Invention Grant
- Patent Title: Detecting distortion using known shapes
-
Application No.: US16196632Application Date: 2018-11-20
-
Publication No.: US10677900B2Publication Date: 2020-06-09
- Inventor: Austin K. Russell , Eric C. Danziger
- Applicant: LUMINAR TECHNOLOGIES, INC.
- Applicant Address: US FL Orlando
- Assignee: Luminar Technologies, Inc.
- Current Assignee: Luminar Technologies, Inc.
- Current Assignee Address: US FL Orlando
- Main IPC: G06T5/00
- IPC: G06T5/00 ; G01S7/497 ; G01S17/42 ; G05D1/02 ; G01S17/89 ; G01S17/87 ; G01S7/481 ; G06N20/00 ; G01S17/06 ; G01S17/58 ; G01S17/86 ; G01S17/931 ; G06T7/246 ; G06T7/73 ; B60W30/095 ; G06K9/00 ; G06T7/521 ; G01S7/48 ; G01S17/08 ; H04N5/232 ; G06K9/62 ; G06T3/60 ; G06T7/20 ; G06T7/00 ; G05D1/00

Abstract:
A computer-implemented method of detecting object distortion. The method includes receiving sensor data generated by one or more sensors of the vehicle. The one or more sensors are configured to sense an environment through which the vehicle is moving by following a scan pattern. The method also includes obtaining, based on the sensor data, a point cloud frame representative of the environment and identifying a point cloud object within the point cloud frame. Additionally, the method includes analyzing the point cloud object to identify a feature of the point cloud object that has an expected shape and comparing the feature of the point cloud object to the expected shape. The method also includes identifying that the point cloud object is distorted based on the feature of the point cloud object not matching the expected shape.
Public/Granted literature
- US20200043146A1 DETECTING DISTORTION USING KNOWN SHAPES Public/Granted day:2020-02-06
Information query