Yong Se Kim

Recognition of form features using convex decomposition

Abstract A convex decomposition method, called the alternating sum of volumes (ASV) method, uses convex hulls and set-difference operations. ASV decomposition may not converge, which severely limits the domain of geometric objects that can be handled. Via the combination of ASV decomposition and remedial partitioning for the non-convergence, a convergent convex decomposition is proposed that is called the alternating sum of volumes with partitioning (ASVP) decomposition. The paper describes how ASVP decomposition is used for the recognition of form features. In this approach to form-feature recognition, volumetric form features which are intrinsic to the shape of a given object are recognized from the boundary information through ASVP decomposition. Moreover, hierarchical relationships between the recognized form features are obtained in addition to global spatial information.

Geometric reasoning for machining features using convex decomposition

Abstract A convex decomposition called the alternating sum of volumes with partitioning (ASVP) is a hierarchical volumetric representation of solid objects obtained from the boundary information. Form features intrinsic to the shape of solid objects can be recognized using ASVP decomposition. As the form-feature decomposition is compact and faithful to the shape of the object, it includes both positive and negative form-feature components. In the paper, conversion procedures from the form-feature decomposition into features useful for machining applications are described. The decomposition is accomplished by taking the complement of the halfspace abstraction of the positive components using the original faces, and combining this with the parent negative components. The resulting negative-feature decomposition represents the removal volume. Alternative decompositions of the removal volume are also obtained by exploiting the geometric information maintained in the decomposition.

Geometric reasoning for mill-turn machining process planning

Abstract An integrated system to support both product and manufacturing process design should be such that (1) the part design can be evaluated and redesigned based on manufacturability analysis and (2) the manufacturing processes can be selected efficiently and flexibly exploiting product information provided by part design representations. In this paper, we describe the feature-based geometric reasoning system for part modeling and process planning as applied to mill-turn machined parts. The feature recognition system based on convex decomposition and the mapping method to relate the negative feature volumes to machining process classes are applied to mill-turn parts. Also, the geometry-based machining precedence relations have been generated for various alternative machining feature decompositions. The above geometric information is input to mill-turn machining process planning to determine machining process sequences and assignment to multiple spindles and turrets.

Incremental and localized update of convex decomposition used for form feature recognition

Abstract Alternating sum of volumes with partitioning (ASVP) decomposition is a volumetric representation of a part obtained from its boundary representation, organizing the faces of the part in an outside-in hierarchy. A form feature decomposition (FFD) which can serve as a central feature representation for various applications is obtained from ASVP decomposition. In a typical design procedure, part designs are analysed for a given application context and redesign suggestions are made; design changes are then propagated from one context to another through the FFD. Thus, incremental update capability is crucial to support design refinement. FFD incremental update can be achieved by incrementally updating the corresponding ASVP decomposition. This paper describes how ASVP incremental update is achieved by exploiting the hierarchical structure of the decomposition. ASVP incremental update seeks to determine the new decomposition tree of a part, after its boundary representation is slightly changed, without computing it from scratch. Based on local face alteration and extremality considerations, we look for active components that only need to be updated. For a connected delta volume, defined as the volumetric difference between the old and new parts, active components are localized in a subtree of the decomposition tree, called active subtree. Then, the new decomposition is obtained by updating only the active subtree in the old decomposition using localized ASVP decomposition operations. The algorithms described in this paper have been implemented and integrated within the form feature recognition system based on ASVP decomposition.

Handling interacting positive components in machining feature reasoning using convex decomposition

Abstract Form features to the product shape can be recognized using a convex decomposition called alternating sum of volumes with partitioning (ASVP). Since the form feature decomposition is compact and faithful to the product shape, it includes both positive and negative components. For machining applications, the positive components are converted into corresponding negative components to represent the removal volume. The positive to negative conversion is done in a top-down manner by abstracting the positive components using halfspaces determined by the original faces and combining with the parent negative component. In this paper, the handling of interacting sibling positive components which have a common parent component in the positive to negative conversion is described. Interacting sibling positive components are sequentially ordered based on face dependency and extremality. Coplanarity of some faces is exploited to obtain a concise negative feature decomposition.

Geometric operations for visual reasoning of a solid from orthographic projections

Abstract The ability of an engineer to visualize and reason about geometric aspects of physical objects and processes is crucial to the success of engineering activities. The article describes the development of geometric operations for a visual reasoning instructional software system using so-called missing view problem. The missing view problem requires construction of a valid solid given two orthographic projections. Sweeping operations, which are the reverse procedure of orthographic projection, are used to solve missing view problems by forming the boundary faces of a solid from the loops of orthographic views. Specifically, this geometric operation which forms a primary user interface of the system is described in detail in the article.