Apparatus and method for boundary learning optimization

    公开(公告)号:US12159226B2

    公开(公告)日:2024-12-03

    申请号:US17139800

    申请日:2020-12-31

    Abstract: A boundary learning optimization tool for training neural networks with accurate models of parameterized flexure using a minimal number of numerically generated performance solutions generated from different design instantiations of those topologies. Performance boundaries are output by the neural network in optimization steps, with geometric parameters varied from smallest allowable feature sizes to largest geometrically compatible feature sizes for given constituent materials. The plotted performance boundaries define the design spaces of flexure systems toward allowing designers to visually identify which geometric versions of their synthesized topologies best achieve desired combinations of performance capabilities.

    Stable shape-reconfigurable structures and mechanisms

    公开(公告)号:US11193551B2

    公开(公告)日:2021-12-07

    申请号:US15498648

    申请日:2017-04-27

    Abstract: In one embodiment, a shape-reconfigurable structure includes a rigid base having first and second ends, a rigid first beam having a lateral end and a central end, the lateral end being connected to the first end of the base, and a rigid second beam having a lateral end and a central end, the lateral end of the second beam being connected to the second end of the base and the central end of the second beam being connected to the central end of the first beam, wherein the structure can be placed in an expanded orientation in which the first and second beams extend outward away from the base and a contracted orientation in which the first and second beams extend inward toward the base.

    Systems and methods for micromechanical displacement-based logic circuits

    公开(公告)号:US10855259B1

    公开(公告)日:2020-12-01

    申请号:US16808629

    申请日:2020-03-04

    Abstract: The present disclosure relates to a micromechanical displacement logic, signal propagation system that makes use of first and second bistable elements, and first and second mounting structures arranged adjacent opposing surfaces of the first bistable element. A plurality of pivotal lever arms are used to support the first bistable element in either one of two positions of equilibrium. A support structure and a compressible flexure element disposed between the support structure and the first mounting structure apply a preload force to the first mounting structure, which imparts the preload force to the first bistable element. The first bistable element is moveable from one of the two stable equilibrium positions to the other in response to an initial signal applied thereto. The preload force, at least one stiffness characteristic of the lever arms, and a compressibility of a compressible coupling element which links the second bistable element to the first, are all selected to tune signal propagation from the first bistable element to the second bistable element.

    APPARATUS AND METHOD FOR BOUNDARY LEARNING OPTIMIZATION

    公开(公告)号:US20210241100A1

    公开(公告)日:2021-08-05

    申请号:US17139800

    申请日:2020-12-31

    Abstract: A boundary learning optimization tool for training neural networks with accurate models of parameterized flexure using a minimal number of numerically generated performance solutions generated from different design instantiations of those topologies. Performance boundaries are output by the neural network in optimization steps, with geometric parameters varied from smallest allowable feature sizes to largest geometrically compatible feature sizes for given constituent materials. The plotted performance boundaries define the design spaces of flexure systems toward allowing designers to visually identify which geometric versions of their synthesized topologies best achieve desired combinations of performance capabilities.

    Stable Shape-Reconfigurable Structures And Mechanisms

    公开(公告)号:US20200025272A1

    公开(公告)日:2020-01-23

    申请号:US15498648

    申请日:2017-04-27

    Abstract: In one embodiment, a shape-reconfigurable structure includes a rigid base having first and second ends, a rigid first beam having a lateral end and a central end, the lateral end being connected to the first end of the base, and a rigid second beam having a lateral end and a central end, the lateral end of the second beam being connected to the second end of the base and the central end of the second beam being connected to the central end of the first beam, wherein the structure can be placed in an expanded orientation in which the first and second beams extend outward away from the base and a contracted orientation in which the first and second beams extend inward toward the base.

    MULTI-STABLE COMPLIANT-MECHANISM MATTRESS FOR BED SORE PREVENTION

    公开(公告)号:US20220175601A1

    公开(公告)日:2022-06-09

    申请号:US17585271

    申请日:2022-01-26

    Abstract: A pressure sore prevention mattress is provided alternating pressure points on the body of an occupant that is easily transformable between two stable configurations. The four-layer mattress construction has a top foam pad, a webbing layer, a dynamic layer of deformable panels trapezoid-shaped four-bar mechanisms that deform to produce an undulating pattern of peaks, and a base. Panel geometry can determine peak height. The mattress has a spatially hollow geometry that allows natural air convection through the breathable top foam pad to enable drying of moisture as well as the use of forced air convective drying or cooling. The mattress can be transformed from one configuration to the other while the occupant is on the mattress.

Patent Agency Ranking