公开(公告)号：EP4320606A1

公开(公告)日：2024-02-14

申请号：EP22719496.6

申请日：2022-04-09

申请人： QUALCOMM Incorporated

发明人： BELLI, Davide ,  MAJOR, Bence ,  DIJKMAN, Daniel Hendricus Franciscus ,  PORIKLI, Fatih Murat

IPC分类号： G06V40/40 ,  G06V10/82 ,  G06V40/12 ,  G06V10/80

公开(公告)号：EP3887857A2

公开(公告)日：2021-10-06

申请号：EP19824145.7

申请日：2019-11-29

申请人： QUALCOMM INCORPORATED

发明人： GOWAIKAR, Radhika Dilip ,  SUKHAVASI, Ravi Teja ,  FONTIJNE, Daniel Hendricus Franciscus ,  MAJOR, Bence ,  ANSARI, Amin ,  LIM, Teck Yian ,  SUBRAMANIAN, Sundar ,  WU, Xinzhou

IPC分类号： G01S13/931 ,  G01S13/86 ,  G01S13/89 ,  G06N3/02 ,  G01S7/295 ,  G06K9/00 ,  G06N3/04 ,  G01S7/41

公开(公告)号：EP3887849A2

公开(公告)日：2021-10-06

申请号：EP19824143.2

申请日：2019-11-29

申请人： QUALCOMM INCORPORATED

发明人： FONTIJNE, Daniel Hendricus Franciscus ,  ANSARI, Amin ,  MAJOR, Bence ,  SUKHAVASI, Ravi Teja ,  GOWAIKAR, Radhika Dilip ,  WU, Xinzhou ,  SUBRAMANIAN, Sundar ,  HAMILTON, Michael John

IPC分类号： G01S7/41 ,  G01S13/86 ,  G01S13/89 ,  G01S13/931 ,  G01S17/89 ,  G01S17/931 ,  G06N3/02 ,  G06K9/00 ,  G06N3/04

公开(公告)号：EP4379580A2

公开(公告)日：2024-06-05

申请号：EP24156992.0

申请日：2019-11-29

申请人： QUALCOMM Incorporated

发明人： FONTIJNE, Daniel Hendricus Franciscus ,  ANSARI, Amin ,  MAJOR, Bence ,  SUKHAVASI, Ravi Teja ,  GOWAIKAR, Radhika Dilip ,  WU, Xinzhou ,  SUBRAMANIAN, Sundar ,  HAMILTON, Michael John

IPC分类号： G06F18/25

CPC分类号： G01S13/931 ,  G01S17/89 ,  G01S13/865 ,  G01S17/931 ,  G01S13/89 ,  G01S13/867 ,  G01S7/417 ,  G01S7/2955 ,  G06N3/08 ,  G06V20/10 ,  G06V20/58 ,  G06V20/70 ,  G06V10/82 ,  G06V10/803 ,  G06V10/764 ,  G06N3/044 ,  G06N3/045 ,  G06F18/251

摘要： Disclosed are techniques for employing deep learning to analyze radar signals. In an aspect, an on-board computer of a host vehicle receives, from a radar sensor of the vehicle, a plurality of radar frames, executes a neural network on a subset of the plurality of radar frames, and detects one or more objects in the subset of the plurality of radar frames based on execution of the neural network on the subset of the plurality of radar frames. Further, techniques for transforming polar coordinates to Cartesian coordinates in a neural network are disclosed. In an aspect, a neural network receives a plurality of radar frames in polar coordinate space, a polar-to-Cartesian transformation layer of the neural network transforms the plurality of radar frames to Cartesian coordinate space, and the neural network outputs the plurality of radar frames in the Cartesian coordinate space.