-
公开(公告)号：US10427305B2
公开(公告)日：2019-10-01
申请号：US15216583
申请日：2016-07-21
Applicant: Autodesk, Inc.
Inventor： Evan Patrick Atherton , David Thomasson , Maurice Ugo Conti , Heather Kerrick
IPC: G05B15/00 , B25J9/16 , B25J19/02 , G06K9/00 , H04N5/232 , G05D1/00 , G06T7/246 , G06F3/01 , H04N5/247
Abstract: A motion capture setup records the movements of an operator, and a control engine then translates those movements into control signals for controlling a robot. The control engine may directly translate the operator movements into analogous movements to be performed by the robot, or the control engine may compute robot dynamics that cause a portion of the robot to mimic a corresponding portion of the operator.
-
公开(公告)号：US10363667B2
公开(公告)日：2019-07-30
申请号：US15363956
申请日：2016-11-29
Applicant: Autodesk, Inc.
Inventor： Evan Atherton , David Thomasson , Heather Kerrick , Maurice Conti
Abstract: One embodiment of the present invention sets forth a technique for determining a location of an object that is being manipulated or processed by a robot. The technique includes capturing a digital image of the object while the object is disposed by the robot within an imaging space, wherein the digital image includes a direct view of the object and a reflected view of the object, detecting a visible feature of the object in the direct view and the visible feature of the object in the reflected view, and computing a first location of the visible feature in a first direction based on a position of the visible feature in the direct view. The technique further includes computing a second location of the visible feature in a second direction based on a position of the visible feature in the reflected view and causing the robot to move the object to a processing station based at least in part on the first location and the second location.
-
公开(公告)号：US10955814B2
公开(公告)日：2021-03-23
申请号：US15495945
申请日：2017-04-24
Applicant: AUTODESK, INC.
Inventor： Evan Atherton , David Thomasson , Maurice Ugo Conti , Heather Kerrick , Nicholas Cote
IPC: G05B19/29 , B23K9/04 , B33Y10/00 , B33Y50/02 , G05B19/4099 , B29C64/386
Abstract: A robot system is configured to fabricate three-dimensional (3D) objects using closed-loop, computer vision-based control. The robot system initiates fabrication based on a set of fabrication paths along which material is to be deposited. During deposition of material, the robot system captures video data and processes that data to determine the specific locations where the material is deposited. Based on these locations, the robot system adjusts future deposition locations to compensate for deviations from the fabrication paths. Additionally, because the robot system includes a 6-axis robotic arm, the robot system can deposit material at any locations, along any pathway, or across any surface. Accordingly, the robot system is capable of fabricating a 3D object with multiple non-parallel, non-horizontal, and/or non-planar layers.
-
公开(公告)号：US10751879B2
公开(公告)日：2020-08-25
申请号：US15995005
申请日：2018-05-31
Applicant: AUTODESK, INC.
Inventor： Hui Li , Evan Patrick Atherton , Erin Bradner , Nicholas Cote , Heather Kerrick
Abstract: One embodiment of the present invention sets forth a technique for controlling the execution of a physical process. The technique includes receiving, as input to a machine learning model that is configured to adapt a simulation of the physical process executing in a virtual environment to a physical world, simulated output for controlling how the physical process performs a task in the virtual environment and real-world data collected from the physical process performing the task in the physical world. The technique also includes performing, by the machine learning model, one or more operations on the simulated output and the real-world data to generate augmented output. The technique further includes transmitting the augmented output to the physical process to control how the physical process performs the task in the physical world.
-
公开(公告)号：US20200264583A1
公开(公告)日：2020-08-20
申请号：US16866442
申请日：2020-05-04
Applicant: Autodesk, Inc.
Inventor： Evan Patrick Atherton , David Thomasson , Maurice Ugo Conti , Heather Kerrick
IPC: G05B19/402 , G05B19/18
Abstract: A robot is configured to assist an end-user with creative tasks. While the end-user modifies the work piece, the robot observes the modifications made by the end-user and determines one or more objectives that the end-user may endeavor to accomplish. The robot then determines a set of actions to perform that assist the end-user with accomplishing the objectives.
-
Patent Agency Ranking
公开(公告)号：US10444716B2
公开(公告)日：2019-10-15
申请号：US15089409
申请日：2016-04-01
Applicant: Autodesk, Inc.
Inventor： Florencio Mazzoldi , Olivier Dionne , Thomas White , Heather Kerrick , Christopher C. Romes
Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for passing actionable information between different buildings to facilitate building management without human intervention include, in one aspect, a method including: determining, in a building information modelling (BIM) system of a first building, a set of rules defining actions to be taken by a building automation system of the first building in response to a defined set of remote information received from a BIM system of a second building, the set of remote information corresponding to one or more sensors in or associated with the second building; receiving data from the BIM system of the second building in accordance with the set of remote information; and using the building automation system of the first building to automatically change configuration, use, or operation of the first building in response to the received data in accordance with the set of rules.
