0
$\begingroup$

Rosanswers logo

I am working on calibrating the pose of some lidar relative to base_link frame. After Calibrating x,y and yaw angle of the lidar, I noticed that there is inconsistency in the data showed by the sensor. By increasing the decay time of the laser scans and rotating the robot in place (having a good odometry) this is what I see in Rviz:

image description

I also compared what the front and back lidar see and this is what I get: image description

The robot is standing in a nearly rectangular area, what I find weird is that in one corner both laser are well aligned while on the other corner the deviation of the scans is huge.

I am wondering if this effect can be produced by a wrong calibration of x,y,yaw values or that it is the result of a tilted lasers not sharing the same scanning plane (roll and pitch are not equal to zero).


Originally posted by Mehdi. on ROS Answers with karma: 3339 on 2016-03-23

Post score: 0

$\endgroup$

1 Answer 1

0
$\begingroup$

Rosanswers logo

What do you mean with "wrong calibration of x,y,yaw values" ? You performed a 3DoF calibration, but you need a full 6DoF calibration here. What you see is, without doubts, an issue caused by having different scanning planes. Which kind of robot are you using?


Originally posted by Augusto Luis Ballardini with karma: 430 on 2016-03-24

This answer was ACCEPTED on the original site

Post score: 0


Original comments

Comment by Mehdi. on 2016-03-24:
Yes well I assumed that the lasers are mounted horizontally, but I guess some error in the pitch and roll cannot be avoided, leading to different scanning planes. The robot I work on is developed at the company where I work.

Comment by Augusto Luis Ballardini on 2016-03-26:
I had the same issues in my lab in the university ;-) You can't assume anything in this robotic-world!

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.