Ask Your Question
1

performance of VSLAM

asked 2011-11-17 17:53:07 -0600

Divya gravatar image

Hello all,

Does anybody know why the VSLAM node in ROS doesn't perform too well when the robot is not moving exceptionally slow or while it takes a sharp turn?

I know there are some limitations to using images to determine position and generating a map.. but if someone could pinpoint the exact shortcomings to the algorithm used in the VLSAM package (or at least tell me what algorithm has been followed in this case), it would be reallly helpful!

I am new to SLAM and ROS.. this may be a very silly question, but it'd really help me learn!

Thanks! Divya

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted
4

answered 2011-11-18 02:47:02 -0600

Mac gravatar image

When the camera is moving quickly (particularly when it's rotating), every stage of VSLAM gets harder. It's harder to extract features, because of camera blur. It's harder to match features, because it's hard to initialize well, and because the viewpoint has changed, fouling up which features are visible. The geometric constraints that can help start to fall apart in the monocular case with rotations.

In general, VSLAM (and monocular VSLAM in particular) are very tricky; one of the reasons the VSLAM package isn't getting attention is because it's very, very hard to make VSLAM systems work reliably for everyone; a package that works great on a PR2, in a particular environment, may need to be re-tuned arbitrarily for your robot in your building; this just isn't something where general-purpose code that Just Works (or even Sorta Just Kinda Works) can be shipped.

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

Stats

Asked: 2011-11-17 17:53:07 -0600

Seen: 194 times

Last updated: Nov 18 '11