using T265 with mobile robots
I have got T265 camera but I'm confused about how to use it for mobile robots. Please, correct me.
As I understand, purpose of T265 is only in better localization, for replacing imu.
So, we had slam algorithms using imu, now we have slam algorithms using T265 with better robustness.
Is it correct?
And the second question, what is better for using in mapping: d435 depth camera or computed depth from T265 fisheye images? What is supposed to use?
-
Intel has an excellent long presentation on YouTube about using T265 with mobile robots.
https://youtube.com/watch?v=62vm0_RZ1nU
The section introducing use of a D435 with the T265 starts at 17 minutes 15 seconds into the video.
T265 can provide very high quality tracking data, though it does not have depth tracking by default - there is a Python script for getting depth from the pair of fisheye cameras, though it is not as accurate as pairing the T265 with a real depth camera such as the D435 (it does not need to be a D435i, as the T265 already has an IMU). Using a dedicated depth camera with the T265 is better than using fisheye.
https://github.com/IntelRealSense/librealsense/blob/master/wrappers/python/examples/t265_stereo.py
Please sign in to leave a comment.
Comments
1 comment