# Denavit-Hartenberg convention

There are two different conventions that can determine DH parameters. What is the difference between Craig's [1, Sec 3.4] convention and the Spong [2, Sec. 3.2] convention? I know that both methods must have the same response.

[1]: Craig, John J. Introduction to robotics: mechanics and control. Addison-Wesley, 1989.

[2]: Spong, Mark W., Seth Hutchinson, and Mathukumalli Vidyasagar. Robot modeling and control. Wiley, 2006.

Craig uses the modified DH parameters, while Spong uses the classic DH parameters.

The difference between them are the locations of the coordinates system attachment to the links: in the modified DH, the coordinates of frame $O_{i}$ is put on axis $i$, while in the classic DH convention the coordinates of frame $O_{i}$ is put on axis $i+1$.

Update (2014.10.26)

Reading Craig's book again, I noted that the modified DH matrix is also different because of the order of the applied transformations. I edited the Wikipedia link with the different DH matrix.

You should read this paper: "Lipkin 2005: A Note on Denavit-Hartenberg Notation in Robotics". It explains the 3 main DH parameter conventions and how they differ.

• That link is down. Any chance you have a nother link for the paper? Nov 5 '15 at 13:54
• I fixed the link to a more official source, although you will have to pay for the download.
– Ben
Sep 2 '16 at 16:01

This is Best Youtube video which gives thorough explanation of Denavit Hartenberg Algorithm and Parameters

Forward Kinematics: Denavit-Hartenberg Convention

Following is youtube channel by Prof. Yang Cao from University of British Columbia which covers all topics in Advanced Robotics. This is best place for learning robotics concepts for Robotics Engineering.

Robot Modeling and Control

Check it out.

• I would humbly also suggest the short video lessons at robotacademy.net.au/?s=kinematics, nearly a million downloads -- Peter Corke, creator of Robot Academy Oct 19 '19 at 12:29