A Framework for Non-Nominal Visualization and Perceived Quality Evaluation

Perceived Quality clusters different aspects that influence the customer’s perception of non-functional quality on a product that are perceive through senses. All together those aspects and the harmony between them reflect the producer’s ability to control product parameters and thereby also mirror the functional quality of the product. High Perceived Quality cannot be added to the product at the end of the developing process. Project prerequisites, system solution, factory capability etc. are criterion to succeed. Therefore, it is important to be able to evaluate Perceived Quality early in the process when product system solutions and architecture are defined, although data maturity is low. This paper presents a comprehensive framework to manage and support evaluation of Perceived Quality aspects in a product development process. The framework is based on an industrial process in combination with recent research within the field. The framework focuses on activities that can be performed at different stages in the developing process based on maturity of the CAD or styling data. For example, if the styling data is divided into different components by split-lines it has reached higher level of maturity then styling data that not has been divided. Consequently, the choice of applied method is based on data maturity, regardless phase in the developing process. The framework contains methods based on several different simulation and analysis techniques. Design methods, Computer-Aided Tolerancing and FEA based non-rigid variation simulation are represented in the framework.

GPU-Based Haptic Simulator for Dental Bone Drilling

Dental bone drilling is an inexact and often a blind art. Dentist risks damaging the invisible tooth roots, nerves and critical dental structures like mandibular canal and maxillary sinus. This paper presents a haptics-based jawbone drilling simulator for novice surgeons. Through the real-time training of tactile sensations based on patient-specific data, improved outcomes and faster procedures can be provided. Previously developed drilling simulators usually adopt penalty-based contact force models and often consider only spherical-shaped drill bits for simplicity and computational efficiency. In contrast, our simulator is equipped with a more precise force model, adapted from the Voxmap-PointShell (VPS) method to capture the essential features of the drilling procedure. In addition, the proposed force model can accommodate various shapes of drill bits. To achieve better anatomical accuracy, our oral model has been reconstructed from Cone Beam CT, using voxel-based method. To enhance the real-time response, the parallel computing power of Graphics Processing Units is exploited through extra efforts for data structure design, algorithms parallelization, and graphic memory utilization. Preliminary results show that the developed system can produce appropriate force feedback at different tissue layers.

Supporting Co-Design of Physical and Control Architectures of Mechatronic Systems

There is a rather recent tendency to define the physical structure and the control structure of a system concurrently when designing the architecture of a product, i.e., to perform codesign. We argue that co-design can only be enabled when the mutual influence between physical system and control is made evident to the designer at an early stage. Though the idea of design integration is not new, to the best of our knowledge, there is no computer tooling that explicitly supports this activity by enabling co-design as stated before. In this paper the authors propose a method for co-design of physical and control architectures as a better approach to design mechatronic systems, allowing to exploit the synergy between software and hardware and detecting certain design problems at an early stage of design. The proposed approach is supported by a set of tools and demonstrated through an example case.

GEO + ES Hybrid Optimization Algorithm Applied to the Parametric Thermal Model Estimation of a 200N Hydrazine Thruster

In the present paper, a hybrid version of the Generalized Extremal Optimization (GEO) and Evolution Strategies (ES) algorithms [1], developed in order to conjugate the convergence properties of GEO with the self-tuning characteristics present in the ES, is applied to the estimation of the temperature distribution of the film cooling near the internal wall of a thruster. The temperature profile is determined through an inverse problem approach using the hybrid. The profile was obtained for steady-state conditions, were the external wall temperature along the thruster is considered as a known input. The Boltzmann’s equation parameters [2], which define the cooling film temperature profile, are the design variables. Results using simulated data showed that this approach was efficient in recuperating those parameters. The approach showed here can be used on the design of thrusters with lower wall temperatures, which is a desirable feature of such devices.

Towards a Framework for Holistic Ideation in Conceptual Design

The long term aim of this research is to develop a framework for holistic ideation. Towards that goal, we are investigating the components of ideation (both logical and intuitive), characterization of ideation states, ideation blocks and strategies used to overcome such blocks. One of the major contributions of this research is the method by which the experiential and the intuitive ideation methods are interlaced together, to support both creativity and functional quality. Another important part of the framework is the identification of creativity blocks and investigation of matching ideation strategies most likely to facilitate progress. This framework will also be used as a research tool to collect large amount of data from designers about their choice of ideation strategies used, and their effectiveness.

Evaluation of Deviation Zone Based on Maximum Conformance to Tolerances

The accurate estimation of the geometric deviations is not possible only by manipulating the Euclidian distances of the discrete measured points from substitute geometry. The real geometric deviations of a measured surface need to be calculated based on the desired tolerance zone of the surface. This fact is usually neglected in common practices in the coordinate metrology of surfaces. The importance of considering the desired tolerance zone in estimation of the optimum deviation zone is demonstrated in this paper. Then a best fit method is presented which complies with the tolerance requirements of the designed surface. The developed fitting methodology constructs a substitute geometry to minimizes the residual deviations corresponding to the given tolerance zone and the needs of down-stream operations that use the results of the inspection process. It is shown how the developed objective function can be adopted for a case of closed-loop manufacturing process, when the under-cut residual deviations of the manufactured part can be corrected by a down-stream operation. In order to validate the proposed methodology, experiments are conducted. The results show a significant reduction of uncertainties in coordinate metrology of geometric surfaces. Implementation of this method directly results in increasing the accuracy of the entire tolerance evaluation process, and less uncertainty in quality control of the manufactured parts.

