Vehicle with visual simultaneous localization and mapping system and dynamic outlier rejection function
By identifying and processing ill-conditioned feature points in the V-SLAM system through dynamic covariance scaling and condition number thresholding, the problem of reduced navigation accuracy is solved, and the performance of the navigation system in environments with impaired signal is improved.
CN122192272APending Publication Date: 2026-06-12GM GLOBAL TECHNOLOGY OPERATIONS LLC
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GM GLOBAL TECHNOLOGY OPERATIONS LLC
- Filing Date
- 2025-02-26
- Publication Date
- 2026-06-12
Smart Images
Figure CN122192272A_ABST
Abstract
A visual simultaneous localization and mapping (V-SLAM) system for a mobile host (e.g., a vehicle) includes a sensor suite and a controller. The sensor suite collects raw input data for estimating host parameters. The controller includes a processor and a computer storage medium ("memory") containing computer-readable instructions. Execution of the instructions by the processor causes the controller to receive the input data, use the input data to estimate a pose of the host and compute three-dimensional (3D) coordinates of feature map points, and use a predetermined linear algebraic property (e.g., a condition number) to identify outliers among the feature map points. The controller also dynamically filters out the outliers using dynamic covariance scaling to generate a filtered set of feature map points, and transmits the set to a navigation system of the mobile host.
Need to check novelty before this filing date? Find Prior Art