IP Library › Granted Patent US 12,071,148
Granted Patent B2
US 12,071,148 · App. 16/827,451 · Granted Aug 27, 2024

Switching decision for vehicle computational offloading to roadside edge server

Inventors: Roger Melen (Mountain View, CA); Dawei Chen (Mountain View, CA); BaekGyu Kim (Mountain View, CA)
B60W50/06G07C5/008B60W50/085G08G1/096811
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,071,148
App. No.
16/827,451
Granted
Aug 27, 2024
Kind
B2
Abstract

The disclosure includes embodiments that provide a switching decision for vehicle computational offloading to a roadside edge server. A method includes causing a sensor set of a connected vehicle to determine a current driving context of the connected vehicle. The method includes comparing the current driving context to a set of historical data to determine a predicted latency for using offboard computing resources of an edge server. The method includes determining that the predicted latency for using the offboard computing resources satisfies a threshold for the predicted latency. The method includes executing a switching decision that includes deciding to use the offboard computing resources of the edge server. The method includes causing the edge server to wirelessly provide digital data generated by the edge server responsive to a calculation. The method includes modifying an operation of the onboard vehicle computer based on the digital data.

Claims (37)

1. A method comprising:

causing a sensor set of a connected vehicle to determine a current driving context of the connected vehicle;

comparing, by an onboard vehicle computer of the connected vehicle, the current driving context to a set of historical data to determine a predicted latency for using offboard computing resources of an edge server which is remote from the connected vehicle, wherein the set of historical data describes, for a plurality of historical switching decisions, real-world historical latencies for using the offboard computing resources in a set of different driving contexts and whether using the offboard computing resources was successful on a decision-by-decision basis;

determining that the predicted latency for using the offboard computing resources satisfies a threshold for the predicted latency;

executing a switching decision that includes deciding to use the offboard computing resources of the edge server based on the comparing of the current driving context to the set of historical data and the determining that the threshold for the predicted latency is satisfied;

causing, responsive to the switching decision, the edge server to wirelessly provide digital data generated by the edge server responsive to a calculation; and

modifying an operation of the onboard vehicle computer based on the digital data generated by the calculation and provided to the connected vehicle.

2. The method of claim 1 , wherein the edge server executes a separate instance of the switching decision and a consensus between the switching decisions of the onboard vehicle computer and the edge server is required in order for the calculation to be executed by the edge server.

3. The method of claim 1 , wherein the edge server provides the digital data responsive to a calculation and further comprising determining whether the calculation improves the operation of the onboard vehicle computer.

4. The method of claim 3 , further comprising updating the historical data responsive to determining that the calculation does not improve the operation of the onboard vehicle computer so that a future switching decision is improved.

5. The method of claim 1 , further comprising determining an actual latency resulting from the switching decision.

6. The method of claim 5 , further comprising determining whether the actual latency is substantially equal to the predicted latency.

7. The method of claim 6 , further comprising updating the historical data responsive to determining that the actual latency is not substantially equal to the predicted latency so that a future switching decision is improved.

8. The method of claim 1 , wherein the digital data describes a sensor measurement recorded by a remote vehicle.

9. The method of claim 1 , wherein the digital data describes an output of a computational process executed by the edge server for the connected vehicle.

10. A non-transitory computer program product installed in an onboard unit of a connected vehicle, the computer program product including codes and routines that are operable, when executed by the onboard unit, to cause the onboard unit to execute routines including:

comparing a current driving context of the connected vehicle to a set of historical data to determine a predicted latency for using offboard computing resources of an edge server, wherein the set of historical data describes, for a plurality of historical switching decisions, real-world historical latencies for using the offboard computing resources in a set of different driving contexts and whether using the offboard computing resources was successful on a decision-by-decision basis;

determining that the predicted latency for using the offboard computing resources satisfies a threshold for the predicted latency;

executing a switching decision that includes deciding to use the offboard computing resources of the edge server based on the comparing and the determining;

responsive to the switching decision, wirelessly receiving digital data from the edge server; and

modifying an operation based on the digital data.

11. The computer program product of claim 10 , wherein the edge server executes a separate instance of the switching decision and a consensus between switching decisions of the onboard unit and the edge server is required in order for a calculation to be executed by the edge server.

12. The computer program product of claim 10 , wherein the edge server provides the digital data responsive to a calculation and the routines further include determining whether the calculation improves the operation of the onboard unit.

13. The computer program product of claim 12 , wherein the routines further include updating the historical data responsive to determining that the calculation does not improve the operation of the onboard unit so that a future switching decision is improved.

14. The computer program product of claim 10 , wherein the routines further include determining an actual latency resulting from the switching decision.

15. The computer program product of claim 14 , wherein the routines further include determining whether the actual latency is substantially equal to the predicted latency.

16. The computer program product of claim 15 , wherein the routines further include updating the historical data responsive to determining that the actual latency is not substantially equal to the predicted latency so that a future switching decision is improved.

17. The computer program product of claim 10 , wherein the digital data describes a sensor measurement recorded by a remote vehicle.

18. The computer program product of claim 10 , wherein the digital data describes an output of a computational process executed by the edge server for the connected vehicle.

19. A system of a connected vehicle comprising:

an onboard vehicle computer communicatively coupled to a non-transitory memory and a sensor set of the connected vehicle, wherein the onboard vehicle computer is operable to retrieve computer-executable code from the non-transitory memory which is operable, when executed by the onboard vehicle computer, to cause the onboard vehicle computer to:

compare a current driving context of the connected vehicle to a set of historical data to determine a predicted latency for using offboard computing resources of an edge server, wherein the set of historical data describes, for a plurality of historical switching decisions, real-world historical latencies for using the offboard computing resources in a set of different driving contexts and whether using the offboard computing resources was successful on a decision-by-decision basis;

determine that the predicted latency for using the offboard computing resources satisfies a threshold for the predicted latency;

execute a switching decision that includes deciding to use the offboard computing resources of the edge server based on the comparing and the determining;

responsive to the switching decision, wirelessly receiving digital data from the edge server; and

modify an operation based on the digital data.

20. The system of claim 19 , wherein the edge server executes a separate instance of the switching decision and a consensus between the switching decisions of the onboard vehicle computer and the edge server is required in order for the calculation to be executed by the edge server.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 8, 2025
From: TOYOTA MOTOR ENGINEERING & MANUFACTURING NORTH AMERICA, INC.
To: TOYOTA JIDOSHA KABUSHIKI KAISHA
Reel/Frame 069782/0390 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 24, 2020
From: KIM, BAEKGYU; CHEN, DAWEI; MELEN, ROGER
To: TOYOTA MOTOR ENGINEERING & MANUFACTURING NORTH AMERICA, INC.
Reel/Frame 052213/0381 →
Continuity (1)
Related Publication 20210291851A1 · Sep 23, 2021