locked
question about 3 DoF robot control software architecture RRS feed

  • Question

  • I am about to plan an architecture of a simple software controlling a 3 DoF serial robot arm.

    Could somebody give me some help or link about how to start planning it, what are the main building blocks or giving some personal experience about this theme. Thanks a lot!

    z

    Monday, March 31, 2008 12:42 PM

Answers

  • Three Degrees of Freedom is not a lot for an arm. Can you explain what they are? I'm assuming that there are probably two rotational joints and a gripper. A joint and a gripper are fundamentally different in their operation. However, in working with a Lynx 6 arm I treated the gripper as a rotational joint from an API point of view because it uses a servo motor and gearing so you can send it a "joint angle" the same as the other joints.

     

    There is an Articulated Arm generic contract in MSRS, but personally I think it is incomplete because it does not allow for coordinated motions.

     

    Have you looked at the Kuka educational tutorials for robotics? They might be more complex than what you want, but it might give you some ideas. Kuka have their own contracts for robotic arms, in part to solve the problem I mentioned in the previous paragraph.

     

    You will also need to develop your Forward and Inverse Kinematic solutions for the arm.

     

    Trevor

     

    Wednesday, April 2, 2008 8:06 AM

All replies

  • Three Degrees of Freedom is not a lot for an arm. Can you explain what they are? I'm assuming that there are probably two rotational joints and a gripper. A joint and a gripper are fundamentally different in their operation. However, in working with a Lynx 6 arm I treated the gripper as a rotational joint from an API point of view because it uses a servo motor and gearing so you can send it a "joint angle" the same as the other joints.

     

    There is an Articulated Arm generic contract in MSRS, but personally I think it is incomplete because it does not allow for coordinated motions.

     

    Have you looked at the Kuka educational tutorials for robotics? They might be more complex than what you want, but it might give you some ideas. Kuka have their own contracts for robotic arms, in part to solve the problem I mentioned in the previous paragraph.

     

    You will also need to develop your Forward and Inverse Kinematic solutions for the arm.

     

    Trevor

     

    Wednesday, April 2, 2008 8:06 AM
  •  

    Sorry for not to answer your post so long. Actually what I found out as a task is a project of some course in my university. Since that time I created this thread, I decided not to plan an architecture of a robot control software, but to analyse and get to know what is going on in MSRS while controlling an articulated arm (like simulationtutorial4).
    I made this model from the control point of view.

     

    http://digitus.itk.ppke.hu/~sepzo/jpg/MySimulation.JPG

     

     

    Unfortunately I could not dig below the Joint.SetAngularDriveOrientation because I cannot find the implementation of Joint class. Do somebody know about that? Is this model more or less correct? How should I extend this ?

    z01ee
    Monday, April 14, 2008 5:25 PM
  • It's not clear to me why you need to "dig below" the level of the Joint Class. You can build on this class if you want to, but generally you just create joints, like in Simulation Tutorial 4.

     

    Trevor

     

     

    Tuesday, April 15, 2008 3:18 AM
  • The reason I would like to dig below is that I would like to model the process when I set some joint angles in my robot and then it really moves there. In general my task is to model a simplified version of the control part's architecture in this specific task(simtutorial4). But I'm sure there are several other directions to dig in this model in order to make it complete. I am waiting for any suggestions or correcting about this. Thanks!

     

     

    Tuesday, April 15, 2008 11:36 AM