SENSOR DATA PROCESSING METHOD AND APPARATUS
    1.
    发明申请

    公开(公告)号:US20200226434A1

    公开(公告)日:2020-07-16

    申请号:US16825145

    申请日:2020-03-20

    Inventor: Wanlong LI Feng WEN

    Abstract: The method includes: determining a reference sensor from at least two sensors and sampling moments of the reference sensor; when obtaining observation data of a sensor K at a moment t1, determining a type of the sensor K; if the sensor K is a first sensor of a first type, determining a sampling moment T1 that is closest to t1 and an interpolation coefficient λ1, and calculating first rotation data and first translation data that are at T1, and a first covariance matrix; if the sensor K is a second sensor of a second type, determining two sampling moments T2 and T3 that are closest to t1 and interpolation coefficients λ2 and λ3, and calculating second rotation data and second translation data that are of the second sensor from T2 to T3, and a second covariance matrix; merging rotation data, translation data, and covariance matrices that are at the sampling moments.

    VISUAL SLAM METHOD AND APPARATUS BASED ON POINT AND LINE FEATURES

    公开(公告)号:US20200218929A1

    公开(公告)日:2020-07-09

    申请号:US16824219

    申请日:2020-03-19

    Abstract: The method is: receiving a visual image frame input by a camera; extracting feature points and feature lines of the visual image frame; predicting a first pose of the camera device by using the feature points; observing an extracted first feature line, to determine a feature line measurement of the first feature line; obtaining a global feature line state vector set of the camera device, where the global feature line state vector set includes feature line state vectors of N historical feature lines, and N is a positive integer; and updating the first pose by using the feature line measurement and the global feature line state vector set. In this way, feature point-based motion estimations and observation features of feature lines observed in an environment are integrated, to update a pose of the camera device in real time, thereby improving visual SLAM accuracy.

Patent Agency Ranking