Localization based on predefined features of the environment
AssigneeMotional AD LLC
InventorsDana Miller, James Esper, Theodore Alden, Asim Ayaz, Michael Oden
Life of this patent
Estimating the term…
Abstract
The subject matter described in this specification is directed to a computer system and techniques for determining a location of an autonomous vehicle. The computer system is configured to determine the location using localization data from multiple data sources. When the localization data from the multiple data sources is unavailable or inaccurate, the computer system is configured to determine the location using predefined features of the environment.
Claims18
A system comprising:
- one or more computer processors;
- and one or more non-transitory media storing instructions which, when executed by the one or more computer processors, cause performance of operations comprising:
- receiving first localization data of a vehicle and second localization data of the vehicle at a first time, wherein the first localization data is from a first source and the second localization data is from a second source different from the first source;
- determining a first difference between the first localization data and the second localization data received at the first time;
- based on determining that the first difference between the first localization data and the second localization data satisfies an accuracy criterion, providing a first estimated location of the vehicle based on the first localization data and the second localization data;
- receiving first localization data of the vehicle and second localization data of the vehicle at a second time;
- determining a second difference between the first localization data and the second localization data received at the second time;
- determining that the second difference between the first localization data and the second localization data at the second time does not satisfy the accuracy criterion;
- and
- based on determining that the second difference between the first localization data and the second localization data at the second time does not satisfy the accuracy criterion:
- detecting, using one or more images of an environment surrounding the vehicle, one or more predefined features of the environment;
- providing a second estimated location of the vehicle based on the one or more predefined features of the environment;
- and
- causing the vehicle to navigate based on the second estimated location of the vehicle.
A method, comprising:
- receiving first localization data of a vehicle and second localization data of the vehicle at a first time, wherein the first localization data is from a first source and the second localization data is from a second source different from the first source;
- determining a first difference between the first localization data and the second localization data received at the first time;
- based on determining that the first difference between the first localization data and the second localization data satisfies an accuracy criterion, providing a first estimated location of the vehicle based on the first localization data and the second localization data;
- receiving first localization data of the vehicle and second localization data of the vehicle at a second time;
- determining a second difference between the first localization data and the second localization data received at the second time;
- determining that the second difference between the first localization data and the second localization data at the second time does not satisfy the accuracy criterion;
- and
- based on determining that the second difference between the first localization data and the second localization data at the second time does not satisfy the accuracy criterion:
- detecting, using one or more images of an environment surrounding the vehicle, one or more predefined features of the environment;
- providing a second estimated location of the vehicle based on the one or more predefined features of the environment;
- and
- causing the vehicle to navigate based on the second estimated location of the vehicle.
One or more non-transitory media storing instructions which, when executed by one or more computing devices, cause performance of operations comprising:
- receiving first localization data of a vehicle and second localization data of the vehicle at a first time, wherein the first localization data is from a first source and the second localization data is from a second source different from the first source;
- determining a first difference between the first localization data and the second localization data received at the first time;
- based on determining that the first difference between the first localization data and the second localization data satisfies an accuracy criterion, providing a first estimated location of the vehicle based on the first localization data and the second localization data;
- receiving first localization data of the vehicle and second localization data of the vehicle at a second time;
- determining a second difference between the first localization data and the second localization data received at the second time;
- determining that the second difference between the first localization data and the second localization data at the second time does not satisfy the accuracy criterion;
- and
- based on determining that the second difference between the first localization data and the second localization data at the second time does not satisfy the accuracy criterion:
- detecting, using one or more images of an environment surrounding the vehicle, one or more predefined features of the environment;
- providing a second estimated location of the vehicle based on the one or more predefined features of the environment;
- and
- causing the vehicle to navigate based on the second estimated location of the vehicle.
Dependencies are inferred from the claim text.