Open mbr4477 opened 4 years ago
For this, I was thinking use depthcam data; create a plane detection node that accepts a downsampled point cloud and always looks for a slanted plane in the relative center of the FOV.
On Mon, Mar 2, 2020 at 8:56 AM Matthew Russell notifications@github.com wrote:
With a 2D lidar, we can't easily determine if an obstacle is a wall or a ramp. We will need to see how significant of an issue this will be.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/UKyKORA/IGV/issues/11?email_source=notifications&email_token=AAKB7DPHCDSPRAO4FAYM3STRFO3IFA5CNFSM4K7UYELKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IRWNJCA, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKB7DLV3OAMCJRPNBIWEUDRFO3IFANCNFSM4K7UYELA .
With a 2D lidar, we can't easily determine if an obstacle is a wall or a ramp. We will need to see how significant of an issue this will be.