Assignment in a vehicular micro cloud based on sensed vehicle maneuvering

1 . A method to improve an operation of a vehicular micro cloud by increasing a continuity of a vehicular micro cloud service provided by the vehicular micro cloud, the method comprising:

maintaining a data structure of stored driving maneuver shapes, wherein each stored driving maneuver shape includes a set of driving maneuvers determined based at least in part on execution of a set of digital twin simulations that model vehicle behaviors in a vehicular micro cloud environment;

sensing a candidate vehicle and a driving maneuver shape of the candidate vehicle;

determining a matching driving maneuver shape from the stored driving maneuver shapes based on each driving maneuver in the driving maneuver shape being within a threshold similarity value of a corresponding matching driving maneuver from the stored driving maneuver shapes;

estimating a next driving maneuver of the candidate vehicle based at least in part on the execution of the set of digital twin simulations that model vehicle behaviors in the vehicular micro cloud environment to predict whether the candidate vehicle will leave the vehicular micro cloud within a predetermined time period; and

assigning a role to the candidate vehicle in the vehicular micro cloud based on the next driving maneuver of the candidate vehicle, wherein the role includes executing one or more vehicular micro cloud tasks whose execution provides the vehicular micro cloud service and handing over the one or more vehicular micro cloud tasks to another vehicle in the vehicular micro cloud upon the candidate vehicle leaving the vehicular micro cloud to maintain the continuity of the vehicular micro cloud service.

2 . The method of claim 1 , wherein estimating the next driving maneuver is based on the set of driving maneuvers in the matching driving maneuver shape including steps not yet performed by the candidate vehicle and wherein the estimate includes a prediction that candidate vehicle will leave the vehicular micro cloud with the next driving maneuver, wherein the steps not yet performed are simulated in the set of digital twin simulations to refine the prediction.

3 . The method of claim 1 , wherein:

estimating the next driving maneuver of the candidate vehicle is performed by a machine-learning model; and

the machine-learning model was trained using one or more of data from a digital twin simulation, observed behavior of drivers that is location dependent, or accident data that is location dependent.

4 . The method of claim 3 , further comprising:

sensing that the candidate vehicle left the vehicular micro cloud; and

modifying parameters of the machine-learning model.

5 . The method of claim 1 , wherein sensing the driving maneuver shape of the candidate vehicle occurs responsive to the candidate vehicle entering the vehicular micro cloud.

6 . The method of claim 1 , wherein the role includes serving as a hub for the vehicular micro cloud.

7 . The method of claim 1 , wherein the role includes executing one or more vehicular micro cloud tasks whose execution provides the vehicular micro cloud service and serving as a central hub that coordinates communication and data exchange among vehicles in the vehicular micro cloud to support the execution of the one or more vehicular micro cloud tasks.

8 . The method of claim 1 , wherein the method is executed by an edge server that is an element of a roadside unit.

9 . The method of claim 1 , wherein the role is determined based at least in part on an execution of the set of digital twin simulations, wherein the set of digital twin simulations model location-dependent driver behaviors.

10 . The method of claim 1 , wherein the next driving maneuver is estimated based at least in part on an execution of the set of digital twin simulations, wherein the execution of the set of digital twin simulations incorporates observed behavior of drivers that is location dependent.

11 . The method of claim 1 , wherein the matching driving maneuver shape is determined based at least in part on an execution of the set of digital twin simulations.

12 . The method of claim 1 , wherein the stored driving maneuver shapes are determined based at least in part on execution of the set of digital twin simulations, wherein the execution of the set of digital twin simulations incorporates accident data that is location dependent.

13 . The method of claim 1 , wherein the next driving maneuver indicates that the candidate vehicle is predicted to leave the vehicular micro cloud within a predetermined time period and the role is assigned based on predicting that the candidate vehicle will leave the vehicular micro cloud within the predetermined time period, wherein the predetermined time period is dynamically adjusted based on outputs from the set of digital twin simulations.

