search

MAPIRlab / rf2o_laser_odometry

Estimation of 2D odometry based on planar laser scans. Useful for mobile robots with innacurate base odometry. For full description of the algorithm, please refer to: Planar Odometry from a Radial Laser Scanner. A Range Flow-based Approach. ICRA 2016 Available at: http://mapir.isa.uma.es/mapirwebsite/index.php/mapir-downloads/papers/217
GNU General Public License v3.0
372 stars 210 forks source link

rf2o with outdoor navigation #26

Closed mhaboali closed 4 years ago

mhaboali commented 4 years ago

Recently, I decided to use rf2o to replace the wheel encoder in my robot_localization EKF node. I'm using Hokuyo as the source of a laser scan topic.

I have tested my system indoor and it works perfectly. But once I moved to the outdoor side, I got this error from rf2o node ([rf2o] ERROR: Eigensolver couldn't find a solution. Pose is not updated)

After searching about that error, I found this commit solves this bug. https://github.com/artivis/rf2o_laser_odometry/commit/ae010765627e0446930599e3376a45ecaea2b422

After building my workspace, I have nothing new, I got the same error again.

I'm using the map server to provide the map topic for both modes(indoor and outdoor) and the outdoor area with almost flat, I suspect that was the issue.

I hope anyone can help me with this issue, I got stuck by this error.

Thanks!

JGMonroy commented 4 years ago

Hi @mhaboali , rf2o is a laser-based odometry estimator. It should work independently of the scenario, either indoor or outdoor, as long as the laser is able to catch/see enough obstacles in the environment. That given said the main issue with outdoor environments comes with the laser itself, not being able to detect enough obstacles from which to discern the odometry movement.

In outdoor environments, you will need a long-range laser to ensure a solution can be found. The problem u mention is the result of the method falling to get a valid solution (odometry estimation), which in most cases is due to lack of laser data.

mhaboali commented 4 years ago

Thanks, @JGMonroy for your reply. I can see what you mean, and I think the issue is coming from the outdoor environment itself as you mentioned.

Thanks, Mohamed