Language models and machine learning frameworks for optimizing vehicle navigation routes and vehicle operator sessions
This disclosure relates to improved techniques for personalizing vehicle routes and operator sessions using pre-trained machine learning language models. In certain embodiments, a language model is trained on operator interaction data to learn operator route preferences for vehicle operators. These learned operator route preferences can be leveraged to optimize and personalize vehicle routes and operator sessions in various ways. Other embodiments are disclosed herein as well.
1. A method implemented via execution of computing instructions by one or more processors and stored on one or more non-transitory computer-readable storage devices, the method comprising:
providing a navigation application comprising:
a client interface that facilitates interactions between a language model and a vehicle operator; and
a route generation engine that is configured to compute vehicle routes;
collecting operator interaction data corresponding to interactions between the vehicle operator and the language model;
determining, by the language model, one or more operator route preferences based, at least in part, on the operator interaction data;
initiating a communication exchange between the language model and the route generation engine to identify a personalized vehicle route based, at least in part, on the operator route preferences determined by the language model; and
outputting, via the navigation application, the personalized vehicle route for the vehicle operator,
wherein:
the navigation application is a ride hailing application that enables the vehicle operator to accept requests for transporting passengers;
the personalized route is utilized in connection with transporting a passenger from an origin location to a destination location;
the one or more operator route preferences include a revenue preference of the vehicle operator;
the route generation engine generates a plurality of candidate vehicle routes based, at least in part, on the origin location and the destination location identified by the language model;
the route generation engine communicates the plurality of candidate vehicle routes to the language model; and
the language model selects the personalized driving route from the plurality of vehicle candidate routes based, at least in part, on the operator route preferences.
2. The method of claim 1 , wherein the language model is a generative pre-trained transformer (GPT) model that is trained to understand human language inputs received from the vehicle operator relating to scheduling vehicle routes and to learn the one or more operator route preferences based on the operator interaction data.
3. The method of claim 1 , wherein:
the ride hailing application executes a surge pricing function that is configured to dynamically adjust prices for passenger rides in a geographic area based on a current supply of automobile vehicle operators or a current demand for the automobile vehicle operators; and
based on the revenue preference of the vehicle operator, the personalized vehicle route is selected based, at least in part, by identifying the geographic regions where the prices for passenger rides are adjusted by the surge pricing function.
4. The method of claim 1 , wherein:
the language model generates scores for the plurality of candidate vehicle routes based, at least in part, on the operator route preferences; and
the personalized vehicle route is generated or selected based, at least in part, on the scores.
5. The method of claim 1 , wherein:
the language model is trained to learn a plurality of operator route preferences selected from a group consisting of: a ride duration preference, a distance preference, an operating area preference, a fuel preference, an intermediate stop preference, a dining preference, a revenue preference, a passenger preference, and a road preference; and
the language model determines or selects the personalized vehicle route based, at least in part, on the plurality of operator route preferences learned by the language model.
6. The method of claim 1 , wherein the language model is trained, at least, in part, on a domain-specific dataset comprising data related to scheduling vehicle routes and interactions with vehicle operators.
7. The method of claim 1 , wherein the language model communicates with the route generation engine to inform the route generation engine of the origin location and the destination location for the personalized vehicle route.
8. The method of claim 1 , where the one or more operator route preferences are derived from:
operator interaction data related to interactions between the vehicle operator and the language model in connection with identifying the personalized vehicle route; and
operator interaction data related to historical interactions between the vehicle operator and the language model.
9. The method of claim 8 , where the one or more operator route preferences are further derived from operator interaction data related to interactions between the vehicle operator and one or more third-party systems or applications.
10. The method of claim 1 , wherein:
the personalized vehicle route is generated in connection with a current operator session for the vehicle operator;
during the current operator session, the language model and the route generation engine communicate to generate a plurality of personalized vehicle routes;
the one or more operator route preferences are utilized to personalize the current operator session; and
the one or more operator route preferences are utilized to identify the plurality of personalized vehicle routes during the current operator session.
11. A system comprising:
one or more processors; and
one or more non-transitory computer-readable storage devices storing computing instructions configured to be executed on the one or more processors and cause the one or more processors to execute functions comprising:
providing a navigation application comprising:
a client interface that facilitates interactions between a language model and a vehicle operator; and
a route generation engine that is configured to compute vehicle routes;
collecting operator interaction data corresponding to interactions between the vehicle operator and the language model;
determining, by the language model, one or more operator route preferences based, at least in part, on the operator interaction data;
initiating a communication exchange between the language model and the route generation engine to identify a personalized vehicle route based, at least in part, on the operator route preferences determined by the language model; and
outputting, via the navigation application, the personalized vehicle route for the vehicle operator,
wherein:
the language model communicates with the route generation engine to inform the route generation engine of an origin location and destination location for the personalized vehicle route;
the route generation engine generates a plurality of candidate vehicle routes based, at least in part, on the origin location and the destination location identified by the language model;
the route generation engine communicates the plurality of candidate vehicle routes to the language model; and
the language model selects the personalized driving route from the plurality of vehicle candidate routes based, at least in part, on the operator route preferences.
