IP Library Granted Patent US 12,736,625
Granted Patent B2
US 12,736,625 · App. 18/421,556 · Granted Sep 15, 2026

Radar apparatus and point cloud generation method

Inventors: Hidekuni Yomo (Kanagawa, JP); Tomohiro Yui (Kanagawa, JP); Takaaki Kishigami (Tokyo, JP); Noriaki Saito (Tokyo, JP)
Assignee: Panasonic Automotive Systems Co., Ltd.
G01S7/415G01S13/536G01S17/931
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,736,625
App. No.
18/421,556
Granted
Sep 15, 2026
Kind
B2
Abstract

A radar apparatus includes signal processing circuitry, which, in operation, generates point cloud data for each frame based on a reflection wave signal resulting from a radar signal reflected by an object; extracting circuitry, which, in operation, extracts a first point cloud corresponding to a stationary object from the point cloud data; and superimposing circuitry, which, in operation, superimposes the point cloud data of a current frame and the first point cloud of a past frame.

Claims (63)

1 . A radar apparatus configured to be mounted on a vehicle, the radar apparatus comprising:

signal processing circuitry, which, in operation, generates point cloud data for each frame of a plurality of frames based on a reflection wave signal resulting from a radar signal reflected by a stationary object and a moving object;

extracting circuitry, which, in operation, extracts a first point cloud corresponding to the stationary object from the point cloud data and a second point cloud corresponding to the moving object from the point cloud data;

superimposing circuitry, which, in operation, superimposes the point cloud data by superimposing the first point cloud and the second point cloud of a current frame and the first point cloud of a past frame, and outputs the superimposed point cloud data to a vehicle controlling circuitry that controls the vehicle based on the superimposed point cloud data, and

removing circuitry, which, in operation,

detects a position of the second point cloud of the current frame is a position of the first point cloud of the past frame, and

removes the first point cloud of the past frame from the superimposed point cloud data, in response to the position of the second point cloud of the current frame being detected as the position of the first point cloud of the past frame.

2 . The radar apparatus according to claim 1 , wherein,

the extracting circuitry extracts a plurality of first point clouds from a plurality of past frames, respectively, and

the radar apparatus further comprises second superimposing circuitry, which, in operation, superimposes the plurality of first point clouds respectively extracted from the plurality of past frames.

3 . The radar apparatus according to claim 2 , further comprising:

deriving circuitry, which, in operation, derives a moving vector of the vehicle; and

shifting circuitry, which, in operation, shifts, based on the moving vector, positions of the plurality of first point clouds respectively extracted from the plurality of past frames, wherein,

the second superimposing circuitry superimposes the plurality of first point clouds with the positions of the plurality of first point clouds shifted.

4 . The radar apparatus according to claim 1 , further comprising:

storing circuitry, which, in operation, stores the first point cloud of the current frame in case a position of the first point cloud of the current frame coincides with an area where the first point cloud of the past frame is not stored among a plurality of areas resulting from dividing a detection target area of the radar apparatus.

5 . The radar apparatus according to claim 1 , further comprising:

clustering circuitry, which, in operation, generates a cluster for the stationary object and the moving object based on the point cloud data outputted from the superimposing circuitry; and

tracking circuitry, which, in operation, performs tracking of the cluster, wherein,

the extracting circuitry determines whether each point cloud in the point cloud data is the first point cloud or the second point cloud based on a result of the tracking.

6 . The radar apparatus according to claim 1 , further comprising:

storing circuitry, which, in operation, stores a plurality of first point clouds over N frames and stores a moving vector of the vehicle over the N frames, N being an integer greater than or equal to 2; and

second superimposing circuitry, which, in operation, superimposes the plurality of first point clouds respectively extracted from the N frames after positions of the plurality of first point clouds are shifted based on the moving vector.

7 . The radar apparatus according to claim 6 , wherein the storing circuitry stops to store the first point cloud and the moving vector over the N frames in a case where a predetermined time elapses after the vehicle changes from a traveling state to a stopped state, or in a case where the vehicle repeats the traveling state and the stopped state a predetermined number of times.

8 . The radar apparatus according to claim 1 , further comprising:

removing circuitry, which, in operation, removes the first point cloud when a position of the vehicle before a power source of the vehicle is stopped is inconsistent with a position of the vehicle after the power source is restarted.

9 . The radar apparatus according to claim 1 , further comprising:

removing circuitry, which, in operation, removes the first point cloud when a position of the first point cloud before a power source of the vehicle is stopped is inconsistent with a position of the first point cloud after the power source is restarted.

10 . The radar apparatus according to claim 1 , further comprising:

deriving circuitry, which, in operation, derives a moving vector of the vehicle; and

compensating circuitry, which, in operation, compensates for the moving vector based on a result of comparing the point cloud data of the current frame and the point cloud data of a previous frame.

