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

Revision history [back]

click to hide/show revision 1
initial version

I’m going to answer my own question here.

I got a really good answer from Martin Cote at Clearpath Robotics (where I purchased my Turtlebot):

Greetings James,

Thank you for contacting Clearpath Robotics support! This is not a dumb question at all, and actually something that has been brought up by a few other customers recently! Sadly this is indeed how all Turtlebots are. The Turtlebots weren't always like this, but right around the time you probably purchased your Turtlebot, this was changed to it's current state. As you may be aware, we're simply a North American distributor for the Turtlebot, and don't have too much design input.

The Kobuki base is made by Yujin Robot, and I reached out to them recently about this issue, so sadly I'm not able to help to much to resolve this, but here is the representatives response for your own insight.

If I remember correctly, the only thing that changed is the dimensions and properties of the passive wheels. This does cause different rocking properties. I do not recall that we did anything to the diff drive mechanism for this particular problem - that is unlikely as it would have been a significant change for moulding/factory.

Getting the wheels right is a pareto-optimal game for a base with four contact points and an unknown centre of mass (knowing the centre of mass, you can get away with three wheels and avoid many problems like cleaning robots do).

Previously with all four contact points close to each other there was little rocking, and this works fine on a perfectly smooth and slippery floor. Put it on carpet though and the passive wheels will create alot of drag. Lay down some small obstacles like cables, and the base won't get over them. Since this made it effectively useless for a large number of users, we had to find that pareto-optimal compromise between stability (minimal rocking) and being functional on many environments. There's not much worse than having to make decisions where you are forced to be average rather than great.

I hope that at least answers you questions, let me know if there's anything else I can assist with!

Best regards, Martin

So, from Googling around it looks like the reason for the change is probably discussed here:

https://github.com/yujinrobot/kobuki/issues/202

Apparently the Turtlebot would get easily stuck without the extra clearance.

I understand the reasoning, but I wish there were a better solution.