Gazebo | Ignition | Community
Ask Your Question

Model moves and rotates when inserted

asked 2016-01-07 07:32:09 -0500

Anand George gravatar image

Hi all

I made a differential drive robot with meshes for bot collision and visual elements. When it is inserted, the robot keep moving in some direction and rotates in its own axis. Both wheels are also rotating even if velocity is not specified. Sometimes it moves forward and backward.Here are some screenshots and the model directory of the case. Properties of each link can be found here. How these errors can be resolved?

Thanks in advance

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2016-01-07 12:05:28 -0500

chapulina gravatar image

It's kind of hard to tell without the plugins. I suggest you inspect your model to be sure all parameters are correct. For example, right clicking on the model and visualizing center of mass and inertia, I could see that you have a very heavy mass on a corner. (The visualizations available will depend on your version of Gazebo)

image description

Another way to find out why the movement is wrong is pausing the simulation then stepping through it slowly to see what's going on.

edit flag offensive delete link more


The same thing is happening even after disabling the plugin. What are those spheres and boxes indicating?

Anand George gravatar imageAnand George ( 2016-01-08 00:12:09 -0500 )edit

Center of mass of every links is located at its origin which are located at a corner of the link( ). Maybe thal causes the problem. How can I make the origin of each link to its original center of mass?

Anand George gravatar imageAnand George ( 2016-01-09 11:30:50 -0500 )edit

First, I suggest using something like MeshLab to recenter your mesh. As for the center of mass, there's a tag **pose** in **inertia** for a **link** which you can set.

Peter Mitrano gravatar imagePeter Mitrano ( 2016-01-09 13:57:55 -0500 )edit

I've set pose for inertial element of every link. Now the center of masses are at the required positions (almost). But still, the model slips(slightly). Also when some velocity is applied to the hinge joint, the model moves abruptly in the 3D space. Why this is happening ?

Anand George gravatar imageAnand George ( 2016-01-10 02:13:24 -0500 )edit

Question Tools



Asked: 2016-01-07 07:32:09 -0500

Seen: 2,760 times

Last updated: Jan 07 '16