-
1.
公开(公告)号:US20250078315A1
公开(公告)日:2025-03-06
申请号:US18240802
申请日:2023-08-31
Applicant: NVIDIA Corporation
Inventor: Dae Jin Kim , Rajath Shetty
Abstract: In various examples, interior sensor calibration using exterior features for autonomous systems and applications is described herein. Systems and methods are disclosed that use one or more features that are located exterior to a vehicle, such as one or more tags located in the environment surrounding the vehicle, to determine one or more values for one or more calibration parameters that calibrate an interior sensor of the vehicle with respect to a reference coordinate system of the vehicle. For instance, such as when the vehicle is located at a calibration station, the sensor may generate sensor data representing a sensor representation, where at least a portion of the sensor representation is associated with a feature that is visible through a transparent component (e.g., a window) of the vehicle. The sensor data may then be used to calibrate the sensor with respect to the coordinate system.
-
公开(公告)号:US20250067581A1
公开(公告)日:2025-02-27
申请号:US18456070
申请日:2023-08-25
Applicant: NVIDIA Corporation
Inventor: Dae Jin Kim , Rajath Shetty
Abstract: In various examples, interior sensor calibration for autonomous systems and applications is described herein. Systems and methods are disclosed that may recalibrate sensors of a vehicle, such as sensors located within the interior of the vehicle, using one or more techniques. For instance, if a sensor is attached to a component within the interior of the vehicle, an additional sensor associated with the component may output data indicating the location and/or orientation of the component within the vehicle. The indicated location and/or orientation of the component may then be used to recalibrate the sensor with respect to a reference coordinate system of the vehicle. For a second example, the sensor may output data representing at least a feature located within the interior of the vehicle. The sensor may then again be recalibrated based at least on a portion of the data that represents the feature.
-