ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Robust robot docking with lidar only

asked 2022-12-30 03:01:43 -0500

rodrigo55 gravatar image

Hello, I'm looking on advice on how I can improve my robot docking algorithm.

I only have a lidar to work with, and I use a kobuki base.

Currently, the robot is able to detect a triangle-shaped dock and move towards the middle of it using a PD controller.

However, this approach does not consider how centered it is with respect to the dock. What I mean is that I have no control to check whether the robot is approaching the dock perpendicularly, which is needed to make contact with the charger.

Any ideas on how to add this into my code?

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2022-12-30 10:06:48 -0500

ljaniec gravatar image

Generally, the solutions I saw in the wild on the Internet used additional infrared sensors coupled to a USB camera, with possible QR/AR tags on the charging station.

If additional sensors aren't possible, perhaps you could try adding an easily identifiable object next to the dock so that you could control the angle as you approach the station? Perhaps this is not even needed if you can calculate the difference between the robot's current position and the position needed at the end of the path - I can see use of Rotation Shim Controller here (if you use Nav2, of course), combined with special behavior when the robot is close to the target, e.g., the robot approaches the station when it is within 1 m of the station, plans a midpoint just before the station and an end point with a tight angle error + a possible angle correction with the other waypoint.

How do you detect this triangle-shaped dock now? Did you transform the LiDAR scan to point clouds and use some known packages to detect objects? You can add a description of your setup as well :)

edit flag offensive delete link more

Question Tools



Asked: 2022-12-30 03:01:43 -0500

Seen: 306 times

Last updated: Dec 30 '22