Systems and Methods for Simulating Traffic Scenes
AssigneeAurora Operations, Inc.
InventorsShuhan Tan, Kelvin Ka Wing Wong, Shenlong Wang, Sivabalan Manivasagam, Mengye Ren, Raquel Urtasun
Life of this patent
Estimating the term…
Abstract
Example aspects of the present disclosure describe a scene generator for simulating scenes in an environment. For example, snapshots of simulated traffic scenes can be generated by sampling a joint probability distribution trained on real-world traffic scenes. In some implementations, samples of the joint probability distribution can be obtained by sampling a plurality of factorized probability distributions for a plurality of objects for sequential insertion into the scene.
Claims21
-20. (canceled)
A computer-implemented method for traffic scene generation, comprising:
- obtaining environmental data descriptive of an environment;
- generating data descriptive of a first object for insertion into a synthesized traffic scene in the environment, the data descriptive of the first object predicted using a machine-learned traffic scene generation framework and based at least in part on the environmental data;
- generating data descriptive of a second object for insertion into the synthesized traffic scene, the data descriptive of the second object predicted using the machine-learned traffic scene generation framework and based at least in part on the environmental data and the data descriptive of the first object;
- and
- outputting data descriptive of the synthesized traffic scene.
A computing system for traffic scene generation, the computing system comprising:
- 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 computing system to perform operations, the operations comprising:
- obtaining environmental data descriptive of an environment;
- generating data descriptive of a first object for insertion into a synthesized traffic scene in the environment, the data descriptive of the first object predicted using a machine-learned traffic scene generation framework and based at least in part on the environmental data;
- generating data descriptive of a second object for insertion into the synthesized traffic scene, the data descriptive of the second object predicted using the machine-learned traffic scene generation framework and based at least in part on the environmental data and the data descriptive of the first object;
- and
- outputting data descriptive of the synthesized traffic scene.
An autonomous vehicle control system comprising:
- one or more machine-learned models that have been trained using simulated sensor data representing at least a portion of a synthesized traffic scene, the simulated sensor data having been generated by performance of operations, the operations comprising:
- obtaining environmental data descriptive of an environment;
- generating data descriptive of a first object for insertion into a synthesized traffic scene in the environment, the data descriptive of the first object predicted using a machine-learned traffic scene generation framework and based at least in part on the environmental data;
- generating data descriptive of a second object for insertion into the synthesized traffic scene, the data descriptive of the second object predicted using the machine-learned traffic scene generation framework and based at least in part on the environmental data and the data descriptive of the first object;
- and
- outputting data descriptive of the synthesized traffic scene.
Dependencies are inferred from the claim text.