Method and device with autonomous driving plan offset adjustment
AssigneeSamsung Electronics Co., Ltd.
InventorsJaemuk Jeong, Jahoo Koo, Seho Shin
Life of this patent
Estimating the term…
Abstract
An electronic device and method for more accurate autonomous driving, where the electronic device includes a sensor disposed in a moving object and configured to generate sensing data, and a processor configured to determine a predicted position of the moving object and a region of interest (ROI) based on a driving plan of the moving object, select a target position offset based on an available sensing region of the sensor in the ROI and the predicted position, and update the driving plan of the moving object based on the target position offset.
Claims22
An electronic system of a vehicle, the electronic system comprising:
- a sensor configured to generate sensing data;
- one or more processors;
- and
- a memory comprising one or more non-transitory storage media that store instructions that, when executed by the one or more processors, configures the one or more processors to:
- while the vehicle is driving along a first trajectory generated by a driving plan, determine a predicted future position of the vehicle and a region of interest (ROI), corresponding to the predicted future position, respectively based on the driving plan of the vehicle;
- select a target position offset based on a determined available sensing region of the sensor in the ROI and the predicted future position, where the target position offset is selected to provide an increased, compared to the determined available sensing region, available sensing region of the sensor;
- and
- update, based on the target position offset, the driving plan of the vehicle to generate a second trajectory for the vehicle to drive along towards a position corresponding to the target position offset.
An electronic device comprising:
- a sensor disposed in a moving object and configured to generate sensing data;
- one or more processors;
- and
- a memory comprising one or more non-transitory storage media that store instructions that, when executed by the one or more processors, causes the one or more processors to:
- while the vehicle is driving along a first trajectory generated by a driving plan, determine a predicted future position of the moving object and a region of interest (ROI), corresponding to the predicted future position, based on the driving plan of the moving object;
- select a target position offset based on a determined available sensing region of the sensor in the ROI and the predicted future position, where the target position offset is selected to provide an increased, compared to the determined available sensing region, available sensing region of the sensor;
- and
- update, based on the target position offset, the driving plan of the moving object to generate a second trajectory for the moving object to move along towards a position corresponding the target position, the updating including a generation of an active sensing layer indicating the position corresponding to the target position offset, and a performance of the update of the driving plan based on the active sensing layer and other layers for driving.
A method implemented by one or more processors, the method comprising:
- while the vehicle is driving along a first trajectory generated by a driving plan, determining a predicted future position of a vehicle and a region of interest (ROI), corresponding to the predicted future position, respectively based on the driving plan of the vehicle;
- selecting a target position offset based on a determined available sensing region of a sensor in the ROI and the predicted future position, where the target position offset is selected to provide an increased, compared to the determined available sensing region, available sensing region of the sensor;
- and
- updating, based on the target position offset, the driving plan of the vehicle to generate a second trajectory for the vehicle to drive along towards a position corresponding to the target position offset.
A method implemented by one or more processors, the method comprising:
- while the vehicle is driving along a first trajectory generated by a driving plan, determining a predicted future position of a moving object and a region of interest (ROI), corresponding to the predicted future position, based on the driving plan of the moving object;
- selecting a target position offset based on a determined available sensing region of a sensor in the ROI and the predicted future position, where the target position offset is selected to provide an increased, compared to the determined available sensing region, available sensing region of the sensor;
- and
- updating, based on the target position offset, the driving plan of the moving object to generate a second trajectory for the moving object to move along towards a position corresponding to the target position offset,
- wherein the target position offset comprises any one of a first offset distance from the sensor in a front of the moving object, a second offset distance from the sensor in a rear of the moving object, a third offset distance from the sensor in a first side of the moving object, or a fourth offset distance from the sensor in a second side of the moving object.
Dependencies are inferred from the claim text.