Vehicle command generation using vehicle-to-infrastructure communications and deep networks

1. A system to command vehicles via a vehicle-to-infrastructure communication network, comprising:

a first roadside computing device positioned at a first location in a vehicular travel path, the first roadside computing device to broadcast one or more data packets comprising a timestamp;

a first vehicle comprising an onboard computing device to:

receive the one or more data packets broadcast by the first roadside computing device;

calibrate, based on the timestamp received via the one or more data packets broadcast by the first roadside computing device, an internal clock of the first vehicle to adjust a clock drift resulting from the internal clock of the first vehicle counting at a different rate relative to the first roadside computing device; and

transmit, responsive to receipt of the one or more data packets and calibration of the internal clock, a status of the first vehicle to the first roadside computing device;

the first roadside computing device to:

receive the status of the first vehicle transmitted by the first vehicle;

generate a second one or more data packets comprising the status of the first vehicle, and information associated with the first location in the vehicular travel path; and

transmit the second one or more data packets to a data processing system;

the data processing system comprising one or more processors and memory, a command generator component and a deep learning engine, the data processing system to:

input the status of the first vehicle and the information associated with the first location into the deep learning engine;

assign, based on an output from the deep learning engine, a label to the first vehicle;

select a vehicle command based on the label; and

transmit the vehicle command to the first vehicle to execute an action for traversing the vehicular travel path.

2. The system of claim 1 , wherein the first roadside computing device comprises a traffic light controller, comprising:

the first roadside computing device to determine a current light status; and

provide the current light status with the information associated with the first location in the vehicular travel path.

3. The system of claim 1 , comprising:

the first roadside computing device to:

receive, within a predetermined time window based on receipt of the status of the first vehicle, a status from each of a plurality of nearby vehicles different from the first vehicle; and

generate the second one or more data packets comprising the status from each of the plurality of nearby vehicles received within the predetermined time window; and

the data processing system to:

input the status of each of the plurality of nearby vehicles into the deep learning engine to assign a label to each of the plurality of nearby vehicles.

4. The system of claim 1 , comprising:

the first roadside computing device to generate the second one or more data packets comprising a status from each of a plurality of nearby vehicles received within a predetermined time window; and

the data processing system to:

input the status of each of the plurality of nearby vehicles into the deep learning engine to assign a label to each of the plurality of nearby vehicles; and

transmit a second vehicle command to at least one vehicle of the plurality of nearby vehicles to cause the at least one vehicle to execute an action for traversing the vehicular travel path.

5. The system of claim 1 , comprising:

the onboard computing device of the first vehicle to synchronize or syntonize the internal clock based on the timestamp to calibrate the internal clock; and

transmit the status of the first vehicle along with a calibrated timestamp produced by the internal clock.

6. The system of claim 1 , comprising:

the data processing system to determine the output via the deep learning engine configured based on a neural network having an input processing layer, an output processing layer, and a plurality of processing layers intermediary to the input layer and the output layer.

7. The system of claim 1 , comprising:

the data processing system to use the deep learning engine to generate the label comprising a driving style.

8. The system of claim 1 , comprising:

the data processing system to select the vehicle command comprising an instruction to cause the onboard computing device of the first vehicle to display a notification via a display device communicatively coupled to the onboard computing device.

9. The system of claim 1 , comprising:

the data processing system to select the vehicle command comprising an instruction to cause the onboard computing device of the first vehicle to adjust a speed of the first vehicle.

10. The system of claim 1 , comprising:

the onboard computing device of the first vehicle to determine the status of the first vehicle comprising at least one of a position, speed, acceleration, turn angle, throttle, or brake.

11. The system of claim 1 , comprising:

the first roadside computing device to store preconfigured information comprising at least one of a road geometry associated with the vehicular travel path, a speed limit, or a traffic sign.

12. The system of claim 1 , comprising:

the first roadside computing device to transmit the information comprising preconfigured information stored in memory of the first roadside computing device and a current condition associated with the first location.

13. The system of claim 1 , comprising the data processing system to:

receive, from a second roadside device, third data packets comprising a second status of the first vehicle; and

generate an updated label for the first vehicle based on the status and the second status.

14. The system of claim 1 , comprising the data processing system to:

assign the label to the first vehicle responsive to receipt of the second one or more data packets comprising the status of the first vehicle, and the information associated with the first location in the vehicular travel path.

15. The system of claim 1 , comprising the data processing system to:

transmit the vehicle command responsive to receipt of the second one or more data packets comprising the status of the first vehicle, and the information associated with the first location in the vehicular travel path.

16. The system of claim 1 , comprising the data processing system to:

transmit the label assigned to the first vehicle to a third-party computing device.

17. A method of commanding vehicles via a vehicle-to-infrastructure communication network, comprising:

broadcasting, by a first roadside computing device positioned at a first location in a vehicular travel path, one or more data packets comprising a timestamp;

receiving, by a first vehicle comprising an onboard computing device, the one or more data packets broadcast by the first roadside computing device;

calibrating, by the onboard computing device of the first vehicle based on the timestamp received via the one or more data packets broadcast by the first roadside computing device, an internal clock of the first vehicle to adjust a clock drift resulting from the internal clock of the first vehicle counting at a different rate relative to the first roadside computing device;

transmitting, by the onboard computing device of the first vehicle responsive to receipt of the one or more data packets and calibration of the internal clock, a status of the first vehicle to the first roadside computing device;

receiving, by the first roadside computing device, the status of the first vehicle;

generating, by the first roadside computing device, a second one or more data packets comprising the status of the first vehicle and information associated with the first location in the vehicular travel path;

transmitting, by the first roadside computing device, the second one or more data packets to a data processing system;

inputting, by the data processing system, the status of the first vehicle and the information associated with the first location into a deep learning engine;

assigning, by the data processing system based on an output from the deep learning engine, a label to the first vehicle;

selecting, by the data processing system, a vehicle command based on the label; and

transmitting, by the data processing system, the vehicle command to the first vehicle to cause the first vehicle to execute an action for traversing the vehicular travel path.

18. The method of claim 17 , wherein the first roadside computing device comprises a traffic light controller, comprising:

determining, by the first roadside computing device, a current light status; and

providing, by the first roadside computing device, the current light status with the information associated with the first location in the vehicular travel path.

19. A system to command vehicles via a vehicle-to-infrastructure communication network, comprising:

a data processing system comprising one or more processors and memory, a command generator component and a deep learning engine, the data processing system to:

receive, from a roadside computing device positioned at a vehicular travel path, one or more data packets comprising a status for a vehicle, a time stamp, and information associated with a location of the vehicle in the vehicular travel path, wherein the vehicle transmits the status to the roadside computing device responsive to calibration, based on a timestamp of the roadside computing device, of an internal clock of the vehicle to adjust a clock drift resulting from the internal clock of the vehicle counting at a different rate relative to the roadside computing device;

input the status of the vehicle and the information associated with the location into the deep learning engine;

assign, based on an output from the deep learning engine, a label to the vehicle;

select a vehicle command based on the label; and

transmit the vehicle command to the first vehicle to execute an action for traversing the vehicular travel path.

20. The system of claim 19 , comprising the data processing system to:

receive, from a second roadside device, second data packets comprising a second status of the vehicle; and

generate an updated label for the vehicle based on the status and the second status.