-
公开(公告)号：US11551037B2
公开(公告)日：2023-01-10
申请号：US16388958
申请日：2019-04-19
发明人： Stefan Gavranovic , Dirk Hartmann
摘要： Provided is a method for determining a physical shape having a predefined physical target property that includes calculating a sensitivity landscape on the basis of a shape data record for the physical shape with the aid of a calculation device. The calculation device is a machine-taught artificial intelligence device. The shape data record identifies locations at or on the physical shape. For a plurality of these locations, the sensitivity landscape respectively indicates how the target property of the physical shape changes if the physical shape changes in the region of the location. Furthermore, the shape data record for the physical shape to be determined is changed on the basis of the sensitivity landscape in such a manner that the predefined physical target property is improved.
-
公开(公告)号：US20190031204A1
公开(公告)日：2019-01-31
申请号：US15963240
申请日：2018-04-26
摘要： A method for performing an optimized control of a complex dynamical system using machine learned, scenario based control heuristics including: providing a simulation model for predicting a system state vector of the dynamical system in time based on a current scenario parameter vector and a control vector; using a Model Predictive Control, MPC, algorithm to provide the control vector during a simulation of the dynamical system using the simulation model for different scenario parameter vectors and initial system state vectors; calculating a scenario parameter vector and initial system state vector a resulting optimal control value by the MPC algorithm; generating machine learned control heuristics approximating the relationship between the corresponding scenario parameter vector and the initial system state vector for the resulting optimal control value using a machine learning algorithm; and using the generated machine learned control heuristics to control the complex dynamical system modelled by the simulation model.
-
公开(公告)号：US11567470B2
公开(公告)日：2023-01-31
申请号：US17310300
申请日：2019-12-17
发明人： Dirk Hartmann , David Bitterolf , Hans-Georg Köpken , Birgit Obst , Florian Ulli Wolfgang Schnös , Sven Tauchmann
IPC分类号： G05B19/402 , G05B19/31
摘要： In order to be able to take into account machining configurations more flexibly, a method for optimizing numerically controlled machining of a workpiece includes ascertaining geometric interaction data. A relationship between a force to be expected and a configuration parameter of the machining is determined on the basis of the interaction data. The force is calculated during the machining on the basis of the relationship and a current value of the at least one configuration parameter. The machining is adapted depending on the calculated force.
-
公开(公告)号：US20220382265A1
公开(公告)日：2022-12-01
申请号：US17775839
申请日：2020-11-18
发明人： Dirk Hartmann , Michael Jaentsch , Tobias Kamps , Birgit Obst , Daniel Regulin , Florian Ulli Wolfgang Schnös , Sven Tauchmann
IPC分类号： G05B19/418
摘要： A method for operating a numerical controlled machine comprising receiving a sequence of control commands which, when executed by a numerical controlled machine, cause the numerical controlled machine to machine a workpiece to obtain a predetermined workpiece geometry, wherein the sequence of control commands includes while machining the workpiece based on the received sequence of control commands measuring a value of a first interaction parameter for a first position of the tool, comparing a measured value of the first interaction parameter for the first position of the tool with the simulated value of the first interaction parameter for the first position of the tool, and determining an adapted value of the second interaction parameter for a following position of the tool based on a result of the comparison.
-
公开(公告)号：US20220284153A1
公开(公告)日：2022-09-08
申请号：US17632481
申请日：2019-08-28
发明人： Stefan Gavranovic , Suraj Ravi Musuvathy , Dirk Hartmann , Peter Nanson , Richard Collins , Hiren Dedhia
摘要： A computing system may include a geometry access engine configured to access geometries associated with a topology optimization process, including an original geometry that represents a design space upon which the topology optimization process applies to as well as a topology optimized geometry that represents an output of the topology optimization process performed for the original geometry. The system may also include geometry processing engine configured to generate a final geometry from the topology optimized geometry, including by conforming the topology optimized geometry to the original geometry at portions of the topology optimized geometry that correspond to fixed regions of the original geometry as well as smoothing the topology optimized geometry at portions that correspond to non-fixed regions of the original geometry.
