-
公开(公告)号：US20230016735A1
公开(公告)日：2023-01-19
申请号：US17837656
申请日：2022-06-10
Applicant: Hitachi, Ltd.
Inventor： Hiroaki ITSUJI , Takumi UEZONO , Kenichi SHIMBO , Tadanobu TOBA
Abstract: Provided is a computer capable of reducing a diagnosis load. For each predetermined diagnosis target node among a plurality of nodes in a neural network, a determination processing unit calculates an expected output value expected as a calculation result of a node calculation process corresponding to the predetermined diagnosis target node, which is obtained when the node calculation process is executed using a predetermined input value. For each diagnosis target node, a generation processing unit generates as a diagnosis program a program for comparing the calculation result of the node calculation process corresponding to the diagnosis target node, which is obtained when the node calculation process is executed by an NN calculation processor using the input value, with the expected output value.
-
公开(公告)号：US20220414207A1
公开(公告)日：2022-12-29
申请号：US17642793
申请日：2020-12-07
Applicant: Hitachi, Ltd.
Inventor： Kenichi SHIMBO , Tadanobu TOBA
IPC: G06F21/55
Abstract: Malware infection or an abnormal operation caused by a malicious attack is detected in real time, even in an electronic device with relatively inferior processing capacity. An electronic device includes an executable code identification unit configured to receive an executable code string output from a processor, and identify at least an execution address in a user program region of an operating system (OS), and an execution address in a kernel region, and a determination unit configured to check a predetermined feature value obtained at a predetermined timing from an identification result obtained by the executable code identification unit, against a predetermined expected value, and determine that an attack happens, in a case where a difference is equal to or larger than a predetermined difference, and the determination unit notifies the processor of a predetermined abnormality notification signal if the determination unit determines that an attack happens.
-
公开(公告)号：US20210357285A1
公开(公告)日：2021-11-18
申请号：US17246940
申请日：2021-05-03
Applicant: Hitachi, Ltd.
Inventor： Hiroaki ITSUJI , Takumi UEZONO , Kenichi SHIMBO , Tadanobu TOBA
Abstract: A program for causing a parallel arithmetic device including a plurality of arithmetic groups to execute parallel arithmetic is input. The program includes information defining each of the following: application arithmetic constituting predetermined processing; redundant arithmetic (which is redundant arithmetic of the application arithmetic and is arithmetic assigned to a surplus core(s) in a diagnosis target arithmetic group); and diagnostic arithmetic (arithmetic that is a comparison of results of the same redundant arithmetic by two or more diagnosis target arithmetic groups and is assigned to surplus cores in an arithmetic group for diagnosis). The surplus core(s) is a core(s) to which no application arithmetic is assigned.
-
4.发明申请公开(公告)号：US20160335145A1
公开(公告)日：2016-11-17
申请号：US15110278
申请日：2014-01-24
Applicant: Hitachi, Ltd.
Inventor： Tadanobu TOBA , Kenichi SHIMBO , Yusuke KANNO , Nobuyasu KANEKAWA , Kotara SHIMAMURA , Hiromichi YAMADA
IPC: G06F11/07 , H03K19/177
CPC classification number: G06F11/079 , G06F11/0727 , G06F11/073 , G06F11/0751 , G06F11/0754 , G06F11/0772 , G06F11/0793 , H03K19/1776 , H03K19/17764
Abstract: The present invention aims to provide a programmable device with a configuration memory that can hold the state of the occurrence abnormal situation that is difficult to assume such as a failure occurring in the programmable device due to the terrestrial radiation of the configuration memory, even during power off, in order to improve the reproducibility in device testing based on the held error information. The programmable device with the configuration memory includes: an error detection section for detecting an error in the configuration memory, and outputting the detected error as well as an address in which the error occurred, as error information; and an error information holding section provided with a non-volatile memory to store the output error information.
Abstract translation: 本发明的目的在于提供一种具有配置存储器的可编程设备，即使在电源期间，由于配置存储器的地面辐射，可以保持难以产生的发生异常情况的状态，例如在可编程设备中发生的故障 以提高基于所保持的错误信息的设备测试中的再现性。 具有配置存储器的可编程设备包括：错误检测部分，用于检测配置存储器中的错误，并将检测到的错误以及发生错误的地址输出作为错误信息; 以及设置有用于存储输出错误信息的非易失性存储器的错误信息保持部。
-
公开(公告)号：US20240176755A1
公开(公告)日：2024-05-30
申请号：US18551650
申请日：2022-04-01
Applicant: HITACHI, LTD.
Inventor： Yutaka UEMATSU , Tadanobu TOBA , Kenichi SHIMBO , Takumi UEZONO
IPC: G06F13/36
CPC classification number: G06F13/36 , G06F2213/40
Abstract: It is possible to reduce the amount of data transmitted from the edge system to the outside while securing the amount of information necessary for purposes such as factor analysis and diagnosis and the cycle of information acquisition. A distributed system includes: an edge device that is a moving body or equipment capable of automatic operation; and a diagnostic data computer. The edge device has: at least one of a movement mechanism and an actuation mechanism for implementing automatic operation; a component corresponding to at least one of the movement mechanism and the actuation mechanism; and an intra-edge controller that controls the component. The intra-edge controller reads state data acquired in the component, aggregates the read state data into component-level state data, determines an allowable latency when transferring the component-level state data to a transfer destination according to a predetermined rule, and transfers the component-level state data to the diagnostic data computer based on the allowable latency.