11 . A point cloud generation method using a radar apparatus configured to be mounted on a vehicle, comprising:

generating, by the radar apparatus, point cloud data for each frame of a plurality of frames based on a reflection wave signal resulting from a radar signal reflected by a stationary object and a moving object;

extracting, by the radar apparatus, a first point cloud corresponding to the stationary object from the point cloud data and a second point cloud corresponding to the moving object from the point cloud data;

superimposing, by the radar apparatus, the point cloud data by superimposing the first point cloud and the second point cloud of a current frame and the first point cloud of a past frame;

detecting, by the radar apparatus, a position of the second point cloud of the current frame is a position of the first point cloud of the past frame;

removing, by the radar apparatus, the first point cloud of the past frame from the superimposed point cloud data, in response to detecting the position of the second point cloud of the current frame is the position of the first point cloud of the past frame; and

outputting, by the radar apparatus, the superimposed point cloud data to a vehicle controlling circuitry that controls the vehicle based on the superimposed point cloud data.

12 . The point cloud generation method according to claim 11 , further comprising:

extracting, by the radar apparatus, a plurality of first point clouds from a plurality of past frames, respectively; and

superimposing, by the radar apparatus, the plurality of first point clouds respectively extracted from the plurality of past frames.

13 . The point cloud generation method according to claim 12 , further comprising:

deriving, by the radar apparatus, a moving vector of the vehicle;

shifting, by the radar apparatus, based on the moving vector, positions of the plurality of first point clouds respectively extracted from the plurality of past frames; and

superimposing, by the radar apparatus, the point cloud data of the current frame and the plurality of first point clouds with the positions of the plurality of first point clouds shifted.

14 . The point cloud generation method according to claim 11 , further comprising:

storing, by the radar apparatus, the first point cloud of the current frame in case a position of the first point cloud of the current frame coincides with an area where the first point cloud of the past frame is not stored among a plurality of areas resulting from dividing a detection target area of the radar apparatus.

15 . The point cloud generation method according to claim 11 , further comprising:

generating, by the radar apparatus, a cluster for the stationary object and the moving object based on the point cloud data superimposed on the first point cloud of the past frame;

performing, by the radar apparatus, tracking of the cluster; and

determining, by the radar apparatus, whether each point cloud in the point cloud data is the first point cloud or the second point cloud based on a result of the tracking.

16 . The point cloud generation method according to claim 11 , further comprising:

storing, by the radar apparatus, a plurality of first point clouds over N frames and storing a moving vector of the vehicle over the N frames, N being an integer greater than or equal to 2; and

superimposing, by the radar apparatus, the plurality of first point clouds respectively extracted from the N frames after positions of the plurality of first point clouds are shifted based on the moving vector.

17 . The point cloud generation method according to claim 16 , further comprising:

stopping storing, by the radar apparatus, the first point cloud and the moving vector over the N frames in a case where a predetermined time elapses after the vehicle changes from a traveling state to a stopped state, or in a case where the vehicle repeats the traveling state and the stopped state a predetermined number of times.

18 . The point cloud generation method according to claim 11 , further comprising:

removing, by the radar apparatus, the first point cloud when a position of the vehicle before a power source of the vehicle is stopped is inconsistent with a position of the vehicle after the power source is restarted.

19 . The point cloud generation method according to claim 11 , further comprising:

removing, by the radar apparatus, the first point cloud when a position of the first point cloud before a power source of the vehicle is stopped is inconsistent with a position of the first point cloud after the power source is restarted.

20 . The point cloud generation method according to claim 11 , further comprising:

deriving, by the radar apparatus, a moving vector of the vehicle; and

compensating, by the radar apparatus, for the moving vector based on a result of comparing the point cloud data of the current frame and the point cloud data of a previous frame.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 6, 2024
From: PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO., LTD.
To: PANASONIC AUTOMOTIVE SYSTEMS CO., LTD.
Reel/Frame 068328/0764 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 12, 2024
From: YOMO, HIDEKUNI; YUI, TOMOHIRO; KISHIGAMI, TAKAAKI; SAITO, NORIAKI
To: PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO., LTD.
Reel/Frame 067970/0673 →
Priority Claims (1)
JP 2023-010158 · Jan 26, 2023 · national
Continuity (1)
Related Publication 20240288547A1 · Aug 29, 2024
References Cited (11)
US 20070030131A1 · Takahama · 2007 [cited by examiner]
US 20180267558A1 · Tiwari · 2018 [cited by examiner]
US 20220135074A1 · Armstrong-Crews · 2022 [cited by examiner]
US 20220383749A1 · Ishikawa · 2022 [cited by examiner]
US 20230213640A1 · Kuroda · 2023 [cited by examiner]
US 20240151813A1 · Fischer · 2024 [cited by examiner]
JP 2011191227A · 2011 [cited by applicant]
JP 5700940B2 · 2015 [cited by applicant]
JP 2017166846A · 2017 [cited by applicant]
JP 2022511990A · 2022 [cited by applicant]
Jin et al., “Comparison of Different Approaches for Identification of Radar Ghost Detections in Automotive Scenarios,” IEEE Radar Conference, May 2021. (7 pages). [cited by applicant]