-
公开(公告)号：US20180065247A1
公开(公告)日：2018-03-08
申请号：US15258957
申请日：2016-09-07
Applicant: Autodesk, Inc.
Inventor： Evan Patrick Atherton , David Thomasson , Heather Kerrick , Maurice Ugo Conti
IPC: B25J9/16 , G05B19/042
CPC classification number: B25J9/163 , G05B19/042 , G05B2219/40264 , G05B2219/40391
Abstract: A control engine is trained to operate a robotic camera according to a variety of different cinematographic techniques. The control engine may reconfigure the robotic camera to respond to a set of cues, to enforce a set of constraints, or to apply one or more characteristic styles. A training engine trains a network within the control engine based on training data that exemplifies cue responses, enforced constraints, and characteristic styles.
-
公开(公告)号：US20160342937A1
公开(公告)日：2016-11-24
申请号：US14720559
申请日：2015-05-22
Applicant: Autodesk, Inc.
Inventor： Heather Kerrick
CPC classification number: G06Q10/087 , G06Q30/0643
Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for identifying, inventorying and managing physical possessions. Image data is captured including one or more images or video of physical possessions of a person that occupy a space. The captured image data is automatically analyzed to recognize physical products included in the image data and to determine a position of each recognized physical product in the space. The recognized physical products are automatically inventoried including preparing a list of products. Each entry in the list includes a product name or identifying information and location information for where the product is located in the space. A representation of the space is generated including respective physical products. A user interface is provided for presenting the inventory including providing the representation of the space and displaying a representation of respective physical products in the space.
Abstract translation: 方法，系统和装置包括在计算机可读存储介质上编码的计算机程序，包括用于识别，盘点和管理物理财产的方法。 拍摄图像数据，包括占用空间的人的物理财产的一个或多个图像或视频。 捕获的图像数据被自动分析以识别包括在图像数据中的物理产品并且确定每个识别的物理产品在该空间中的位置。 公认的物理产品被自动盘点，包括准备产品清单。 列表中的每个条目都包括产品名称或产品位于该空间中的标识信息和位置信息。 产生空间的表示，包括各自的物理产品。 提供用于呈现库存的用户界面，包括提供空间的表示并在空间中显示各个物理商品的表示。
-
公开(公告)号：US11953879B2
公开(公告)日：2024-04-09
申请号：US17015000
申请日：2020-09-08
Applicant: AUTODESK, INC.
Inventor： Evan Patrick Atherton , David Thomasson , Maurice Ugo Conti , Heather Kerrick , Nicholas Cote , Hui Li
IPC: G05B19/4099 , B22D23/00 , B33Y50/00 , B23K9/04
CPC classification number: G05B19/4099 , B22D23/003 , B33Y50/00 , B23K9/044 , G05B2219/49023 , G06T2219/008
Abstract: An agent engine allocates a collection of agents to scan the surface of an object model. Each agent operates autonomously and implements particular behaviors based on the actions of nearby agents. Accordingly, the collection of agents exhibits swarm-like behavior. Over a sequence of time steps, the agents traverse the surface of the object model. Each agent acts to avoid other agents, thereby maintaining a relatively consistent distribution of agents across the surface of the object model over all time steps. At a given time step, the agent engine generates a slice through the object model that intersects each agent in a group of agents. The slice associated with a given time step represents a set of locations where material should be deposited to fabricate a 3D object. Based on a set of such slices, a robot engine causes a robot to fabricate the 3D object.
-
公开(公告)号：US11679506B2
公开(公告)日：2023-06-20
申请号：US17691838
申请日：2022-03-10
Applicant: AUTODESK, INC.
Inventor： Hui Li , Evan Patrick Atherton , Erin Bradner , Nicholas Cote , Heather Kerrick
CPC classification number: B25J9/1671 , B25J9/161 , B25J9/163 , B25J9/1605 , G05B19/41885 , G06F30/20 , G06N3/044 , G06N3/08 , G06N20/00 , G06T17/00 , G05B2219/32017 , G05B2219/35353
Abstract: One embodiment of the present invention sets forth a technique for generating simulated training data for a physical process. The technique includes receiving, as input to at least one machine learning model, a first simulated image of a first object, wherein the at least one machine learning model includes mappings between simulated images generated from models of physical objects and real-world images of the physical objects. The technique also includes performing, by the at least one machine learning model, one or more operations on the first simulated image to generate a first augmented image of the first object. The technique further includes transmitting the first augmented image to a training pipeline for an additional machine learning model that controls a behavior of the physical process.
-
-
-
-
-
-
-
-
-
Search Results
34
records
(took 0.015 seconds)
