Chain reaction contraptions, commonly referred to as Rube Goldberg machines, achieve simple tasks in an intentionally complex fashion via a cascading sequence of events. They are fun, engaging and satisfying to watch. Physically realizing them, however, involves hours or even days of manual trial-and-error effort. The main difficulties lie in predicting failure factors over long chains of events and robustly enforcing an expected causality between parallel chains, especially under perturbations of the layout. We present a computational framework to help design the layout of such contraptions by optimizing their robustness to possible assembly errors. Inspired by the active learning paradigm in machine learning, we propose a generic sampling-based method to progressively approximate the success probability distribution of a given scenario over the design space of possible scene layouts. The success or failure of any given simulation is determined from a user-specified causal graph enforcing a time ordering between expected events. Our method scales to complex causal graphs and high dimensional design spaces by dividing the graph and scene into simpler sub-scenarios. The aggregated success probability distribution is subsequently used to optimize the entire layout. We demonstrate the use of our framework through a range of real world examples of increasing complexity, and report significant improvements over alternative approaches. Code and fabrication diagrams are available on the project page.

Computing multiple contact points between geometric models evolved in a virtual 3D environment is central to many robotic simulation applications. While this task can be performed efficiently and robustly between complex polyhedra, using the exact analytic geometric models issued by CAD modelers still suffers from efficiency limitations. Yet models composed of smooth surfaces are required to ensure smooth contact constraints, thus avoiding possible numerical artifacts which may dramatically affect the behavior of the system in the case of functional contacts. This paper builds on the observation that industrial CAD models are mostly composed of simple surfaces to perform an off-line identification of similar features and build a bounding volume hierarchy in order to locate potential contacts. Those are then computed by dedicated analytic methods, or an iterative root-finder, depending on the actual geometric representations of the features. In the context of dynamic simulation of robotic tasks, our method exhibits interactive computation times while naturally providing better result accuracy than existing polyhedron-specific algorithms.

This paper presents a prototype of a hands-on immersive peripheral device for controlling a virtual hand with high dexterity. Based on the results of users' tests on previous versions of our device and on the analysis of a manipulation task, this prototype is as easy as a mouse to use and allows the control of a high number of degrees of freedom (dofs) with tactile feedback. Design issues, physical phenomena and physiological behaviors are tightly linked and highly influence interaction. The goals corresponding to these issues include the choice of sensors' technology and their position on the device, low efforts exerted while using the device, relevant multi-sensorial feedback, performance of achieved tasks. An example of a grasping task illustrates the effectiveness of our device to achieve intuitive and efficient interactions, bringing new insights for collaborative interaction.

The concept of product view and the corresponding required models is a basic constituent of a product development process where shape transformations are repeatedly encountered. Here, topological and geometric issues are identified and major shape modeling principles are studied. As a result, some core concepts of a framework for product view integration are proposed and justified. Notions of mixed shape representation and layered topological representation are briefly outlined as part of the proposed framework. Categories of operators enabling the required shape transformations are also shortly introduced as element of the proposed framework.

Non-manifold models are frequently encountered in engineering simulations and design as well as in computer graphics. However, these models lack shape characterization for modelling and searching purposes. Topological properties act as a kernel for deriving key features of objects. Here we propose a classification for the non-manifold singularities of non-manifold objects through continuous shape transformations of 2-manifolds without boundary up to the creation of non-manifold singularities. As a result, the non-manifold objects thus created can be categorized and contribute to the definition of a general purpose taxonomy for non-manifold shapes.

Even if today's CAD systems can easily represent free-form shapes by means of NURBS surfaces, their definition and modification still require a deep knowledge and a great skill in the manipulation of the underlying mathematical models. This paper presents an attempt to bring the feature concepts, well-known in the classical mechanical domain, to the free-form domain. The paper extends our previous work on fully free-form features to include parameterised feature templates. The free-form shapes are obtained by deformation according to specific constraint lines taking part to the feature templates definition. The feature template is adapted to the user-specified parameter values by our deformation engine, which can applied either to surfaces and curves. The method is illustrated with examples obtained with our prototype software.

