The field of solid modeling has developed a variety of techniques for unambiguous representations of three-dimensional objects. Feature recognition is a sub-discipline of solid modeling that focuses on the design and implementation of algorithms for detecting manufacturing information from solid models produced by computer-aided design (CAD) systems. Examples of this manufacturing information include features such as holes, slots, pockets and other shapes that can be created on modern computer numerically controlled machining systems. Automated feature recognition has been an active research area in solid modeling for many years and is considered to be a critical component for integration of CAD and computer-aided manufacturing. The paper gives an overview of the state-of-the-art in feature recognition research. Rather than giving an exhaustive survey, we focus on the three of the major algorithmic approaches for feature recognition: graph-based algorithms, volumetric decomposition techniques, and hint-based geometric reasoning. For each approach, we present a detailed description of the algorithms being employed along with some assessments of the technology. We conclude by outlining important open research and development issues.

Manufacturing feature recognition from solid models: a status report

A method has been developed that decomposes a polyhedron into maximal cells by intersecting it with half spaces of its faces having concave edges. Every set of such half spaces that would result in maximal convex cells is found very efficiently without actual intersection operations by examining the relationships among the half spaces and the neighbourhoods of concave edges. One application of this decomposition method is recognition of intersecting features for process planning. Recognition of intersecting features has been a major difficulty in automating process planning. With this decomposition method, a delta volume is decomposed into maximal convex cells. By subtracting maximal convex cells from each other in different orders, multiple interpretations of features are generated. Since the number of interpretations, which may reach N! for N maximal convex cells, can be very large and thus selecting one of them for machining may be difficult, we tried to generate a machining sequence directly from maximal convex cells of a delta volume using a small number of heuristics on machining. The result was the same machining sequence as the one suggested by a machinist.

Volume decomposition and feature recognition: part 1—polyhedral objects

A method has been developed that decomposes a curved object into volumes, called maximal volumes, with the half-spaces of the object. A maximal volume has as few concave edges as possible without introducing additional halfspaces. The object is first decomposed into minimal cells by extending the surfaces of the object. Combinations of these minimal cells are then composed to form maximal volumes. The combinations of such minimal cells that result in maximal volumes are found by traversing a search tree which is pruned by examining the relationships among these minimal cells. One application of this decomposition method is recognition of machining features. With this decomposition method, a delta volume is decomposed into maximal volumes. Many maximal volumes are readily recognizable as features with graph matching. Compound features can be recognized by combining maximal volumes. By subtracting maximal volumes from each other in different orders and applying graph matching to the resulting volumes, multiple interpretations of features can be generated.

Volume decomposition and feature recognition, part II: curved objects

Feature-based modelling is the emerging technology to integrate design and engineering activities. It enriches product data representation with semantic information, that allows more efficient and direct communication between engineering processes. Nevertheless, major industrial problems arise from the fact that the communication between these processes is still not based on sharing the same information within a computer-internal product model. The lack of industrial product data exchange between various CAD/CAE applications is owing to the fact that there are diverse data formats and standards, giving rise to conversion problems. The ISO standard STEP provides a unique, comprehensive technology to solve most of the industrial problems in data exchange. However, it still does not provide all the resources needed to achieve a high-level, semantic data communication between engineering processes. In this article, we propose a solution to link the design process (CAD) with downstream engineering processes (CAE) based on an integrated feature-based modelling approach that supports both design-by-features and feature recognition. Integration is thereby achieved through a homogeneous, multiple view feature-based representation of the part model, called intermediate model, which is shared among the various applications. A distributed, object-oriented feature-based system architecture is described. Herein, the intermediate modeller server acts as a service provider to network-based distributed applications. These share the intermediate model and are able to extract their specific view from the semantic data of the product. A prototype implementation of the integrated feature-based modelling kernel and its integration into the distributed environment is also outlined.

Design and engineering process integration through a multiple view intermediate modeller in a distributed object-oriented system environment

State of The Art-Intense Review on Artificial Intelligence Systems Application in Process Planning and Manufacturing

In a design by feature system, the database may contain features that are not known to a process planning system. Also, additional features arise from the shape and size of the workpiece from which the part is to be machined. Production rules are inadequate for dealing with these undocumented features. In order to get machining features, one must enhance and transform the design database. A volume decomposition method called minimum convex decomposition by halfspace partitioning has been developed to recognize machining features. First, the total volume to be removed by machining is obtained by subtracting the part from the stock. This volume is decomposed into minimum convex cells by halfspace partitioning at every concave edge. A method called maximum convex cell composition is developed to generate all alternative volume decompositions. The composing subvolumes are classified based on degree of freedom analysis. Manufacturing knowledge is characterized in fundamental terms, based on tool and workpiece motion (deep knowledge), by means of algebraic expressions. Inverse mapping of decomposed volumes into these expressions determines feasibility of various machining operations. The method deals effectively with feature interactions and results in alternative machining sequences.

Arvind Shirur

Papers

Determination Of Machining Volumes From Extensible Sets Of Design Features

Machining algebra for mapping volumes to machining operations for developing extensible generative CAPP

Machining Algebra for Mapping Volumes to Machining Operations

User Interface Design for Feature Modelers and Feature Based Applications