Fleet Maintenance Management for Autonomous Vehicles

• US 20190197798A1
• Filed: 12/22/2017
• Published: 06/27/2019
• Est. Priority Date: 12/22/2017
• Status: Active Grant

• Current Assignee: LYFT Incorporated
• Original Assignee: LYFT Incorporated
• Inventors: Abari, Farzad Cyrus Foroughi, Friedman, Aaron Jacob Levine, Houston, John, Reis, Adam Hart, Starns, Alexander Thomas
• Granted Patent: US 10,726,644 B2
• Time in Patent Office: Days
• CPC Class Codes: G05D 1/0088 characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00
• CPC Class Codes: G05D 1/0291 Fleet control monitoring fleets in traffic control systems for road vehicles G08G1/127, G08G1/127
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/02 Reservations, e.g. for tickets, services or events
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis
• CPC Class Codes: G06Q 10/20 Administration of product repair or maintenance
• CPC Class Codes: G06Q 50/40 Business processes related to the transportation industry shipping G06Q10/083
• CPC Class Codes: G07C 5/006 Indicating maintenance
• CPC Class Codes: G07C 5/0808 Diagnosing performance data testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005

Abstract

In particular embodiments, a computing system may generate a prediction of requests for autonomous vehicles in a fleet of collectively managed autonomous vehicles based on a current condition and a future event, the prediction including a predicted request level and a predicted duration of the request level. The system may receive status information from fleet vehicles and identify a vehicle in need of service. The system may receive status information from service centers and identify a service center to service the vehicle. The system may determine a time at which to service the vehicle at the identified service center based on the generated prediction of requests, schedule the vehicle for service at the identified service center at the determined time, and instruct the vehicle to drive to the service center to be serviced at the determined time. In particular embodiments, the prediction of requests may be generated using machine learning.

Claims

1. A method comprising, by a computing system:

generating a prediction of requests for autonomous vehicles in a fleet of collectively managed autonomous vehicles based on a current condition and a future event, the prediction comprising a predicted request level and a predicted duration of the predicted request level;

receiving respective status information from the autonomous vehicles in the fleet of autonomous vehicles;

identifying an autonomous vehicle in the fleet of autonomous vehicles in need of service based on the status information received from the autonomous vehicle;

determining a time at which the autonomous vehicle is to be serviced at an identified service center based on the generated prediction of requests;

scheduling the autonomous vehicle for service at the identified service center at the determined time; and

instructing the autonomous vehicle to drive to the identified service center to be serviced at the determined time.

2. The method of claim 1, wherein generating the prediction of requests is further based on historic data including a number of requests for autonomous vehicles in a region in which the fleet of autonomous vehicles operates and a number of autonomous vehicles available to fulfill requests in the region in which the fleet of autonomous vehicles operates.

3. The method of claim 1, wherein:

generating the prediction of requests is further based on respective characteristics of one or more autonomous vehicles in the fleet of autonomous vehicles including the autonomous vehicle, the characteristics including at least one of a vehicle type, a vehicle class, or an optional vehicle feature; and

determining the time at which the autonomous vehicle is to be serviced is further based on a characteristic of the autonomous vehicle.

4. The method of claim 1, wherein determining the time at which the autonomous vehicle is to be serviced comprises at least one of:

determining, based on the generated prediction of requests, a default interval between service appointments for each of one or more autonomous vehicles in the fleet of autonomous vehicles including the autonomous vehicle; and

determining, based on the generated prediction of requests, an order in which multiple autonomous vehicles in the fleet of autonomous vehicles are to be serviced.

5. The method of claim 1, wherein:

the generated prediction of requests comprises a respective predicted request level for each of multiple locations within a region in which the fleet of autonomous vehicles operates; and

the method further comprises;

determining, based on the generated prediction of requests, a location within the region at which the autonomous vehicle is to be positioned subsequent to receiving service at the identified service center; and

sending instructions to the autonomous vehicle to drive to the determined location subsequent to receiving service at the identified service center.

6. The method of claim 1, further comprising:

receiving respective status information from at least one service center; and

identifying the service center at which the autonomous vehicle is to be serviced based on the status information received from the at least one service center.

7. The method of claim 6, wherein:

identifying the service center is further based on at least one of a particular service to be performed on the autonomous vehicle, and a capability of the identified service center to perform the particular service; and

determining the time at which the autonomous vehicle is to be serviced is based on at least one of capacity of the identified service center to perform the particular service at the determined time, an estimated amount of time required to perform the particular service at the identified service center, and an estimated travel time from a current location of the autonomous vehicle to the identified service center.

8. The method of claim 1, wherein:

the method further comprises determining that the number of autonomous vehicles in the fleet of autonomous vehicles available to fulfill requests exceeds the predicted request level;

a plurality of services are needed by the autonomous vehicle; and

scheduling the autonomous vehicle for service comprises scheduling the autonomous vehicle for all of the plurality of needed services.

9. The method of claim 1, wherein:

the method further comprises determining that the predicted request level exceeds the number of autonomous vehicles in the fleet of autonomous vehicles available to fulfill requests;

a plurality of services are needed by the autonomous vehicle; and

scheduling the autonomous vehicle for service comprises scheduling the autonomous vehicle for fewer than all of the plurality of services needed by the autonomous vehicle.

