Autonomous vehicles featuring machine-learned yield model

An autonomous vehicle, comprising:

• a machine-learned yield model configured to receive and process feature data descriptive of objects perceived by the autonomous vehicle and, in response, provide yield decisions for the autonomous vehicle relative to the objects, the machine-learned yield model having been trained by a machine learning computing system, the machine learning computing system configured to:
• generate a plurality of synthetic autonomous vehicle trajectories;
• simulate an outcome for each synthetic autonomous vehicle trajectory;
• assign a yield label for each synthetic autonomous vehicle trajectory based at least in part on the outcome for such synthetic autonomous vehicle trajectory;
• train the machine-learned yield model based at least in part on the yield labels assigned to the synthetic autonomous vehicle trajectories;
• one or more processors;
• and
• one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the one or more processors to:
• obtain first feature data descriptive of a first object perceived by the autonomous vehicle;
• input the first feature data into the machine-learned yield model;
• receive a first yield decision as an output of the machine-learned yield model, wherein the first yield decision describes a yield behavior for the autonomous vehicle relative to the first object;
• generate a motion plan that executes the yield behavior described by the first yield decision output by the machine-learned yield model;
• and control motion of the autonomous vehicle based at least in part motion plan.

A computer-implemented method, comprising:

• generating, by a computing system that comprises one or more computing devices, a plurality of synthetic autonomous vehicle trajectories;
• simulating, by the computing system, an outcome for each synthetic autonomous vehicle trajectory;
• assigning, by the computing system, a yield label for each synthetic autonomous vehicle trajectory based at least in part on the outcome for such synthetic autonomous vehicle trajectory;
• training, by the computing system, a machine-learned yield model based at least in part on the yield labels assigned to the synthetic autonomous vehicle trajectories;
• obtaining, by the computing system, first feature data descriptive of a first object perceived by an autonomous vehicle;
• inputting, by the computing system, the first feature data into the machine-learned yield model;
• receiving, by the computing system, a first yield decision as an output of the machine learned yield model, wherein the first yield decision describes a yield behavior for the autonomous vehicle relative to the first object;
• generating, by the computing system, a motion plan that executes the yield behavior described by the first yield decision output by the machine-learned yield model;
• and
• controlling, by the computing system, motion of the autonomous vehicle based at least in part motion plan.