The Experts below are selected from a list of 135 Experts worldwide ranked by ideXlab platform
Lionel Fine - One of the best experts on this subject based on the ideXlab platform.
-
Idealized models for FEA derived from generative modeling processes based on extrusion primitives
Engineering with Computers, 2015Co-Authors: Flavien Boussuge, Jean-claude Léon, Stefanie Hahmann, Lionel FineAbstract:Shape idealization transformations are very common operations when adapting a CAD component to FEA requirements. Here, an idealization approach is proposed that is based on generative shape processes used to decompose an initial B-Rep solid, i.e., extrusion processes with material addition are used to segment a solid. The corresponding extrusion primitives form the basis of candidate sub-domains for idealization and their connections conveyed through the generative processes they belong to, bringing robustness to set up the appropriate connections between idealized sub-domains. This is made possible because the connections between extrusion primitives have an explicit geometric representation and can be used to bound the connections between idealized sub-domains. Taking advantage of an existing Construction Tree as available in a CAD software does not help much because it may be complicated to use it for idealization processes because this Tree structure is not unique. Using generative processes attached to an object that is no longer reduced to a single Construction Tree but to a graph containing all non-trivial Construction Trees, is more useful for the engineer to evaluate variants of idealization. From this automated decomposition, each primitive is subjected to a morphological analysis to define whether it can idealized or not. Subsequently, geometric interfaces between primitives form also a graph that can be used to process the connections between the idealized sub-domains generated from the primitives. These interfaces are taken into account to determine more precisely the idealizable sub-domains and their contours when primitives are incrementally merged to come back to produce the global morphological analysis of the initial object. A user-defined threshold is used to tune the morphological analysis with respect to further user parameters. Finally, the idealizable sub-domains and their connections are processed to locate the mid-surfaces and connect them using generic criteria that the user can tune locally using complementary criteria.
-
IMR - Idealized models for FEA derived from generative modeling processes based on extrusion primitives
Engineering With Computers, 2014Co-Authors: Flavien Boussuge, Jean-claude Léon, Stefanie Hahmann, Lionel FineAbstract:Shape idealization transformations are very common operations when adapting a CAD component to FEA requirements. Here, an idealization approach is proposed that is based on generative shape processes used to decompose an initial B-Rep solid, i.e., extrusion processes with material addition are used to segment a solid. The corresponding extrusion primitives form the basis of candidate sub-domains for idealization and their connections conveyed through the generative processes they belong to, bringing robustness to set up the appropriate connections between idealized sub-domains. This is made possible because the connections between extrusion primitives have an explicit geometric representation and can be used to bound the connections between idealized sub-domains. Taking advantage of an existing Construction Tree as available in a CAD software does not help much because it may be complicated to use it for idealization processes because this Tree structure is not unique. Using generative processes attached to an object that is no longer reduced to a single Construction Tree but to a graph containing all non-trivial Construction Trees, is more useful for the engineer to evaluate variants of idealization. From this automated decomposition, each primitive is subjected to a morphological analysis to define whether it can idealized or not. Subsequently, geometric interfaces between primitives form also a graph that can be used to process the connections between the idealized sub-domains generated from the primitives. These interfaces are taken into account to determine more precisely the idealizable sub-domains and their contours when primitives are incrementally merged to come back to produce the global morphological analysis of the initial object. A user-defined threshold is used to tune the morphological analysis with respect to further user parameters. Finally, the idealizable sub-domains and their connections are processed to locate the mid-surfaces and connect them using generic criteria that the user can tune locally using complementary criteria.
-
Extraction of generative processes from B-Rep shapes and application to idealization transformations
Computer-aided Design, 2013Co-Authors: Flavien Boussuge, Jean-claude Léon, Stefanie Hahmann, Lionel FineAbstract:A Construction Tree is a set of shape generation processes commonly produced with CAD modelers during a design process of B-Rep objects. However, a Construction Tree does not bring all the desired properties in many configurations: dimension modifications, idealization processes, etc. Generating a non trivial set of generative processes, possibly forming a Construction graph, can significantly improve the adequacy of some of these generative processes to meet user's application needs. This paper proposes to extract generative processes from a given B-rep shape as a high-level shape description. To evaluate the usefulness of this description, finite element analyses (FEA) and particularly idealizations are the applications selected to evaluate the adequacy of additive generative processes. Non trivial Construction Trees containing generic extrusion and revolution primitives behave like well established CSG Trees. Advantageously, the proposed approach is primitive-based, which ensures that any generative process of the Construction graph does preserve the realizability of the corresponding volume. In the context of FEA, connections between idealized primitives of a Construction graph can be efficiently performed using their interfaces. Consequently, generative processes of a Construction graph become a high-level object structure that can be tailored to idealizations of primitives and robust connections between them.
O.d. Jones - One of the best experts on this subject based on the ideXlab platform.
