Trajectory Prediction for Autonomous Devices

A computing system comprising:

• one or more processors;
• one or more non-transitory computer-readable media storing instructions that when executed by the one or more processors cause the one or more processors to perform operations comprising:
• accessing trajectory data and goal path data, wherein the trajectory data comprises information associated with a predicted trajectory of an object, the predicted trajectory comprising a plurality of waypoints respectively associated with a plurality of waypoint position uncertainty distributions, and wherein the goal path data comprises information associated with a goal path and comprises a plurality of locations that the object is predicted to travel;
• determining a plurality of solution waypoints for the object based at least in part on application of one or more optimization techniques to the plurality of waypoints and the plurality of waypoint position uncertainty distributions, wherein the one or more optimization techniques comprise one or more operations to maximize a probability of each of the plurality of solution waypoints;
• and
• generating stitched trajectory data based at least in part on the plurality of solution waypoints, wherein the stitched trajectory data comprises information associated with a stitched trajectory associated with one or more portions of the plurality of solution waypoints and the goal path.

An autonomous vehicle comprising:

• one or more processors;
• one or more non-transitory computer-readable media storing instructions that when executed by the one or more processors cause the one or more processors to perform operations comprising:
• accessing trajectory data and goal path data, wherein the trajectory data comprises information associated with a predicted trajectory of an object, the predicted trajectory comprising a plurality of waypoints respectively associated with a plurality of waypoint position uncertainty distributions, and wherein the goal path data comprises information associated with a goal path and comprises a plurality of locations that the object is predicted to travel;
• determining a plurality of solution waypoints for the object based at least in part on application of one or more optimization techniques to the plurality of waypoints and the plurality of waypoint position uncertainty distributions, wherein the one or more optimization techniques comprise one or more operations to maximize a probability of each of the plurality of solution waypoints;
• and
• generating stitched trajectory data based at least in part on the plurality of solution waypoints, wherein the stitched trajectory data comprises information associated with a stitched trajectory associated with one or more portions of the plurality of solution waypoints and the goal path.