Robotics

[last updated: 2021-04-05]
go to: ...
-----

Random first steps in a huge field..,.

• Inverse Kinematics:
  • This allows you to specify the desired end position of your robot joint of interest, in x, y, z coordinates,
    and calculate the movements and rotations of the other joints in your robot in order to achieve it.

  • Given a robot manipulator that has some number of connected "joints" or nodes
    and each node has a position in 3d space
    and each node (except the end effector) will have some actuator, perhaps extension, perhaps rotation...
  • and the "last" node is the one of interest: the "end effector."
  • It is desired to derive a set of equations that take as input the final/desired/target position of the end effector
    and output the extensions and rotations of each of the node actuators
    that are necessary to produce the desired target position of the end effector.
  • Derive the equations:
    • This first method uses trigonometry:
    • Start by drawing/visualizing the assemblage of nodes.
    • Specify nomenclature for axes and positions of each node:
      • So for example, suppose you have a manipulator that contains 3 nodes:
        The first/base/ground node will have position X₀, Y₀, Z₀
        The second node will have position X₁, Y₁, Z₁
        The third node will have position X₂, Y₂, Z₂
        The end-effector will have position X₃, Y₃, Z₃
      • You must further specify the direction in 3d space of each axis on each node.
        [Note: online tutorials for this procedure usually specify node axis directions independently of other nodes. Not sure why...]
    • Specify the length of the connector between the nodes, which may be fixed or not.

.

.

.

eof