-
Galton Watson Fractal Signals
2007 IEEE International Conference on Acoustics Speech and Signal Processing - ICASSP '07, 2007Co-Authors: Geoffrey Decrouez, Pierre-olivier Amblard, Jean-marc Brossier, O.d. JonesAbstract:Iterated function systems (EFS) is a relevant model to produce fractal functions, whether deterministic (with strict self-similarity) or random (self-similar up to probability distribution). The basic idea of such a Construction is to start with an initial function and then compress, dilate and translate it such that by doing so over and over again, we end up with a self-similar signal. This Construction relies on a Construction Tree which has always been deterministic in the literature for signals. Here we introduce new fractals, called Galton Watson fractals, as fixed points of IFS with a random underlying Construction Tree and deterministic operators. We give a proof of the existence and uniqueness of a fixed point at the random and distribution level.
James R. Wilson - One of the best experts on this subject based on the ideXlab platform.
-
Bending and kissing: Computing self-contact configurations of planar loops with revolute joints
2009 IEEE International Conference on Robotics and Automation, 2009Co-Authors: Lee Rudolph, Sam Dorsey-gordon, Dylan Glotzer, Dan Menard, Jon Moran, James R. WilsonAbstract:In recent work, we introduced the notion of a Construction Tree of simplices for a linkage L under distance constraints, and showed that the deformation space DSpace(L) of such an L (i.e., its configuration space CSpace(L) modulo rigid motions of ambient space respecting all system specifications) carries geometrically-defined simplex-based parameters that endow it with a ldquopractically piecewise-convexrdquo structure. Here we present parametrizations of contact deformations of planar loops with revolute joints. We show that the bending and kissing loci, which include the self-contact subspace DSContact (generally as a strict subset) can be efficiently described by triangle-based parameters. These results further demonstrate the effectiveness of the simplex-based approach.
-
ICRA - Bending and kissing: Computing self-contact configurations of planar loops with revolute joints
2009 IEEE International Conference on Robotics and Automation, 2009Co-Authors: Lee Rudolph, Sam Dorsey-gordon, Dylan Glotzer, Dan Menard, Jon Moran, James R. WilsonAbstract:In recent work, we introduced the notion of a Construction Tree of simplices for a linkage L under distance constraints, and showed that the deformation space DSpace(L) of such an L (i.e., its configuration space CSpace(L) modulo rigid motions of ambient space respecting all system specifications) carries geometrically-defined simplex-based parameters that endow it with a “practically piecewise-convex” structure. Here we present parametrizations of contact deformations of planar loops with revolute joints. We show that the bending and kissing loci, which include the self-contact subspace DSContact (generally as a strict subset) can be efficiently described by triangle-based parameters. These results further demonstrate the effectiveness of the simplex-based approach.
Geoffrey Decrouez - One of the best experts on this subject based on the ideXlab platform.
-
Galton?Watson iterated function systems
Journal of Physics A, 2009Co-Authors: Geoffrey Decrouez, Pierre-olivier Amblard, Jean-marc Brossier, Owen JonesAbstract:Iterated function systems (IFS) are interesting parametric models for generating fractal sets and functions. The general idea is to compress, deform and translate a given set or function with a collection of operators and to iterate the procedure. Under weak assumptions, IFS possess a unique fixed point which is in general fractal. IFS were introduced in a deterministic context, then were generalized to the random setting on abstract spaces in the early 1980 s. Their adaptation to random signals was carried out by Hutchinson and R?schendorff [9] by considering random operators. This study extends their model with not only random operators but also a random underlying Construction Tree. We show that the corresponding IFS converges under certain hypothesis to a unique fractal fixed point. Properties of the fixed point are also described.
-
Galton Watson Fractal Signals
2007 IEEE International Conference on Acoustics Speech and Signal Processing - ICASSP '07, 2007Co-Authors: Geoffrey Decrouez, Pierre-olivier Amblard, Jean-marc Brossier, O.d. JonesAbstract:Iterated function systems (EFS) is a relevant model to produce fractal functions, whether deterministic (with strict self-similarity) or random (self-similar up to probability distribution). The basic idea of such a Construction is to start with an initial function and then compress, dilate and translate it such that by doing so over and over again, we end up with a self-similar signal. This Construction relies on a Construction Tree which has always been deterministic in the literature for signals. Here we introduce new fractals, called Galton Watson fractals, as fixed points of IFS with a random underlying Construction Tree and deterministic operators. We give a proof of the existence and uniqueness of a fixed point at the random and distribution level.
-
ICASSP (3) - Galton Watson Fractal Signals
2007 IEEE International Conference on Acoustics Speech and Signal Processing - ICASSP '07, 2007Co-Authors: Geoffrey Decrouez, Pierre-olivier Amblard, Jean-marc Brossier, Owen JonesAbstract:Iterated function systems (EFS) is a relevant model to produce fractal functions, whether deterministic (with strict self-similarity) or random (self-similar up to probability distribution). The basic idea of such a Construction is to start with an initial function and then compress, dilate and translate it such that by doing so over and over again, we end up with a self-similar signal. This Construction relies on a Construction Tree which has always been deterministic in the literature for signals. Here we introduce new fractals, called Galton Watson fractals, as fixed points of IFS with a random underlying Construction Tree and deterministic operators. We give a proof of the existence and uniqueness of a fixed point at the random and distribution level.