-
Patent Agency Ranking
公开(公告)号：US20220366735A1
公开(公告)日：2022-11-17
申请号：US17709506
申请日：2022-03-31
Applicant: HITACHI, LTD.
Inventor： Tadanobu TOBA , Kenichi SHIMBO , Yutaka UEMATSU
Abstract: A distributed system includes an edge device which is at least one of an automatically operable moving body and equipment, and a diagnostic data computer, in which the edge device converts state data representing a state of a component that is a constituent of a replaceable unit forming the edge device, into diagnostic data according to a predetermined diagnostic algorithm, and the diagnostic data computer acquires the diagnostic data from a first edge device, and specifies a second edge device that employs a component related to the component employed in he first edge device.
-
公开(公告)号：US20220060399A1
公开(公告)日：2022-02-24
申请号：US17355251
申请日：2021-06-23
Applicant: HITACHI, LTD.
Inventor： Kenichi SHIMBO , Tadanobu TOBA , Masaharu KINOSHITA
Abstract: A data processing method is performed by an edge device acquiring collected data from a collection target and a first computer capable of communicating with the edge device. The method includes: a first calculation process of, by the edge device, storing the collected data in a secure region to which referring of internally stored information from outside is not allowable and calculating first data which has a data amount less than the collected data and is irreversible in the secure region based on the stored collected data; a first communication process of, by the edge device, transmitting the first data calculated through the first calculation process to the first computer; and a second calculation process of, by the first computer, calculating second data based on the first data transmitted from the edge device through the first communication process.
-
公开(公告)号：US20230286542A1
公开(公告)日：2023-09-14
申请号：US18111010
申请日：2023-02-17
Applicant: Hitachi, Ltd.
Inventor： Tadanobu TOBA , Kenichi SHIMBO , Yutaka UEMATSU , Takumi UEZONO
IPC: B60W60/00
CPC classification number: B60W60/0015 , B60W60/001 , B60W2556/45 , B60W2556/10 , B60W2556/40 , B60W2555/20
Abstract: Provided are an edge device and a distributed system capable of efficiently collecting situation data when an unexpected event occurs in an automatically operable moving body and contributing to improvement of safety. An edge device according to the invention includes: a sensor provided in an automatically operable moving body; a recognition unit configured to recognize an object based on input data from the sensor; a determination unit configured to determine a recognition result of the recognition unit; a drive mechanism control unit configured to control the drive mechanism based on a determination result of the determination unit; an unexpected event determination unit configured to determine whether an unexpected event has occurred based on information from the recognition unit and the determination unit; a data shaping unit in which when it is determined that an unexpected event has occurred, the data shaping unit shapes, as communication data, a recognition result of the recognition unit, or the input data used for recognition by the recognition unit, and a determination result of the determination unit, or a calculation history up to the determination result; and a communication unit configured to transmit, to the cloud server, the communication data shaped by the data shaping unit.
-
9.发明申请公开(公告)号：US20220405581A1
公开(公告)日：2022-12-22
申请号：US17642972
申请日：2020-12-07
Applicant: Hitachi, Ltd.
Inventor： Tadanobu TOBA , Takumi UEZONO , Kenichi SHIMBO , Hiroaki ITSUJI , Nozomi KASAHARA , Yutaka UEMATSU
Abstract: A neural network that can detect abnormality of itself while suppressing redundancy of a scale is realized. A neural network optimization system includes a definition data analysis unit configured to analyze learned neural network definition data, an internode dependence degree analysis unit configured to generate dependence degree information indicating an internode dependence degree in a learned neural network defined by the learned neural network definition data, based on an analysis result of the learned neural network definition data, and a sensitive node extraction unit configured to extract a sensitive node in the learned neural network based on the dependence degree information.
-
公开(公告)号：US20220261986A1
公开(公告)日：2022-08-18
申请号：US17535779
申请日：2021-11-26
Applicant: Hitachi, Ltd.
Inventor： Takumi UEZONO , Tadanobu TOBA , Kenichi SHIMBO , Hiroaki ITSUJI
Abstract: An increase in a diagnosis load can be reduced. A diagnosis pattern generating unit generates a diagnosis pattern including a plurality of data sets for diagnosing whether a processing result of calculation processing by a subset of a plurality of intermediate nodes included in a learned neural network is correct. An intermediate node calculation component identifying unit identifies a node-core relationship that is a correspondence relationship between the intermediate node and a calculation component that executes calculation processing by the intermediate node. A diagnosis pattern reducing unit reduces the number of the plurality of data sets based on the node-core relationship.
-
-
-
-
-
-
-
-
-
Search Results
19
records
(took 0.013 seconds)
