US 12,435,989 B2
Semantic map and point cloud map construction method, apparatus and storage medium
Guangwei Liu, Beijing (CN); and Ji Zhao, Beijing (CN)
Assigned to Beijing Tusen Zhitu Technology Co., Ltd., Beijing (CN)
Filed by Beijing Tusen Zhitu Technology Co., Ltd., Beijing (CN)
Filed on Jul. 29, 2022, as Appl. No. 17/816,274.
Claims priority of application No. 202110873182.3 (CN), filed on Jul. 30, 2021.
Prior Publication US 2023/0083965 A1, Mar. 16, 2023
Int. Cl. G01C 21/00 (2006.01); G06T 7/11 (2017.01); G06T 7/73 (2017.01)
CPC G01C 21/3811 (2020.08) [G01C 21/3867 (2020.08); G06T 7/11 (2017.01); G06T 7/73 (2017.01); G06T 7/74 (2017.01); G06T 2207/10028 (2013.01); G06T 2207/30256 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A map construction method, comprising:
acquiring sensor data collected by a preset sensor;
establishing a pose constraint relationship of a movable object and a semantic constraint relationship of a semantic object according to the sensor data;
performing a joint optimization solution according to the pose constraint relationship and the semantic constraint relationship to determine a semantic result of the semantic object and a pose result of the movable object; and
constructing a semantic map and a point cloud map according to the semantic result and the pose result;
wherein the sensor data comprises an image collected by an image collection apparatus and a point cloud collected by a point cloud collection apparatus, and the establishing the semantic constraint relationship of the semantic object according to the sensor data comprises:
determining an observed value of a 3D profile corresponding to each semantic object according to the image and the point cloud; and
establishing the semantic constraint relationship based on the observed value of the 3D profile.