Systems and methods for improving vehicle predictions using point representations of scene

A system comprising:

• one or more processors and one or more computer-readable non-transitory storage media coupled to the one or more processors, the one or more computer-readable non-transitory storage media comprising instructions operable, when executed by the one or more processors, to cause the system to:
• receive sensor data of an environment external to a vehicle generated by one or more sensors of the vehicle;
• identify, based on the sensor data, one or more objects in the environment;
• generate, based on the one or more objects, a set of points that represent the environment, including transforming a spatial representation of the objects into a plurality of points by converting one or more bounding boxes representing the one or more objects into coordinates in a top-down view of the environment, wherein each object of the one or more objects includes one or more corresponding points in the set of points, and wherein the plurality of points represent only corners of a polygon in the top-down view for the one or more objects;
• generate a prediction for at least one of the one or more objects in the environment by processing the set of points;
• and
• cause the vehicle to perform one or more operations based on the prediction.

One or more computer-readable non-transitory storage media embodying software including instructions that is operable when executed, cause one or more processors to perform operations comprising:

• receiving sensor data of an environment external to a vehicle generated by one or more sensors of the vehicle;
• identifying, based on the sensor data, one or more objects in the environment;
• generating, based on the one or more objects, a set of points that represent the environment, including transforming a spatial representation of the objects into a plurality of points by converting one or more bounding boxes representing the one or more objects into coordinates in a top-down view of the environment, wherein each object of the one or more objects includes one or more corresponding points in the set of points, and wherein the plurality of points represent only corners of a polygon in the top-down view for the one or more objects;
• generating a prediction for at least one of the one or more objects in the environment by processing the set of points;
• and
• causing the vehicle to perform one or more operations based on the prediction.