Boundary Element Parallel Computation for 3D Elastostatics Using CUDA

Finite Element Method (FEM) is pervasively used in most of 3D elastostatic numerical simulations, in which Computer Aided Design (CAD) models need to be converted into mesh models first and then enriched with semantic data (e.g. material parameters, boundary conditions). The interaction between CAD models and FEM models stated above is very intensive. Boundary Element Method (BEM) has been used gradually instead of FEM in recent years because of its advantage in meshing. BEM can reduce the dimensionality of the problem by one so that the complexity in mesh generation can be decreased greatly. In this paper, we present a Boundary Element parallel computation method for 3D elastostatics. The parallel computation runs on Graphics Processing Unit (GPU) using Computing Unified Device Architecture (CUDA). Three major components are included in such method: (1) BEM theory in 3D elastostatics and the boundary element coefficient integral methods, (2) the parallel BEM algorithm using CUDA, and (3) comparison the parallel BEM using CUDA with conventional BEM and FEM respectively by examples. The dimension reduction characteristics of BEM can dispose the 3D elastostatic problem by 2D meshes, therefore we develop a new faceting function to make the ACIS facet meshes suitable for Boundary Element Analysis (BEA). The examples show that the GPU parallel algorithm in this paper can accelerate BEM computation about 40 times.

VR-SMART: A Virtual Reality System for Smart Home Environments

Virtual reality (VR) technologies and systems have the potential to play a key role in assisting disabled inhabitants of smart home environments with instrumental activities of daily living (IADLs). While immersive environments have useful applications in the fields of gaming, simulation, and manufacturing, their capabilities have been largely untapped in smart home environments. We have developed an integrated CAD and virtual reality system which assists a smart home resident in locating and navigating to objects in the home. Using the methods presented in this paper, a room modeled in a CAD system is imported into a virtual environment, which is linked to an audio query-response interface. The user’s head and room objects are fitted with the sensors which are part of a six DOF motion tracking system. Methods have been created to allow the inhabitant to move objects around in the room and then later issue an audio query for the location of the object. The system generates an audio response with the object’s position relative to the person’s current position and orientation. As he approaches the object, information is derived from the virtual models of both the room and the objects within the room to provide better guidance. The ability of the VR-SMART system to guide a resident to an object was tested by mounting a head mounted display (HMD) on a user located in a room. This allowed the user to navigate through the virtual world that simulated the room he occupied, thereby providing a way to test the positional accuracy of the virtual system. Results of the testing in the immersive environment showed that although the overall system shows promise at a 30% success rate, the success of the system depends on the accuracy and calibration of the tracking system. In order to improve the success of the system, we explored the precision of a second motion capture system, with more accurate results. Results confirmed that the VR-SMART system could significantly improve the assistance of disabled people in finding objects easily in the room when implemented only as an assistive system without the head-mounted display.

A Framework for Self-Realizing Process Models for Additive Manufacturing

This research discusses a framework for automating process model realization for additive manufacturing. The models map relationships from design requirements to process variables and can be utilized for future process planning. A repository is employed to collect data and contains previous process plans and corresponding design requirements. The framework organizes data through a statistical clustering method and builds regression models using a multi-layer neural network. Hierarchical and k-means clustering methods are employed in series to manage the data. A two layer neural network and augmented training algorithm are employed to build process models. The framework has been tested with Stereolithography and Selective Laser Sintering process planning problems to demonstrate its usefulness.

An Ontology-Based Online Community to Maintain Engineering Knowledge in a Training Domain

The research presented in this paper seeks to develop an ontology-based online community for knowledge exchange between expert engineers and new engineers. We call this community CREEK (Community for Retention of Engineering dEsign Knowledge). This paper seeks to develop methods and tools related to knowledge acquisition, knowledge modeling, knowledge management, and knowledge presentation that can support activities in this community in engineering design and training domains. An important consideration is to design and deploy the online community and the underlying ontology model such that they will not exist in isolation but will be connected to a product data ontology and a training ontology. In our previous work engineering knowledge related to product data in engineering design and assembly has been modeled. In this work procedural knowledge and knowledge in the training domains related to these procedures are also modeled and populated using ontologies. We have designed an architecture that will allow the ontology of the on-line community to tap into the engineering knowledge from these two supporting domains. In addition, there is a connection the other way too. The online community also allows new knowledge to be captured from experts and be (semi-)automatically transferred to product design and training domains.