Home /Research /Robot Manipulator Control and Computational Cost
MANIPULATION

Robot Manipulator Control and Computational Cost

Yehong Zhang, Richard P. Paul

Year
1988
Citations
6
Access
Open access

Abstract

In this paper robot control algorithms are considered from the point of view of computational complexity. All the algorithms provide for hybrid position and force control which is appropriate to unstructured environments. Both joint coordinate and generalized coordinate motion schemes are considered. The findings are interesting in that the peripheral processing involved in all methods, sometimes, swamps the orders of magnitude differences in computation involved in the inner loops. It is also interesting to note the similarity in complexity between direct methods and differential. Between joint coordinate methods and generalized coordinate methods there is a general two to one ratio in complexity. In general a 0.75 M Flop, 1.25 Mip machine is required to provide for a control rate of 250 hz.

Keywords

GRASPComputational complexity theoryComputer scienceComputationPosition (finance)Coordinate systemRobotControl (management)Artificial intelligencePoint (geometry)

Related papers

Browse all MANIPULATION papers