利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

11 条记录 (耗时 0.019 秒)

    Track segment cleaning of tracked objects
    1.
    发明授权
    Track segment cleaning of tracked objects 有权

    公开(公告)号：US11625909B1

    公开(公告)日：2023-04-11

    申请号：US17736735

    申请日：2022-05-04

    申请人： Motional AD LLC

    发明人： Lubing Zhou ,  Jiong Yang ,  Yilaun Guo

    IPC分类号： G06V10/80 ,  G06V20/58 ,  G06T7/246 ,  G06V10/98 ,  G06V10/26 ,  G06V10/774

    摘要： Provided are methods for track segment cleaning of tracked objects using neural networks, which can include detecting a first track segment and a second track segment. The method includes applying a machine learning model trained to determine if the first track segment and second track segment capture real objects and if the first track segment and the second track segment are representative of an identical object exterior to a vehicle. The method further includes combining the first track segment and the second track segment to form a single track segment having a single trajectory in response to the first track segment and the second track segment being determined to be representative of the identical object. Systems and computer program products are also provided.

    LIDAR POINT CLOUD SEGMENTATION USING BOX PREDICTION
    3.
    发明申请
    LIDAR POINT CLOUD SEGMENTATION USING BOX PREDICTION 有权

    公开(公告)号：US20220357453A1

    公开(公告)日：2022-11-10

    申请号：US17740604

    申请日：2022-05-10

    申请人： Motional AD LLC

    发明人： Lubing Zhou ,  Xiaoli Meng ,  Karan Rajendra Shetti

    IPC分类号： G01S17/89 ,  G01S17/931

    摘要： In an example method, a perception system receives first data representing a point cloud having a plurality of points, and clusters the points into a plurality of clusters. The clusters include a first cluster representing a first portion of an object and a second cluster representing a second portion of the object. Further, the perception system generates a first bounding box enclosing at least the first cluster and the second cluster, and generates a second bounding box enclosing at least the first cluster and the second cluster. The perception system selects either the first bounding box or the second bounding box, and outputs second data representing the object. The second data includes an indication of the selected bounding box and an indication of the object.

    Deep Learning Based Beam Control for Autonomous Vehicles
    4.
    发明公开
    Deep Learning Based Beam Control for Autonomous Vehicles 审中-公开

    公开(公告)号：US20230150418A1

    公开(公告)日：2023-05-18

    申请号：US17528701

    申请日：2021-11-17

    申请人： Motional AD LLC

    发明人： Lubing Zhou ,  Xiaoli Meng ,  Karan Rajendra Shetti

    IPC分类号： B60Q1/08 ,  B60Q1/14 ,  G06K9/00

    CPC分类号： B60Q1/085 ,  B60Q1/143 ,  G06K9/00825

    摘要： Provided are systems and methods for a deep learning based beam control. Sensor data associated with the environment and the corresponding detected objects from a perception system are obtained. Object features and image features are extracted. The extracted object features and image features are fused into fused features. A beam control status is predicted according to the fused features, wherein the beam control status indicates a high beam illumination intensity or a low beam illumination intensity of a light emitting device.

    TRACK REFINEMENT NETWORKS
    10.
    发明公开
    TRACK REFINEMENT NETWORKS 审中-公开

    公开(公告)号：US20240126268A1

    公开(公告)日：2024-04-18

    申请号：US18073104

    申请日：2022-12-01

    申请人： Motional AD LLC

    发明人： Jiong Yang ,  Lubing Zhou

    IPC分类号： G05D1/02 ,  G06V10/32 ,  G06V10/774

    CPC分类号： G05D1/0248 ,  G05D1/0212 ,  G06V10/32 ,  G06V10/774 ,  G06V2201/07

    摘要： Provided are methods for a track refinement network. In examples, center boxes are obtained from a record of driving data, wherein a center box is a center of a sequence of boxes along a track, and the track is associated with a tracked object detected within the sequence of boxes, each respective box comprising a center, a size, and an orientation. Track windows are generated around respective center boxes, wherein a track window corresponds to a respective center box along the track. Track windows are cropped and normalized with respect to center boxes to enable single refinement model for multiple object classes. Point cloud features and trajectory features are extracted from the cropped and normalized track windows. The point cloud features and trajectory features are input into a track refinement network, wherein the track refinement network uses features from the entire track to output a refined center, a refined size, and a refined orientation of each respective center box.