Abstract
A configuration-space-based approach for analyzing the interactions and mobility of objects in quasi-static contact is described. This analysis is motivated by a class of articulated robot motion-planning problems that are not handled by current planning systems. Examples are a snake-like robot that crawls inside a tunnel by bracing against the tunnel walls, a limbed robot (analogous to a monkey) that climbs a trussed structure by pushing and pulling, and a dextrous robotic hand that moves its fingers along an object while holding it stationary. In these examples, one must plan the robot motion to satisfy high-level goals while maintaining quasistatic stability. Planning the hand-hold states (analogous to the hand-holds used by rock climbers between dynamically moving states) where the robot mechanism is at a static equilibrium is considered primarily. The results obtained are some of the first steps necessary to develop planning paradigms for this class of problems. Although the authors have the general class of quasistatic planning problems in mind, the authors use the language of grasping for discussion.

This publication has 7 references indexed in Scilit: