Visual reasoning is an essential skill for many disciplines in engineering, architecture, and design. The underlying cognitive processes of visual reasoning form a basis in various problem-solving processes. We describe an intelligent tutoring system for visual reasoning that uses the missing view problem. This system, called Intelligent Visual Reasoning Tutor (IVRT), can adaptively support different learners' needs, track learners' progress, and provide active critiquing. IVRT uses a two-level reasoning architecture, combining geometric reasoning and semantic technologies, which enables the development of ITS for 3D geometry domains. We discuss IVRT's system architecture and implementation, which includes a learning contents model based on skills, lessons, and problems, aid a learner model that measures domain competence as a set of skills. Learning contents and pedagogical leaching strategy rules arc stored in standard OWL ontologies, which can be customized by the teacher.

Duplicate designs consume a large amount of enterprise resources during product development. Automatic search for similar parts is an effective solution for design reuse. Previous studies have only concerned similarity assessment based on complete 3D models, which may produce unsatisfactory result in practice. This paper proposes a novel scheme which incorporates the concept of LOD (levels of detail) into 3D part search. The scheme allows searching with different LOD variants created from the negative feature tree (NFT) of a solid model. A back-propagation artificial neural network is established to combine the D2-based similarity evaluation at each level of NFT. A human cognition model (HCM) is obtained by training the network with a set of data generated from a human experiment of similarity ranking. Search examples based on HCM show that the proposed scheme provides a practical tool for retrieval of similar part models.

We present an ontology of objects, relations among objects, and generic shape representation that supports form-function reasoning. By reasoning from the generic functions of objects to their geometric shape requirements, we deduce the generic shape representation of everyday objects. This is a complex kind of reasoning that combines diverse knowledge sources and principles. We model the results of this reasoning process as a justification graph of individual reasoning steps, which explicitly links the attributes of objects and their relations to the corresponding geometric shape elements. This object ontology uses OWL Full metamodeling techniques to achieve the necessary level of expressiveness while maintaining a generic representation. We give an example for the Table class, showing its decomposition into functions, features, and relations, and its form-function reasoning process.

A learning diagnosis system collects data from a learner's learning process, and analyzes it to build a suitable model for the learner, which can then be incorporated into an intelligent tutoring system to provide customized tutoring services. However, if the collected data reflects inconsistent learner behaviors or unpredictable learning tendencies, then the reliability of the learner model is degraded. in this paper, the outliers in the learner's data are eliminated by a k-NN method. We apply this method to an experimental data set obtained using DOLLS-HI a learner diagnosis system that uses housing interior learning contents to diagnose learning styles. The resulting diagnosis model shows improved reliability than before eliminating the outliers.

A learning diagnosis system collects data from a learner's learning process, and analyzes it to build a suitable model for the learner, which can then be incorporated into an intelligent tutoring system to provide customized tutoring services. However, if the collected data reflects inconsistent learner behaviors or unpredictable learning tendencies, then the reliability of the learner model is degraded. In this paper, the outliers in the learner's data are eliminated by a k-NN method. We apply this method to an experimental data set obtained using DOLLS-HI, a learner diagnosis system that uses housing interior learning contents to diagnose learning styles. The resulting diagnosis model shows improved reliability than before eliminating the outliers.

A pen container showing a relief on

We present an ontology of objects, functions, and generic shape representation that supports form-function reasoning. By reasoning from the mechanical and other functions of objects to their geometric shape requirements, we deduce the generic shape representation of objects, which we represent as a partial boundary representation composed of primitive geometric shape elements and their spatial and other relations. We use this ontology to model a knowledge base of everyday objects, including their generic shapes. This ontology can support applications such as product design and object recognition.

Within an intelligent tutoring system framework, the teaching strategy engine stores and executes teaching strategies. A teaching strategy is a kind of procedural knowledge, generically an if-then rule that queries the learner’s state and performs teaching actions. We develop a concrete implementation of a teaching strategy engine based on an automatic conversion from SWRL to Jess. This conversion consists of four steps: (1) SWRL rules are written using Protégé’s SWRLTab editor; (2) the SWRL rule portions of Protégé’s OWL file format are converted to SWRLRDF format via an XSLT stylesheet; (3) SweetRules converts SWRLRDF to CLIPS/Jess format; (4) syntax-based transformations are applied using Jess meta-programming to provide certain extensions to SWRL syntax. The resulting rules are then added to the Jess run-time environment. We demonstrate this system by implementing a scenario with a set of learning contents and rules, and showing the run-time interaction with a learner.

We present an ontology to represent generic teaching strategies in an intelligent tutoring system framework. The learning contents, learner's state, and system actions are modeled in just enough detail to support the definition of teaching strategies. Teaching strategies are organized into a hierarchy of teaching goals, and are represented using SWRL rules.

Next generation process planning systems should be capable of dealing with industrial demands of versatility, flexibility, and agility for product manufacturing. Development of process planning system is heavily dependent on feature recognition, but presently there is no satisfactory feature recognition system relying on a single method. In this paper, we describe a hybrid feature recognition method for machining features that combines three feature recognition technologies: graph-based, convex volume decomposition, and maximal volume decomposition. Based on an evaluation of the strengths and weaknesses of these methods, we integrate them in a sequential workflow, such that each method recognizes features according to its strengths, and successively simplifies the part model for the following methods. We identify two anomalous cases arising from the application of maximal volume decomposition, and discuss their cure by introducing limiting halfspaces. All recognized features are combined into a unified hierarchical feature representation, which captures feature interaction information, including geometry-based machining precedence relations.

IVRT is an ITS for visual reasoning, using the missing view problem. It combines an ITS framework with a solid modeling kernel that supports hintgenerating rules using geometric reasoning. We develop an ontology for IVRT’s hint generation rules, and a separate ontology for IVRT’s teaching strategy. Teaching strategy rules are stored in a custom text format, with compilation to Jess. The ability to visualize and reason about geometric aspects of 3D objects is critical for success in many disciplines in engineering and architecture. The missing view problem [1], shown in Figure 1, is typically used in visual reasoning instruction. Two consistent, principal orthographic views are given, and the learner must provide the third view corresponding to a valid 3D solid object. Top View Front View Side View Solid (pictorial view)

A mobile robot that interacts with its environment needs a machine-understandable representation of objects and their usages. We present an ontology of objects, with generic shape representations obtained through form-function reasoning. Sets of objects are associated with typical human activities, which supports context understanding. We describe an efficient ontology document storage system, which is based on stable and well-known relational databases. We first design a relational data schema appropriate for Web Ontology Language (OWL) documents, and then develop a transformation mechanism from OWL documents to the relational schema.

Visual reasoning is an essential skill for many disciplines in engineering and architecture. We describe an intelligent tutoring system for visual reasoning that uses the missing view problem, a learning contents model based on skills, lessons, and problems, and a learner model that measures domain competence as a set of skills. Learning contents and pedagogical teaching strategy are stored in ontologies, which can be customized by the teacher.

We describe a hybrid feature recognition method for machining features that integrates three distinct feature recognition methods: graph matching, cell-based maximal volume decomposition, and negative feature decomposition using convex decomposition. Each of these methods has strengths and limitations, which are evaluated separately. We integrate these methods in a sequential workflow, such that each method recognizes features according to its strengths, and successively simplifies the part model for the following methods. We identify two anomalous cases in the application of maximal volume decomposition, and their cure by introducing limiting halfspaces. Feature volumes recognized by all three methods are then combined into a unified hierarchical feature representation, which captures feature interaction information, including geometry-based machining precedence relations.

