Feature modelling is a promising approach for object modelling since it provides a systematic representation reflecting the functionality of the object. Traditional feature technology is largely confined to modelling regular shaped objects due to the limited geometric nature of feature representation. The configuration part of the semantics of a feature in a regular shaped object does not play a strong role in the representation. However, the same approach is not applicable to the sculptured object class using free-form surfaces. The configuration of a sculptured object is unique in a particular application domain and the feature model needs to emphasize the semantics of the model and its constituent features....[ Read more ]

Feature modelling is a promising approach for object modelling since it provides a systematic representation reflecting the functionality of the object. Traditional feature technology is largely confined to modelling regular shaped objects due to the limited geometric nature of feature representation. The configuration part of the semantics of a feature in a regular shaped object does not play a strong role in the representation. However, the same approach is not applicable to the sculptured object class using free-form surfaces. The configuration of a sculptured object is unique in a particular application domain and the feature model needs to emphasize the semantics of the model and its constituent features.

This thesis presents a framework of semantic feature language, which consists of two components: the language grammar and vocabularies. The grammar of the language is a set of production rules to control the syntactic information of the language. The vocabularies account for the semantic meaning. The semantics of the object are an aggregation of the semantics of these individual components. The semantic feature language reflects the application since the semantic meanings of the vocabularies vary with the application. The semantic feature model is represented by a three-layer structure semantic network. The first layer gives the abstract description of the object, the semantics of various parts of the object are obtained from the second layer. These two layers provide the configuration of the object and the detail semantics relating to its application. The geometry of the object is defined in the third layer as auxiliary information for the application specified by the semantics. The semantic and geometric constraints imposed by the semantic feature language are presented. The validity, sufficiency and adequacy of the semantic feature language are also discussed.

The smoothness of the surface geometry of a sculptured object is a key factor in its definition. The order one continuity requirement covering the geometric constraint for Non-uniform rational B-splines surface in a sculptured object model is presented.

The semantic feature language is shown to provide a unified approach for the modelling of regular shaped objects and sculptured objects. Two case studies of a mannequin and a bolt are used to illustrate the usage of the semantic feature language.