Next: Sensing for Intelligent Up: Methodology Previous: Methodology

Intelligent Control

Discrete event dynamic systems (DEDS) are aptly suited for modeling robotic observers as they provide a means for tracking the continuous, discrete and symbolic aspects of the scene under consideration [9][8][5]. In our system, the DEDS controller is able to model and report the state of the inspection process.

In inspection, the DEDS guides the sensors to the parts of the objects where discrepancies occur between the reference model and the recovered structure. The DEDS formulation can also compensate for noise in sensor readings [14]. Recovered data from the sensing module is then used to drive the CAD module. The DEDS sensing agent is thus used to collect data of a passive element for designing structures.

A dynamic-recursive implementation of a DEDS algorithm is used to facilitate the state recovery of the inspection process. The dynamic recursive context for finite state machines (DRFSM) is a particularly apt choice for the exploration of a set of machined parts which exhibit a recursive nature. A feature which is contained within another feature will generally be of a smaller scale than its parent, even though it might be of a similar type and require the same sensing technique. With the DRFSM implementation, the smaller scale feature can be explored using the same technique, but with more appropriate parameters [16][17][13][18][15].

The mechanical parts we are sensing have no interaction between the boundaries of their machined features and thus may be considered to be inside or outside one another. Although by relying on this assumption we are placing a significant constraint on the class of parts which we can inspect and reverse engineer, this class still represents a large number of real-world machined parts. Given that the focus of our project is on the system's control environment, we feel that this is acceptable.

Within our system, the recursive nature of these parts is used to keep track of features and their relationships to each other. Each feature is assigned a symbol. If a feature contains another feature , we store that relationship in a ``part string'' as .



Next: Sensing for Intelligent Up: Methodology Previous: Methodology


sobh@bridgeport.edu
Mon Sep 19 19:38:55 MDT 1994