中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

56 records (took 0.029 seconds)

    NEURAL NETWORK VALIDATION SYSTEM
    32.
    发明申请
    NEURAL NETWORK VALIDATION SYSTEM 有权

    公开(公告)号：US20230139521A1

    公开(公告)日：2023-05-04

    申请号：US17517260

    申请日：2021-11-02

    Applicant: GM Global Technology Operations LLC

    Inventor： Wei Tong ,  Shige Wang ,  Ramesh Sethu ,  Jeffrey D. Scheu ,  Prashanth Radhakrishan ,  Upali P. Mudalige ,  Ryan Ahmed

    IPC: G05D1/02 ,  G06K9/62 ,  G06N3/04

    Abstract: A system comprises a computer including a processor and a memory. The memory includes instructions such that the processor is programmed to: receive, at a first neural network, unlabeled sensor data, wherein the first neural network generates output based on the unlabeled sensor data, receive, at a second neural network, the unlabeled sensor data, wherein the second neural network generates output based on the unlabeled sensor data during a validation mode, the second neural network different from the first neural network, compare the output generated by the first neural network with the output generated by the second neural network, and generate an alert when a difference between the output generated by the first neural network and the output generated by the second neural network is greater than a predetermined comparison threshold.

    Method of controlling a vehicle to adjust perception system energy usage
    33.
    发明授权
    Method of controlling a vehicle to adjust perception system energy usage 有权

    公开(公告)号：US10915159B2

    公开(公告)日：2021-02-09

    申请号：US15944003

    申请日：2018-04-03

    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC

    Inventor： Yiran Hu ,  Steven E. Muldoon ,  Wei Tong ,  Shige Wang

    IPC: G01M17/00 ,  G06F7/00 ,  G06F11/30 ,  G07C5/00 ,  G06F1/3287 ,  G06F17/00

    Abstract: A method of controlling a vehicle includes determining a current operating situation of the vehicle, and identifying a subset of a plurality of sensors of the vehicle needed to provide data to enable a vehicle control function for the current operating situation of the vehicle. A remainder of the plurality of sensors is disengaged to reduce electric energy usage by the vehicle while the vehicle is operating in the current operating situation of the vehicle. A sampling rate for the selected subset of sensors may be reduced to further reduce energy usage of the vehicle. Additionally, an energy reduction processing strategy may be implemented to reduce a processor frequency or a voltage of a computing device used to provide the vehicle control function to further reduce energy usage of the vehicle.

    Fixed-point quantization in neural networks for vehicle perception systems
    34.
    发明授权
    Fixed-point quantization in neural networks for vehicle perception systems 有权

    公开(公告)号：US10909390B2

    公开(公告)日：2021-02-02

    申请号：US16170478

    申请日：2018-10-25

    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC

    Inventor： Shuqing Zeng ,  Wei Tong ,  Shige Wang ,  Roman L. Millett

    IPC: G06N3/08 ,  G06K9/00 ,  H03M7/24

    Abstract: Examples of techniques for using fixed-point quantization in deep neural networks are disclosed. In one example implementation according to aspects of the present disclosure, a computer-implemented method includes capturing a plurality of images at a camera associated with a vehicle and storing image data associated with the plurality of images to a memory. The method further includes dispatching vehicle perception tasks to a plurality of processing elements of an accelerator in communication with the memory. The method further includes performing, by at least one of the plurality of processing elements, the vehicle perception tasks for the vehicle perception using a neural network, wherein performing the vehicle perception tasks comprises quantizing a fixed-point value based on an activation input and a synapse weight. The method further includes controlling the vehicle based at least in part on a result of performing the vehicle perception tasks.

    High Precision Low Bit Convolutional Neural Network
    36.
    发明申请
    High Precision Low Bit Convolutional Neural Network 审中-公开

    公开(公告)号：US20200065661A1

    公开(公告)日：2020-02-27

    申请号：US16107315

    申请日：2018-08-21

    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC

    Inventor： Wei Tong ,  Shuqing Zeng ,  Upali P. Mudalige ,  Shige Wang

    IPC: G06N3/08 ,  G06N3/04 ,  G06K9/00 ,  G06K9/38

    Abstract: Described herein are systems, methods, and computer-readable media for generating and training a high precision low bit convolutional neural network (CNN). A filter of each convolutional layer of the CNN is approximated using one or more binary filters and a real-valued activation function is approximated using a linear combination of binary activations. More specifically, a non-1×1 filter (e.g., a k×k filter, where k>1) is approximated using a scaled binary filter and a 1×1 filter is approximated using a linear combination of binary filters. Thus, a different strategy is employed for approximating different weights (e.g., 1×1 filter vs. a non-1×1 filter). In this manner, convolutions performed in convolutional layer(s) of the high precision low bit CNN become binary convolutions that yield a lower computational cost while still maintaining a high performance (e.g., a high accuracy).

    Autonomous vehicle adaptive parallel image processing system
    37.
    发明授权
    Autonomous vehicle adaptive parallel image processing system 有权

    公开(公告)号：US10572748B2

    公开(公告)日：2020-02-25

    申请号：US15833382

    申请日：2017-12-06

    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC

    Inventor： Shige Wang ,  Wei Tong ,  Shuqing Zeng ,  Roman Millett

    IPC: G06K9/00 ,  G06F9/48 ,  G05D1/02

    Abstract: An adaptive parallel imaging processing system in a vehicle is provided. The system may include, but is not limited to, a plurality of processors and a resource management system including, but not limited to, an execution monitor, the execution monitor configured to calculate an average utilization of each of the plurality of processors over a moving window, and a service scheduler controlling a request queue for each of the plurality of processors, the service scheduler scheduling image processing tasks in the respective request queue for the each of the plurality of processors based upon the average utilization of each of the plurality of processors, the capabilities of each of the plurality of processors, and a priority associated with each image processing task, wherein an autonomous vehicle control system is configured to generate the instructions to control the at least one vehicle system based upon the processed image processing tasks.

    Architecture and apparatus for advanced arbitration in embedded controls
    38.
    发明授权
    Architecture and apparatus for advanced arbitration in embedded controls 有权

    公开(公告)号：US09996431B2

    公开(公告)日：2018-06-12

    申请号：US15078255

    申请日：2016-03-23

    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC

    Inventor： Shige Wang ,  Xiaofeng F. Song ,  Xian Zhang

    IPC: G06F11/10 ,  G06F11/18 ,  G05B19/042

    CPC classification number: G06F11/182 ,  G05B19/0421 ,  G06F2201/805 ,  G06F2201/82

    Abstract: A method of arbitrating conflicting outputs in a redundant control system. Execution data of a task executed by each controller in the redundant control system is recorded. The execution data includes an initial timestamp of each execution stream, identification of critical functions in each execution stream, and parameter values used by the critical functions. A path executed by each controller is identified based only on the critical functions executed for each execution stream. The recorded execution data of each executed path is applied to an arbitration module. An output result from one of the respective controllers selecting, by an arbitration module, based on the recorded execution data of each executed path. The output result of the selected controller is communicated to a next module for further processing.

Patent Agency Ranking