Method, system and equipment for acquiring drivable area based on multi-frame superposition and medium

The invention provides a method, a system and equipment for acquiring a drivable area based on multi-frame superposition, and a medium. The method comprises the following steps: processing a collected 3D point cloud of a current frame to obtain a segmentation result of the 3D point cloud of the curr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ZHAO YALI, WU LIANSONG, QIN HUAN, LIANG SHUANG, LI HAOPENG, QIU ZHONGYUAN
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention provides a method, a system and equipment for acquiring a drivable area based on multi-frame superposition, and a medium. The method comprises the following steps: processing a collected 3D point cloud of a current frame to obtain a segmentation result of the 3D point cloud of the current frame; filling the current frame grid map with the current frame 3D point cloud segmentation result, and outputting current frame confidence information; according to the grid position transfer information obtained by carrying out iterative matching on the previous frame of grid map and the current frame of grid map according to the pose matrix, transferring the previous frame of confidence information into the current frame of grid map to obtain a multi-frame superposed grid map; according to the initialized state transition matrix, the confidence information of the current frame and the confidence information of the previous frame, obtaining prior confidence information and posterior confidence information of