Dynamic routing for self-driving vehicles
AssigneeUatc, LLC
InventorsAnthony Stentz, Bryan Nagy
Life of this patent
Estimating the term…
Abstract
A system can analyze a sensor view of a surrounding area of a self-driving vehicle (SDV). Based on analyzing the sensor view, the system can determine that a decision point along a current route of the SDV exceeds a predetermined risk threshold, and diverge the SDV from the current route based at least in part on the decision point exceeding the predetermined risk threshold.
Claims20
A system comprising:
- one or more processors;
- and
- a memory storing an instruction set that, when executed by the one or more processors causes the system to:
- analyze a sensor view of a surrounding area of an autonomous vehicle by comparing the sensor view with a localization map comprising data defining the surrounding area, wherein the localization map identifies multiple objects of interest along a road segment that the autonomous vehicle traverses;
- based on comparing the sensor view with the localization map, identify a decision point along a current route of the autonomous vehicle, the decision point indicating a location at which a plurality of route options exist due to a possible hazard or obstacle at the location;
- determine that the decision point along the current route of the autonomous vehicle satisfies a predetermined risk threshold;
- and
- diverge the autonomous vehicle from the current route based at least in part on the decision point satisfying the predetermined risk threshold.
A non-transitory computer-readable medium storing an instruction set that, when executed by one or more processors, causes the one or more processors to:
- analyze a sensor view of a surrounding area of an autonomous vehicle by comparing the sensor view with a localization map comprising data defining the surrounding area, wherein the localization map identifies multiple objects of interest along a road segment that the autonomous vehicle traverses;
- based on comparing the sensor view with the localization map, identify a decision point along a current route of the autonomous vehicle, the decision point indicating a location at which a plurality of route options exist due to a possible hazard or obstacle at the location;
- determine that the decision point along the current route of the autonomous vehicle satisfies a predetermined risk threshold;
- and
- diverge the autonomous vehicle from the current route based at least in part on the decision point satisfying the predetermined risk threshold.
A computer-implemented method of operating an autonomous vehicle, the method being performed by one or more processors and comprising:
- analyzing a sensor view of a surrounding area of the autonomous vehicle by comparing the sensor view with a localization map comprising data defining the surrounding area, wherein the localization map identifies multiple objects of interest along a road segment that the autonomous vehicle traverses;
- based on comparing the sensor view with the localization map, identifying a decision point along a current route of the autonomous vehicle, the decision point indicating a location at which a plurality of route options exists due to a possible hazard or obstacle at the location;
- determining that the decision point along the current route of the autonomous vehicle satisfies a predetermined risk threshold;
- and
- diverging the autonomous vehicle from the current route based at least in part on the decision point satisfying the predetermined risk threshold.
Dependencies are inferred from the claim text.