-
6.发明申请公开(公告)号：US20220250324A1
公开(公告)日：2022-08-11
申请号：US17622264
申请日：2020-05-15
IPC分类号： B29C64/35 , B29C64/393 , B22F10/68 , B22F10/85 , B33Y40/20 , B33Y50/02 , B08B13/00 , B08B7/02
摘要： To separate excess material, the component moved by a movement device that is controlled by movement data, and a fill level of the component with material is measured. A process for emptying material from the component simulated for each different initial fill level with material, wherein movement data, which specify a simulated movement of the component, and a simulated fill level progression resulting from the simulated movement are assigned to the associated initial fill level. In addition, a corresponding initial fill level is selected in accordance with the measured fill level, and the movement device is controlled by movement data which are assigned to the selected initial fill level. The fill level is measured and compared to a simulated fill level progression assigned to the selected initial fill level. The steps of selecting a corresponding initial fill level (SAFG) and controlling the movement device (BV) are carried out.
-
公开(公告)号：US10953891B2
公开(公告)日：2021-03-23
申请号：US15963240
申请日：2018-04-26
摘要： A method using machine learned, scenario based control heuristics including: providing a simulation model for predicting a system state vector of the dynamical system in time based on a current scenario parameter vector and a control vector; using a Model Predictive Control, MPC, algorithm to provide the control vector during a simulation of the dynamical system using the simulation model for different scenario parameter vectors and initial system state vectors; calculating a scenario parameter vector and initial system state vector a resulting optimal control value by the MPC algorithm; generating machine learned control heuristics approximating the relationship between the corresponding scenario parameter vector and the initial system state vector for the resulting optimal control value using a machine learning algorithm; and using the generated machine learned control heuristics to control the complex dynamical system modelled by the simulation model.
-
公开(公告)号：US20190325270A1
公开(公告)日：2019-10-24
申请号：US16388958
申请日：2019-04-19
发明人： Stefan Gavranovic , Dirk Hartmann
摘要： Provided is a method for determining a physical shape having a predefined physical target property that includes calculating a sensitivity landscape on the basis of a shape data record for the physical shape with the aid of a calculation device. The calculation device is a machine-taught artificial intelligence device. The shape data record identifies locations at or on the physical shape. For a plurality of these locations, the sensitivity landscape respectively indicates how the target property of the physical shape changes if the physical shape changes in the region of the location. Furthermore, the shape data record for the physical shape to be determined is changed on the basis of the sensitivity landscape in such a manner that the predefined physical target property is improved.
-
9.发明申请公开(公告)号：US20220347923A1
公开(公告)日：2022-11-03
申请号：US17620869
申请日：2020-05-15
IPC分类号： B29C64/35 , B29C64/153 , B29C64/393 , B33Y40/00 , B33Y50/02
摘要： In order to separate excess material from an additively manufactured component, spatially resolved structural data on the component are received. On the basis of the structural data, a process for emptying material from the component is simulated, wherein a sequence of emptying poses of the component is determined. For an associated emptying pose: the component is moved into the associated emptying pose in accordance with the simulated emptying process, movement of material is detected by sensors, as a result of a detection of a decrease in the movement of material, a trigger signal is generated, and a movement of the component into a subsequent emptying pose is initiated by the trigger signal, the trigger signal being considered higher priority than the simulated emptying process.
-
公开(公告)号：US20220180027A1
公开(公告)日：2022-06-09
申请号：US17601480
申请日：2020-03-23
发明人： Dirk Hartmann , Sanjeev Srivastava
IPC分类号： G06F30/27
摘要： A computer-implemented method and apparatus for generating a design for a technical system or a product is provided. Depending on a set of first parameters, specifying physical properties, and second parameters, specifying perceptible properties of the technical system or product, a design is generated for the technical system or product. A performance indicator that evaluates a physical performance of the generated design is obtained. The generated design of the technical system or product is presented to a user and perception data in response to the presentation of the generated design are measured by a perception capturing unit and a perception evaluation indicator is deduced from the measured perception data. An optimized design is determined by iteratively optimizing the performance indicator and/or the perception evaluation indicator by an optimization algorithm. The method and apparatus enable an autonomous closed design loop taking human perception into account.
-
-
-
-
-
-
-
-
-
搜索结果
14 条记录 (耗时 0.013 秒)
