is no more warning, and the lost rate drops to 0.5% for front and 5% for down. so the lost is mainly due to the uart. For the uart port, we use the uart0 from orin nx which original is 1V8, we use a levelshifter to change to 3V3. Is there anything wrong with our uart configuration? |
Agent comment from Leon in Zendesk ticket #126459: °°° |
We are using the PSDK 3.9.2. Our paltform is M3E. For perception image, we subscribe only front direction channel. We print the image data id inside the callback function, and find this id is not always continuous, for example: ... 1000 1001 1002 1002 1004 ... , which means 1002th is repeated twice and 1003th frame is lost. Average shows for every 100 frames this lost happens once. Is this correct or anything wrong with our configurations?
The text was updated successfully, but these errors were encountered: