Understanding and applying a robot ping-pong player's expert controller

Abstract

Our robot ping-pong system attains good performance in a complex non-linear environment subject t o tight temporal constraints. We will show how the expert controller combines approximate estimates and feedback t o arrive a t a suitable task plan, emphasizing feasibility rather than strict optimality. We will discuss the characteristics of the expert controller t h a t cause it t o work, and their relationship t o other tasks.

High-speed Visual Servoing of PKMs

“…This application considers an eye-in-hand configuration within dynamic look-and-move position-based scheme [6]. The task is defined as a 3D visual tracking task, keeping a constant relationship between the camera and the moving target (ball).…”

“…A number of visual servo systems using model based tracking to estimate the pose of the object have been reported. Andersson presents one particular application: a ping-pong playing robot [5] [6]. The system uses a Puma robot and four video cameras.…”

Robot manipulators

“…Early examples of dynamic robot control include the ping-pong playing robot at Bell Labs [3], and the juggling robot developed by Koditschek et al [4], [5]. Another example of dynamic systems are walking robots [6].…”

Balancing of a planar bouncing object

“…This system was further analyzed by [9]. [1] designed a pingpong playing robot which used a simple model of point-mass collision to predict the motion of the ball after striking. [18] attempts to characterize the qualitative behavior change in the motion of objects upon collision.…”

“…Define the constraint surface to be the set of configurations for which the object touches the table without penetration for a given table orientation (see [11] for a discussion of configuration space). Then according to equation (1), the object will reflect about the normal to the constraint surface. This is an intuitive extension of the usual example of a perfectly elastic, frictionless point mass impact against a flat barrier.…”