10. The method of claim 1, wherein:

generating the prediction comprises using a machine-learning model to generate the predicted request level or the predicted duration of the predicted request level; and

the machine-learning model is trained using a set of training data, at least one training sample in the set of training data being associated with the current condition or the future event.

11. The method of claim 1, wherein:

the future event comprises a known upcoming event likely to affect the number of requests in a region in which the fleet of autonomous vehicles operates or a predicted upcoming event likely to affect the number of requests in the region in which the fleet of autonomous vehicles operates; and

the current condition comprises a condition affecting traffic in the region in which the fleet of autonomous vehicles operates.

12. A system comprising:

one or more processors and one or more computer-readable non-transitory storage media coupled to one or more of the processors, the one or more computer-readable non-transitory storage media comprising instructions operable when executed by one or more of the processors to cause the system to perform operations comprising;

generating a prediction of requests for autonomous vehicles in a fleet of collectively managed autonomous vehicles based on a current condition and a future event, the prediction comprising a predicted request level and a predicted duration of the predicted request level;

receiving respective status information from the autonomous vehicles in the fleet of autonomous vehicles;

identifying an autonomous vehicle in the fleet of autonomous vehicles in need of service based on the status information received from the autonomous vehicle;

determining a time at which the autonomous vehicle is to be serviced at an identified service center based on the generated prediction of requests;

scheduling the autonomous vehicle for service at the identified service center at the determined time; and

causing instructions to be communicated to the autonomous vehicle to drive to the identified service center to be serviced at the determined time.

13. The system of claim 12, wherein generating the prediction of requests is further based on one or more of:

historic data including a number of requests for autonomous vehicles in a region in which the fleet of autonomous vehicles operates and a number of autonomous vehicles available to fulfill requests in the region in which the fleet of autonomous vehicles operates; and

respective characteristics of one or more autonomous vehicles in the fleet of autonomous vehicles including the autonomous vehicle, the characteristics including at least one of a vehicle type, a vehicle class, or an optional vehicle feature.

14. The system of claim 12, wherein the processors are further operable when executing the instructions to perform operations comprising:

scheduling the autonomous vehicle for all of a plurality of services needed by the autonomous vehicle when the number of autonomous vehicles in the fleet of autonomous vehicles available to fulfill requests exceeds the predicted request level; and

scheduling the autonomous vehicle for fewer than all of the plurality of services needed by the autonomous vehicle when the predicted request level exceeds the number of autonomous vehicles in the fleet of autonomous vehicles available to fulfill requests.

15. The system of claim 12, wherein:

the processors are further operable when executing the instructions to perform operations comprising identifying the service center at which the autonomous vehicle is to be serviced based on at least one of status information received from the identified service center, a particular service to be performed on the autonomous vehicle, and a capability of the identified service center to perform the particular service; and

determining the time at which the autonomous vehicle is to be serviced is based on at least one of capacity of the identified service center to perform the particular service at the determined time, an estimated amount of time required to perform the particular service at the identified service center, and an estimated travel time from a current location of the autonomous vehicle to the identified service center.

16. The system of claim 12, wherein:

generating the prediction comprises using a machine-learning model to generate the predicted request level or the predicted duration of the predicted request level; and

the machine-learning model is trained using a set of training data, at least one training sample in the set of training data being associated with the current condition or the future event.

17. One or more computer-readable non-transitory storage media embodying software that is operable when executed to cause one or more processors to perform operations comprising:

generating a prediction of requests for autonomous vehicles in a fleet of collectively managed autonomous vehicles based on a current condition and a future event, the prediction comprising a predicted request level and a predicted duration of the predicted request level;

receiving respective status information from the autonomous vehicles in the fleet of autonomous vehicles;

identifying an autonomous vehicle in the fleet of autonomous vehicles in need of service based on the status information received from the autonomous vehicle;

determining a time at which the autonomous vehicle is to be serviced at an identified service center based on the generated prediction of requests;

scheduling the autonomous vehicle for service at the identified service center at the determined time; and

causing instructions to be communicated to the autonomous vehicle to drive to the identified service center to be serviced at the determined time.

18. The one or more computer-readable non-transitory storage media of claim 17, wherein generating the prediction of requests is further based on one or more of:

historic data including a number of requests for autonomous vehicles in a region in which the fleet of autonomous vehicles operates and a number of autonomous vehicles available to fulfill requests in the region in which the fleet of autonomous vehicles operates; and

respective characteristics of one or more autonomous vehicles in the fleet of autonomous vehicles including the autonomous vehicle, the characteristics including at least one of a vehicle type, a vehicle class, or an optional vehicle feature.

19. The one or more computer-readable non-transitory storage media of claim 17, wherein:

when executed the software is further operable to cause the one or more processors to perform operations comprising identifying the service center at which the autonomous vehicle is to be serviced based on at least one of status information received from the identified service center, a particular service to be performed on the autonomous vehicle, and a capability of the identified service center to perform the particular service; and

determining the time at which the autonomous vehicle is to be serviced is based on at least one of capacity of the identified service center to perform the particular service at the determined time, an estimated amount of time required to perform the particular service at the identified service center, and an estimated travel time from a current location of the autonomous vehicle to the identified service center.

20. The one or more computer-readable non-transitory storage media of claim 17, wherein:

generating the prediction comprises using a machine-learning model to generate the predicted request level or the predicted duration of the predicted request level; and

the machine-learning model is trained using a set of training data, at least one training sample in the set of training data being associated with the current condition or the future event.