Flavien Boussuge - One of the best experts on this subject based on the ideXlab platform.
-
Idealized models for FEA derived from generative modeling processes based on extrusion primitives
Engineering with Computers, 2015Co-Authors: Flavien Boussuge, Jean-claude Léon, Stefanie Hahmann, Lionel FineAbstract:Shape idealization transformations are very common operations when adapting a CAD component to FEA requirements. Here, an idealization approach is proposed that is based on generative shape processes used to decompose an initial B-Rep solid, i.e., extrusion processes with material addition are used to segment a solid. The corresponding extrusion primitives form the basis of candidate sub-domains for idealization and their connections conveyed through the generative processes they belong to, bringing robustness to set up the appropriate connections between idealized sub-domains. This is made possible because the connections between extrusion primitives have an explicit geometric representation and can be used to bound the connections between idealized sub-domains. Taking advantage of an existing Construction Tree as available in a CAD software does not help much because it may be complicated to use it for idealization processes because this Tree structure is not unique. Using generative processes attached to an object that is no longer reduced to a single Construction Tree but to a graph containing all non-trivial Construction Trees, is more useful for the engineer to evaluate variants of idealization. From this automated decomposition, each primitive is subjected to a morphological analysis to define whether it can idealized or not. Subsequently, geometric interfaces between primitives form also a graph that can be used to process the connections between the idealized sub-domains generated from the primitives. These interfaces are taken into account to determine more precisely the idealizable sub-domains and their contours when primitives are incrementally merged to come back to produce the global morphological analysis of the initial object. A user-defined threshold is used to tune the morphological analysis with respect to further user parameters. Finally, the idealizable sub-domains and their connections are processed to locate the mid-surfaces and connect them using generic criteria that the user can tune locally using complementary criteria.
-
IMR - Idealized models for FEA derived from generative modeling processes based on extrusion primitives
Engineering With Computers, 2014Co-Authors: Flavien Boussuge, Jean-claude Léon, Stefanie Hahmann, Lionel FineAbstract:Shape idealization transformations are very common operations when adapting a CAD component to FEA requirements. Here, an idealization approach is proposed that is based on generative shape processes used to decompose an initial B-Rep solid, i.e., extrusion processes with material addition are used to segment a solid. The corresponding extrusion primitives form the basis of candidate sub-domains for idealization and their connections conveyed through the generative processes they belong to, bringing robustness to set up the appropriate connections between idealized sub-domains. This is made possible because the connections between extrusion primitives have an explicit geometric representation and can be used to bound the connections between idealized sub-domains. Taking advantage of an existing Construction Tree as available in a CAD software does not help much because it may be complicated to use it for idealization processes because this Tree structure is not unique. Using generative processes attached to an object that is no longer reduced to a single Construction Tree but to a graph containing all non-trivial Construction Trees, is more useful for the engineer to evaluate variants of idealization. From this automated decomposition, each primitive is subjected to a morphological analysis to define whether it can idealized or not. Subsequently, geometric interfaces between primitives form also a graph that can be used to process the connections between the idealized sub-domains generated from the primitives. These interfaces are taken into account to determine more precisely the idealizable sub-domains and their contours when primitives are incrementally merged to come back to produce the global morphological analysis of the initial object. A user-defined threshold is used to tune the morphological analysis with respect to further user parameters. Finally, the idealizable sub-domains and their connections are processed to locate the mid-surfaces and connect them using generic criteria that the user can tune locally using complementary criteria.
-
Extraction of generative processes from B-Rep shapes and application to idealization transformations
Computer-aided Design, 2013Co-Authors: Flavien Boussuge, Jean-claude Léon, Stefanie Hahmann, Lionel FineAbstract:A Construction Tree is a set of shape generation processes commonly produced with CAD modelers during a design process of B-Rep objects. However, a Construction Tree does not bring all the desired properties in many configurations: dimension modifications, idealization processes, etc. Generating a non trivial set of generative processes, possibly forming a Construction graph, can significantly improve the adequacy of some of these generative processes to meet user's application needs. This paper proposes to extract generative processes from a given B-rep shape as a high-level shape description. To evaluate the usefulness of this description, finite element analyses (FEA) and particularly idealizations are the applications selected to evaluate the adequacy of additive generative processes. Non trivial Construction Trees containing generic extrusion and revolution primitives behave like well established CSG Trees. Advantageously, the proposed approach is primitive-based, which ensures that any generative process of the Construction graph does preserve the realizability of the corresponding volume. In the context of FEA, connections between idealized primitives of a Construction graph can be efficiently performed using their interfaces. Consequently, generative processes of a Construction graph become a high-level object structure that can be tailored to idealizations of primitives and robust connections between them.