Ballistic estimation of vehicle data

1. A system for ballistically estimating vehicle data, comprising:

a communication interface configured to receive a first vehicle measurement taken at a first time point and a second vehicle measurement taken at a second time point;

a plurality of processors, configured to:

compute, by a first processor of the plurality of processors, a first set of vehicle data based on the first vehicle measurement;

estimate, by a second processor of the plurality of processors, a second set of vehicle data for the second time point based on the first set of vehicle data and a model;

compute, by the first processor, a third set of vehicle data based on the second vehicle measurement; and

update the model based on a comparison between the second set of vehicle data and the third set of vehicle data; and

a storage configured to store the first and second vehicle measurements, the model, and the first, second, and third sets of vehicle data,

wherein the second processor estimates the second set of vehicle data at a faster speed than the first processor computes the third set of vehicle data.

2. The system of claim 1 , wherein the first and second vehicle measurements are image frames of a video feed captured by a sensor corresponding to the first time point and the second time point, respectively.

3. The system of claim 2 , wherein the sensor is a camera or a LiDAR.

4. The system of claim 1 , wherein the model is a learning network defined by a plurality of model parameters.

5. The system of claim 4 , wherein to update the model, the plurality of processors are configured to update the plurality of model parameters.

6. The system of claim 1 , wherein the model is configured to estimate the second set of vehicle data by modifying the first set of vehicle data based on movement characteristics of a vehicle.

7. The system of claim 6 , wherein the movement characteristics include a position, a velocity, an acceleration, or a direction of the vehicle.

8. The system of claim 1 , wherein the model is updated using a stochastic gradient descent method.

9. The system of claim 1 , wherein the model is updated to reduce a difference between the second set of vehicle data and the third set of vehicle data.

10. The system of claim 1 , wherein the first and second vehicle measurements are measurements of a vehicle taken at another vehicle, wherein the system is in the another vehicle.

11. A method for ballistically estimating vehicle data, comprising:

receiving, via a communication interface, a first vehicle measurement taken at a first time point and a second vehicle measurement taken at a second time point;

computing, by a first processor, a first set of vehicle data based on the first vehicle measurement;

estimating, by a second processor, a second set of vehicle data for the second time point based on the first set of vehicle data and a model;

computing, by the first processor, a third set of vehicle data based on the second vehicle measurement; and

updating, by the first or second processor, the model based on a comparison between the second set of vehicle data and the third set of vehicle data,

wherein the second processor estimates the second set of vehicle data at a faster speed than the first processor computes the third set of vehicle data.

12. The method of claim 11 , wherein the first and second vehicle measurements are image frames of a video feed captured by a sensor corresponding to the first time point and the second time point, respectively.

13. The method of claim 11 , wherein the model is a learning network defined by a plurality of model parameters.

14. The method of claim 13 , wherein updating the model includes updating the plurality of model parameters.

15. The method of claim 11 , further comprising estimating the second set of vehicle data by modifying the first set of vehicle data based on movement characteristics of a vehicle.

16. The method of claim 15 , wherein the movement characteristics include a position, a velocity, an acceleration, or a direction of the vehicle.

17. The method of claim 11 , wherein the model is updated using a stochastic gradient descent method.

18. The method of claim 11 , wherein the model is updated to reduce a difference between the second set of vehicle data and the third set of vehicle data.

19. A non-transitory computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a plurality of processors, performs a method for ballistically estimating vehicle data, comprising:

receiving a first vehicle measurement taken at a first time point and a second vehicle measurement taken at a second time point;

computing, by a first processor of the plurality of processors, a first set of vehicle data based on the first vehicle measurement;

estimating, by a second processor of the plurality of processors, a second set of vehicle data for the second time point based on the first set of vehicle data and a model;

computing, by the first processor, a third set of vehicle data based on the second vehicle measurement; and

updating the model based on a comparison between the second set of vehicle data and the third set of vehicle data,

wherein the second processor estimates the second set of vehicle data at a faster speed than the first processor computes the third set of vehicle data.

20. The non-transitory computer-readable medium of claim 19 , wherein the first and second vehicle measurements are image frames of a video feed captured by a sensor corresponding to the first time point and the second time point, respectively.