A robot that acts within an everyday environment needs a machine-understandable representation of objects and their features, shapes, and usages. We report on the development of a generic ontology of objects, and the use of this ontology to instantiate a knowledge base of everyday physical objects. Generic shape representation of objects and features is obtained through formfunction reasoning to deduce geometric shape requirements from an object’s mechanical and other functions, which supports object recognition. Associational knowledge between objects captures typical associations among groups of objects that are commonly used together, and associations between sets of objects and typical human activities, which supports context understanding.

We describe an automatic machining tool path generation method that integrates local and global tool path planning for machining features. From the solid model and the tolerance specifications of the part, we automatically recognize machining features, and obtain the geometry-based precedence relations between these features. A separate process planning module uses this information to determine the machining sequence, tool selections, and machining conditions. From the resulting process plan, we then generate machining tool paths for each set-up, combining local and global tool paths. Machining features are expanded through their fictitious faces to obtain feature free spaces. These comprise the cells of a free space decomposition, which enables the use of established robot motion planning techniques. Global tool paths between features are generated incrementally by searching the adjacency graph of feature free spaces, which represents the free space of the part at each step of the process plan. Local tool paths for each machining feature are generated by successive offsetting operations. The start and end positions for each feature's local tool paths are selected using a heuristic method to minimize the cost of each segment of the global tool path. This feature recognition method and the automatic tool path generation method are being developed as modules of a comprehensive machining process planning system.

This paper presents a method to generate machining precedence relations systematically based on the geometric information of the part. The feature recognition method using Alternating Sum of Volumes with Partitioning (ASVP) Decomposition is applied to obtain a Form Feature Decomposition (FFD) of a part model. Form features are classified into a taxonomy of atomic machining features to which machining process information has been associated. Geometry-based precedence relations between features are systematically generated using the face dependency information obtained by ASVP Decomposition and the features' associated machining process information. Multiple sets of precedence relations are generated as alternative precedence trees based on the feature types and machining process considerations. These precedence trees can be further enhanced with precedence relations from tolerance specifications and machining expertise. Machining sequence planning can be performed for each of these precedence trees while minimizing the number of tool changes. The precedence trees may then be evaluated based on machining cost and other criteria. The precedence-reasoning module is currently being implemented within a comprehensive computer-aided process planning system.

Abstract We describe an automatic machining tool path generation method that combines local tool path planning for machining features with global tool path planning. From the solid model and the tolerance specifications of the part, machining features are automatically recognized, and geometry-based precedence relations are obtained between these features. From this information, the machining sequence, tool selections, and machining conditions are determined. Machining tool paths are then generated automatically for each setup, combining local and global tool paths. Local tool paths to machine each feature are generated using successive offsetting operations. Global tool paths between features are generated incrementally by searching the adjacency graph of feature free spaces, which represents the current free space of the part. Feature free spaces are obtained by expanding the machining features through their fictitious faces. The start and end positions for the local tool paths of each feature are selected based on a heuristic method to minimize the cost of each segment of the global tool path. This automatic tool path generation method is currently being developed as part of a comprehensive machining process planning system.

This paper presents a feature-based method to obtain assembly mating relations between a set of polyhedral components. Previous research in assembly planning has assumed that these assembly mating relations, or equivalent information such as the final assembled configuration of the components, is provided as part of the problem input. This paper addresses the case where only the component geometry is provided, which can arise in product maintenance or emergency repair applications where the originally designed assembly mating relations are no longer valid, and in robotic autonomous construction applications. The basis of the method is to represent components in terms of form features using a feature recognition method based on Alternating Sum of Volumes with Partitioning (ASVP) decomposition. Feature recognition is applied to each component to obtain its Form Feature Decomposition (FFD), which is a hierarchy of positive and/or negative form features. Positive form features of each component are compared with negative form features of other components to obtain mating relations between pairs of features. Multiple feature matings between the same pair of components are merged into component matings, which comprise the assembly mating relations. A backtracking search generates all feasible assembly configurations from the component matings.