Time optimal control of SCARA robots

Abstract

Previous research by the authors has aimed at understanding and characterizing time optimal trajectories of two large classes of robots, polar coordinate robots and elbow robots. This paper is addressed to a third large class of robots, the SCARA robots. The nature of the time optimal control from point to point in the work space is studied by decomposing maneuvers into categories of maneuvers that aim at changing one spatial position variable, or one robot joint variable only. From this, a heuristic understanding of maneuvers involving changes in more variables is obtained. A preliminary comparison of the time optimal performance of the SCARA robot and the elbow robot is given, and the elbow robot is seen to often have speed advantages. The decoupling of the vertical and horizontal motions of the SCARA, which has advantages in simplicity of dynamics and control, is seen to prevent motors from helping one another to speed up maneuvers.