In this paper we present a GPU-based method for removing shape details of 3D models. 3D models used in finite element analysis (FEA) are often either constructed for the purpose of manufacturing, or a result of 3D scanning. The models therefore contain shape details that are neither important for FEA nor compatible with the mechanical hypotheses. Vertex removal is a popular method for removing geometrical details where vertices are removed one by one, provided certain constraints are satisfied. The constraints can either be based purely on geometrical properties, or also on mechanical ones. The computations required in this process can be time consuming, especially if mechanical constraints are involved. The main idea behind our method is to perform the computations for all the vertices in parallel using graphics hardware, and then use the CPU to maintain the data structure representing the triangulation. As a result, simplification functions can stay interactive while incorporating complementary mechanically-based criteria in addition to the geometric ones involved in shape transformation.

In order to better integrate downstream applications - like simulation or virtual reality based applications - within design cycles, geometric data extracted from the digital mock-up need to be prepared and adapted to the needs and the constraints of these applications. In this paper, we propose to formalize the preparation process from the digital mock-up to an adapted database (AD). These processes allow not only transforming geometric data, but also creating a link between the initial data and the AD. The current computer implementation shows how this processes can be described and saved, in order to automate and capitalize processes, reduce preparation time, and finally enhance downstream applications integration within design.

Fully free form deformation features (/spl delta/-F/sup 4/) have been proposed to overcome the limits of low-level manipulations of free form surfaces. They correspond to shapes obtained by deformation of a surface part according to geometric constraints. In our approach, a /spl delta/-F/sup 4/ is a result of the indirect manipulation of external forces applied to the nodes of a bar network coupled to the control polyhedron of a B-spline surface. The solution of the equation system corresponding to the constraint specifications, often under-constrained, requires the definition of an optimisation problem where an additional objective function has to be minimised. In this paper, we propose a new formulation of this optimisation problem where the proposed objective functions can be defined as a multiple combination of various local quantities. They can be related either to the geometry of the bar network (e.g. the length of a bar or the displacement of a node), or to its mechanical magnitudes (e.g. the external force applied at a node or a bar deformation energy). Different types of combinations are also proposed and classified according to the induced level of multi-minimisations. In this way, the shape of a /spl delta/-F/sup 4/ can be controlled globally, with a unique minimisation, or locally with different minimisations applied to sub-domains of the surface.

Collaborative design is a major industrial issue, where people from different expertise fields strongly need to cooperate. New Internet Collaboration Tools (ICT) possibilities now allow design processes to be hold over distributed teams. Our goal is to define a collaborative environment based on Internet technologies: Techniques, methods, and tools for CSCW in design. From the design experiment of an electromechanical plunger hold over Internet and involving mechanical and electrical engineers, this paper looks at their collaborative design activity. This leads us to present now communication tools, dedicated to specific tasks that are actually not supported, especially shared concept formalization among technical experts.

This paper addresses the designers' work and in particular the way they express an object shape through character lines. Based on this concept, a deformation feature is presented. The proposed approach includes curvilinear constraints to match the designer's requirements and techniques for the quality and accuracy of the produced model. This method is applied to B-spline surfaces coupled with a mechanical model of a bar network; geometric constraints are automatically and adaptively added to monitor and control the deformation process. A few examples show the results obtained by this deformation feature with the help of curvature maps, used to analyze the influence of the constraints and the resulting quality.

Within the scope of simulation of assembly/disassembly operations, flexible modelling for virtual reality environments is described. The study analyzes the adequacy of various mechanical models in the specific context of oil hoses. The model behaviour is checked against experiments for validation. As an issue of this comparison, a mechanical model is proposed through the concept of a dual mechanical approach. The adequacy of this last model is deduced from the experimental data. The mechanical model is coupled with a geometric model for free-form surfaces. The deformation criterion uses a functional based on the external forces applied to the mechanical model. This model is expressed as a prescribed displacement problem in order to obtain a more accurate model. The model proposed is also capable of handling interaction between the flexible area of the hose and rigid parts of its environment which model other components of an engine bay. To this end, the mechanical model is enriched with a contact boundary condition which preserves the coherence of the approach. The proposed approach is also a proposal for a methodology to model mechanical flexible parts for virtual reality applications. The model set up has been inserted into a virtual reality environment for efficiency evaluation.