Qualitative analysis of task kinematics for compliant motion planning

Abstract

In compliant motmion planning, a. iiiodel of t,a.sk kin(.mat,ics facilit,ates t,he specification of robot, control strat,egies. This paper 1iresrnt.s a qualitmat,ive geoiiiet,ric reasoning approa.ch to 6he arialysis of robot, t.a.sk kiiieinatics. By clualit,ative geometric reasoning, we m m n t.he process of deriving a solut,ion based on a set, of geomet,ric rules in which t,he concept.^ are described in t,ernis of symbolic q~a l i t~a t i v e va.lues and partia.1orderitig relat,ioiis. The qualit,ative kiiieinatic analysis scheme permit,s comput.a.t~ioiiaIly efficient,, approsimate solution of kiiieiiiatic problems without using exact, iiiformat,ioii on t,he geomet,ry of meclianisnis under consideration a.nd provides a hasis for t,lie functional siiiiulat,ioii of mechanisms.

Obtaining Optimal Mobile-Robot Paths with Nonsmooth Anisotropic Cost Functions Using Qualitative-State Reasoning

“…We will only consider here landmark values that are headings, either of the vehicle or surface at a point, since we assume a terrain area in which is constant. Our landmark values will not be evenly spaced as in the qualitative path planning in configuration space of [Liu and Daneshmend 1991], but will be the meaningful headings of the domain, as this will permit fewer of them and better domain representation. So for our vehicle model, we will define the following qualitative states, where is the azimuth angle measured clockwise from uphill with defined range : S1.…”

“…( [Liu and Daneshmend, 1991] and [Holmes and Jungert, 1988] use this idea for isotropic obstacle-avoidance path planning.) Then for a particular start state, the optimal path to a goal can be found by a well-behaved optimization local to a state sequence.…”