-
41.
公开(公告)号:WO2022148565A1
公开(公告)日:2022-07-14
申请号:PCT/EP2021/080253
申请日:2021-11-01
Applicant: VOLKSWAGEN AKTIENGESELLSCHAFT
Inventor: MÜNNING, Daniel , BASSE, Daniel
IPC: G08G1/0967 , G08G1/0962 , B60W50/14 , H04W4/44 , H04W4/46 , G06N3/00
Abstract: ie Erfindung betrifft ein Verfahren zum Betreiben eines Assistenzsystems (2) eines Kraftfahrzeugs (1), bei welchem mittels zumindest einer Erfassungseinrichtung (3) des Assistenzsystems (2) eine Umgebung (9) des Kraftfahrzeugs (1) erfasst wird und in Abhängigkeit von der erfassten Umgebung (9) ein Umgebungsmodell (10) mittels einer elektronischen Recheneinrichtung (8) des Assistenzsystems (2) erzeugt wird, wobei das Umgebungsmodell (10) auf einer Anzeigeeinrichtung (11) des Assistenzsystems (2) angezeigt wird, mittels einer Schwarmdatenempfangseinrichtung (4) des Assistenzsystems (2) Schwarmdaten (5) bezüglich einer Vielzahl von weiteren Kraftfahrzeugen (6) empfangen werden und auf Basis der Schwarmdaten (5) zumindest eine Umgebungsinformation (12, 13, 14, 15) erzeugt wird und die Umgebungsinformation (12, 13, 14, 15) bei der Erzeugung des Umgebungsmodell (10) berücksichtigt wird und mit auf der Anzeigeeinrichtung (11) angezeigt wird. Ferner betrifft die Erfindung ein Computerprogrammprodukt sowie ein Assistenzsystem (2).
-
42.
公开(公告)号:WO2022147583A2
公开(公告)日:2022-07-07
申请号:PCT/US2022/026730
申请日:2022-04-28
Applicant: FUTUREWEI TECHNOLOGIES, INC.
Inventor: MORTAZAVI, Masood Seyed , YAN, Ning , AU YEUNG, Adrian , VAN, Richard , CHEN, Ziru , MENG, Lin
IPC: G06F16/906 , G06F30/27 , G06F30/337 , G06F30/392 , G06N3/04 , G06N3/00 , G06N7/00 , G06N3/08 , G06N3/006 , G06N3/0445 , G06N3/0454 , G06N3/084 , G06N7/005
Abstract: According to embodiments, a reinforcement learning (RL) agent running on at least one processor receives a constant input. The RL agent includes a deep neural network (DNN), and the DNN may include one or more layers of bi-directional gated recurrent units (Bi-GRUs) and one or more self-attention or transformer layers. The RL agent outputs N sets of distributions. Each set of the N sets of distributions corresponds to a different object of the N interacting objects and includes a first distribution in a first direction and a second distribution in a second direction. The RL agent generates a batch of sample configurations based on the N sets of distributions. Each sample configuration of the batch includes identified or selected locations of the N interacting objects. The RL agent outputs the batch of sample configurations to an evaluator. The RL agent updates parameters of the DNN based on a loss function.
-
公开(公告)号:WO2022144796A1
公开(公告)日:2022-07-07
申请号:PCT/IB2021/062420
申请日:2021-12-29
Applicant: SOPHIE'S BIONUTRIENTS PTE. LTD.
Inventor: WANG, Yao-Hsin , TSUEI, Kirin
Abstract: A method executed by an engine of a computing device for simulating microalgae fermentation is described. The engine receives an input from a user. The input includes an identification of a microalgae, an identification of a culture media, an identification of an enclosure, and an identification of fermentation conditions for the microalgae when the microalgae is located in the culture media and when the microalgae and the culture media are located in the enclosure. An algorithm of the engine is used to simulate fermentation of the microalgae. A result of the simulation is displayed to the user.
-
公开(公告)号:WO2022132615A1
公开(公告)日:2022-06-23
申请号:PCT/US2021/063020
申请日:2021-12-13
Applicant: MICRON TECHNOLOGY, INC.
Inventor: RUSSO, Kathryn H. , SIMSEK-EGE, Fatma Arzum , YAN, Yixin , GHOSH, Gitanjali T. , WANG, Libo
Abstract: Methods, devices, and systems related to managing energy using artificial intelligence (AI) are described. In an example, a method can include receiving first signaling including data representing an energy input at a processing resource of a computing device from a radio in communication with a processing resource of an energy source, receiving second signaling including user data at the processing resource of the computing device from a memory of the computing device, inputting the user data and the data representing the energy input into an AI model at the processing resource of the computing device, generating data representing a command as an output of the AI model, and transmitting third signaling including the data representing the command to the processing resource of the energy source from the processing resource of the computing device via the radio in communication with the processing resource of the energy source.
-
公开(公告)号:WO2022131985A1
公开(公告)日:2022-06-23
申请号:PCT/SE2021/050062
申请日:2021-01-29
Applicant: SWEGREEN AB
Inventor: DALIN, Linus
IPC: A01G9/24 , A01G31/04 , A01G9/18 , F24F3/14 , F24F11/30 , G05B13/00 , G05B13/04 , G05B19/00 , G06N3/00 , G06N7/00 , G06N20/00 , G06Q50/02 , A01G9/246 , A01G9/26 , F24F13/22 , F24F3/044
Abstract: The application relates to a method for optimization of plant and driving parameters for a cultivation plant (1) comprising a cultivation room (1 a), an air circuit (2), a water circuit (3), a heat circuit (4), a light circuit (5), a nutrition circuit (6), a spacing circuit (7), a control unit (8) for controlling the circuits, and a plant yield control unit (10) for determining growth and/or quality of plants, where the method comprises training the control unit (8) using a training cultivation plant (1) with a training control unit (8), corresponding circuits (2, 3, 4, 5, 6, 7) and plant yield control unit (10). The training control unit (8) analyses which driving parameters that have a positive or negative effect on plant growth and/or quality, and changes driving parameters until a predetermined level of plant growth and/or quality is attained, upon which the updated driving parameters are sent to the control unit (8) for use in the cultivation plant (1). The application also relates to a cultivation plant (1) comprising a cultivation room (1a) and an adjacent facility (11) that are arranged to exchange resources such as air and heat.
