公开(公告)号：US11162788B2

公开(公告)日：2021-11-02

申请号：US16718172

申请日：2019-12-17

申请人： DeepMap Inc.

发明人： Chen Chen ,  Gregory Coombe

IPC分类号： G01C21/00 ,  G01C11/12 ,  G06T7/73 ,  G06T7/68 ,  G06K9/00 ,  G06T7/55 ,  G06T17/05 ,  G01C11/30 ,  G06T7/246 ,  G06K9/46 ,  G01C11/06 ,  G01C21/36 ,  G06T7/11 ,  G01C21/32 ,  G05D1/00 ,  G05D1/02 ,  G06T7/70 ,  G06T7/593 ,  G06K9/62 ,  B60W40/06 ,  G01S19/42 ,  G08G1/00 ,  G06T17/20 ,  G01S19/47 ,  G01S19/46 ,  G01S17/89

摘要： A high-definition map system receives sensor data from vehicles traveling along routes and combines the data to generate a high definition map for use in driving vehicles, for example, for guiding autonomous vehicles. A pose graph is built from the collected data, each pose representing location and orientation of a vehicle. The pose graph is optimized to minimize constraints between poses. Points associated with surface are assigned a confidence measure determined using a measure of hardness/softness of the surface. A machine-learning-based result filter detects bad alignment results and prevents them from being entered in the subsequent global pose optimization. The alignment framework is parallelizable for execution using a parallel/distributed architecture. Alignment hot spots are detected for further verification and improvement. The system supports incremental updates, thereby allowing refinements of subgraphs for incrementally improving the high-definition map for keeping it up to date.

公开(公告)号：US20200225032A1

公开(公告)日：2020-07-16

申请号：US16718172

申请日：2019-12-17

申请人： DeepMap Inc.

发明人： Chen Chen ,  Gregory Coombe

IPC分类号： G01C11/12 ,  G06T7/73 ,  G06T7/68 ,  G06K9/00 ,  G06T7/55 ,  G06T17/05 ,  G01C11/30 ,  G06T7/246 ,  G06K9/46 ,  G01C11/06 ,  G01C21/36 ,  G06T7/11 ,  G01C21/32 ,  G05D1/00 ,  G05D1/02 ,  G06T7/70 ,  G06T7/593 ,  G06K9/62 ,  B60W40/06 ,  G01S19/42 ,  G08G1/00 ,  G06T17/20 ,  G01C21/00

摘要： A high-definition map system receives sensor data from vehicles traveling along routes and combines the data to generate a high definition map for use in driving vehicles, for example, for guiding autonomous vehicles. A pose graph is built from the collected data, each pose representing location and orientation of a vehicle. The pose graph is optimized to minimize constraints between poses. Points associated with surface are assigned a confidence measure determined using a measure of hardness/softness of the surface. A machine-learning-based result filter detects bad alignment results and prevents them from being entered in the subsequent global pose optimization. The alignment framework is parallelizable for execution using a parallel/distributed architecture. Alignment hot spots are detected for further verification and improvement. The system supports incremental updates, thereby allowing refinements of subgraphs for incrementally improving the high-definition map for keeping it up to date