14 . A system comprising:

a non-transitory memory;

a vehicle control system; and

a processor communicatively coupled to the non-transitory memory and the vehicle control system, wherein the non-transitory memory stores computer readable code that is operable, when executed by the processor, to cause the processor to execute operations including:

maintaining a data structure of stored driving maneuver shapes, wherein each stored driving maneuver shape includes a set of driving maneuvers determined based at least in part on execution of a set of digital twin simulations that model vehicle behaviors in a vehicular micro cloud environment;

sensing a candidate vehicle and a driving maneuver shape of the candidate vehicle;

determining a matching driving maneuver shape from the stored driving maneuver shapes based on each driving maneuver in the driving maneuver shape being within a threshold similarity value of a corresponding matching driving maneuver from the stored driving maneuver shapes;

estimating a next driving maneuver of the candidate vehicle based at least in part on the execution of the set of digital twin simulations that model vehicle behaviors in the vehicular micro cloud environment to predict whether the candidate vehicle will leave the vehicular micro cloud within a predetermined time period; and

assigning a role to the candidate vehicle in the vehicular micro cloud based on the next driving maneuver of the candidate vehicle, wherein the role includes executing one or more vehicular micro cloud tasks whose execution provides the vehicular micro cloud service and handing over the one or more vehicular micro cloud tasks to another vehicle in the vehicular micro cloud upon the candidate vehicle leaving the vehicular micro cloud to maintain the continuity of the vehicular micro cloud service.

15 . The system of claim 14 , wherein estimating the next driving maneuver is based on the set of driving maneuvers in the matching driving maneuver shape including steps not yet performed by the candidate vehicle and wherein the estimate includes a prediction that candidate vehicle will leave the vehicular micro cloud with the next driving maneuver, wherein the steps not yet performed are simulated in the set of digital twin simulations to refine the prediction.

16 . The system of claim 14 , wherein:

estimating the next driving maneuver of the candidate vehicle is performed by a machine-learning model; and

the machine-learning model was trained using one or more of data from a digital twin simulation, observed behavior of drivers that is location dependent, or accident data that is location dependent.

17 . The system of claim 16 , wherein the operations further comprise:

sensing that the candidate vehicle left the vehicular micro cloud; and

modifying parameters of the machine-learning model.

18 . The system of claim 14 , wherein sensing the driving maneuver shape of the candidate vehicle occurs responsive to the candidate vehicle entering the vehicular micro cloud.

19 . The system of claim 14 , wherein the role includes serving as a hub for the vehicular micro cloud.

20 . A computer program product including computer code stored on a non-transitory memory that is operable, when executed by an onboard vehicle computer of a vehicle, to cause the onboard vehicle computer to execute operations including:

maintaining a data structure of stored driving maneuver shapes, wherein each stored driving maneuver shape includes a set of driving maneuvers determined based at least in part on execution of a set of digital twin simulations that model vehicle behaviors in a vehicular micro cloud environment;

sensing a candidate vehicle and a driving maneuver shape of the candidate vehicle;

determining a matching driving maneuver shape from the stored driving maneuver shapes based on each driving maneuver in the driving maneuver shape being within a threshold similarity value of a corresponding matching driving maneuver from the stored driving maneuver shapes;

estimating a next driving maneuver of the candidate vehicle based at least in part on the execution of the set of digital twin simulations that model vehicle behaviors in the vehicular micro cloud environment to predict whether the candidate vehicle will leave the vehicular micro cloud within a predetermined time period; and

assigning a role to the candidate vehicle in the vehicular micro cloud based on the next driving maneuver of the candidate vehicle, wherein the role includes executing one or more vehicular micro cloud tasks whose execution provides the vehicular micro cloud service and handing over the one or more vehicular micro cloud tasks to another vehicle in the vehicular micro cloud upon the candidate vehicle leaving the vehicular micro cloud to maintain the continuity of the vehicular micro cloud service.