-
公开(公告)号:WO2022129867A1
公开(公告)日:2022-06-23
申请号:PCT/GB2021/053189
申请日:2021-12-07
Applicant: UCL BUSINESS LTD
Inventor: MANESCU, Petru , FERNANDEZ-REYES, Delmiro
Abstract: A computer implemented method of controlling a microscope (632) is provided. The method comprises capturing an image (631) within a field of view of a lens of the microscope (632) configured to view a sample on a motorised stage (633) of the microscope (632). The image comprises a portion of the sample. The image (631) is provided to an artificial neural network (610). An action (611) for moving the motorised stage (633) is determined in dependence on an output of the artificial neural network (610). The motorised stage (633) is moved automatically in accordance with the action (611).
-
公开(公告)号:WO2022127425A1
公开(公告)日:2022-06-23
申请号:PCT/CN2021/128829
申请日:2021-11-04
Applicant: 杭州大拿科技股份有限公司
Abstract: 本公开涉及一种题目辅助方法、装置和系统,所述题目辅助方法包括:获取题目影像,并根据题目影像识别出题目内容;根据题目内容产生解题答案和解题过程,其中,解题过程包括解题步骤或者解题过程包括解题步骤和至少部分解题步骤的步骤说明;将解题过程设置在相应的显示层级中;以及根据默认的显示规则和接收到的显示指令中的至少一个来显示对应的显示层级。
-
公开(公告)号:WO2022120200A2
公开(公告)日:2022-06-09
申请号:PCT/US2021/061855
申请日:2021-12-03
Applicant: GOOGLE LLC
Inventor: AHARONI, Asaf , LEVIATHAN, Yaniv , SEGALIS, Eyal , ELIDAN, Gal , GOLDSHTEIN, Sasha , AMIAZ, Tomer , COHEN, Deborah
IPC: G06N3/08 , G06F40/56 , G06F40/35 , H04M3/493 , G10L13/027 , G06F3/16 , H04L51/02 , G06Q30/00 , G10L25/30 , G10L15/18 , G06N3/04 , G06N3/00 , G06F16/3329 , G06F3/167 , G06F40/279 , G06N3/0454 , G06N3/084 , G06Q10/02 , G06Q10/10 , G06Q30/016 , G06Q30/0601 , G06Q50/12 , G06Q50/32 , G10L15/1822 , H04M3/4936
Abstract: Implementations are directed to providing a voice bot development platform that enables a third-party developer to train a voice bot based on training instance(s). The training instance(s) can each include training input and training output. The training input can include a portion of a corresponding conversation and a prior context of the corresponding conversation. The training output can include a corresponding ground truth response to the portion of the corresponding conversation. Subsequent to training, the voice bot can be deployed for conducting conversations on behalf of a third-party. In some implementations, the voice bot is further trained based on a corresponding feature emphasis input that attentions the voice bot to a particular feature of the portion of the corresponding conversation. In some additional or alternative implementations, the voice bot is further trained to interact with third-party system(s) via remote procedure calls (RPCs).
-
公开(公告)号:WO2022119580A1
公开(公告)日:2022-06-09
申请号:PCT/US2020/064722
申请日:2020-12-12
Applicant: GOOGLE LLC
Inventor: AHARONI, Asaf , LEVIATHAN, Yaniv , SEGALIS, Eyal , ELIDAN, Gal , GOLDSHTEIN, Sasha , AMIAZ, Tomer , COHEN, Deborah
IPC: G06N3/08 , G06F40/56 , G06F40/35 , H04L12/58 , H04M3/493 , G10L13/027 , G06F3/16 , G06Q30/00 , G10L25/30 , G10L15/18 , G06N3/04 , G06N3/00
Abstract: Implementations are directed to providing a voice bot development platform that enables a third-party developer to train a voice bot based on training instance(s). The training instance(s) can each include training input and training output. The training input can include a portion of a corresponding conversation and a prior context of the corresponding conversation. The training output can include a corresponding ground truth response to the portion of the corresponding conversation. Subsequent to training, the voice bot can be deployed for conducting conversations on behalf of a third-party. In some implementations, the voice bot is further trained based on a corresponding feature emphasis input that attentions the voice bot to a particular feature of the portion of the corresponding conversation. In some additional or alternative implementations, the voice bot is further trained to interact with third-party system(s) via remote procedure calls (RPCs).
-
公开(公告)号:WO2022106858A2
公开(公告)日:2022-05-27
申请号:PCT/GB2021/053033
申请日:2021-11-23
Applicant: BOTSANDUS LTD
Inventor: DANESCU, Andrei , NEGOITA, Adrian , MACLEOD, Matthew
Abstract: A method for measuring and analysing packages particularly for optimising storage in a warehouse or for packing for transporting. The method follows the steps of placing packages on a floor surface moving a robot to adjacent one of the packages. Lidar or similar is used to measuring distance data from the robot to the package at a multiple heights from at least 0.5m above the floor surface. The robot then follows a path around all or part of the packages gathering further distance data as it travels. The distance data is then used to determine or infer the shape of the sides of the package and from their estimate volumetric dimensions of the package. The volumetric dimensions are then used to optimise storage of the package with other packages.
-
-
-
-
-
-
-
-
-