12. The system of claim 11 , wherein the language model is a generative pre-trained transformer (GPT) model that is trained to understand human language inputs received from the vehicle operator relating to scheduling vehicle routes and to learn the one or more operator route preferences based on the operator interaction data.
13. The system of claim 11 , wherein:
the navigation application is a ride hailing application that enables the vehicle operator to accept requests for transporting passengers; and
the personalized route is utilized in connection with transporting a passenger from the origin location to the destination location.
14. The system of claim 13 , wherein:
the one or more operator route preferences include a revenue preference of the vehicle operator;
the ride hailing application executes a surge pricing function that is configured to dynamically adjust prices for passenger rides in a geographic area based on a current supply of automobile vehicle operators or a current demand for the automobile vehicle operators; and
based on the revenue preference of the vehicle operator, the personalized vehicle route is selected based, at least in part, by identifying the geographic regions where the prices for passenger rides are adjusted by the surge pricing function.
15. A method implemented via execution of computing instructions by one or more processors and stored on one or more non-transitory computer-readable storage devices, the method comprising:
providing a navigation application comprising:
a client interface that facilitates interactions between a language model and a vehicle operator; and
a route generation engine that is configured to compute vehicle routes;
collecting operator interaction data corresponding to interactions between the vehicle operator and the language model;
determining, by the language model, one or more operator route preferences based, at least in part, on the operator interaction data;
initiating a communication exchange between the language model and the route generation engine to identify a personalized vehicle route based, at least in part, on the operator route preferences determined by the language model; and
outputting, via the navigation application, the personalized vehicle route for the vehicle operator,
wherein:
the route generation engine generates a plurality of candidate vehicle routes based, at least in part, on an origin location and a destination location identified by the language model;
the route generation engine communicates the plurality of candidate vehicle routes to the language model; and
the language model selects the personalized driving route from the plurality of vehicle candidate routes based, at least in part, on the operator route preferences.
16. The method of claim 15 , wherein the language model is a generative pre-trained transformer (GPT) model that is trained to understand human language inputs received from the vehicle operator relating to scheduling vehicle routes and to learn the one or more operator route preferences based on the operator interaction data.
17. The method of claim 15 , wherein:
the language model is trained to learn a plurality of operator route preferences selected from a group consisting of: a ride duration preference, a distance preference, an operating area preference, a fuel preference, an intermediate stop preference, a dining preference, a revenue preference, a passenger preference, and a road preference; and
the language model determines or selects the personalized vehicle route based, at least in part, on the plurality of operator route preferences learned by the language model.
18. The method of claim 15 , wherein the language model is trained, at least, in part, on a domain-specific dataset comprising data related to scheduling vehicle routes and interactions with vehicle operators.
19. The method of claim 15 , wherein the language model communicates with the route generation engine to inform the route generation engine of the origin location and the destination location for the personalized vehicle route.
20. The method of claim 15 , wherein:
the language model generates scores for the plurality of candidate vehicle routes based, at least in part, on the operator route preferences; and
the personalized vehicle route is generated or selected based, at least in part, on the scores.
21. The method of claim 15 , where the one or more operator route preferences are derived from:
operator interaction data related to historical interactions between the vehicle operator and the language model; or
operator interaction data related to interactions between the vehicle operator and one or more third-party systems or applications.
22. The method of claim 15 , wherein:
the personalized vehicle route is generated in connection with a current operator session for the vehicle operator;
during the current operator session, the language model and the route generation engine communicate to generate a plurality of personalized vehicle routes;
the one or more operator route preferences are utilized to personalize the current operator session; and
the one or more operator route preferences are utilized to identify the plurality of personalized vehicle routes during the current operator session.
23. A system comprising:
one or more processors; and
one or more non-transitory computer-readable storage devices storing computing instructions configured to be executed on the one or more processors and cause the one or more processors to execute functions comprising:
providing a navigation application comprising:
a client interface that facilitates interactions between a language model and a vehicle operator; and
a route generation engine that is configured to compute vehicle routes;
collecting operator interaction data corresponding to interactions between the vehicle operator and the language model;
determining, by the language model, one or more operator route preferences based, at least in part, on the operator interaction data;
initiating a communication exchange between the language model and the route generation engine to identify a personalized vehicle route based, at least in part, on the operator route preferences determined by the language model; and
outputting, via the navigation application, the personalized vehicle route for the vehicle operator,
wherein:
the route generation engine generates a plurality of candidate vehicle routes based, at least in part, on an origin location and a destination location identified by the language model;
the route generation engine communicates the plurality of candidate vehicle routes to the language model; and
the language model selects the personalized driving route from the plurality of vehicle candidate routes based, at least in part, on the operator route preferences.
24. The system of claim 23 , wherein the language model is a generative pre-trained transformer (GPT) model that is trained to understand human language inputs received from the vehicle operator relating to scheduling vehicle routes and to learn the one or more operator route preferences based on the operator interaction data.
25. The system of claim 23 , wherein:
the navigation application is a ride hailing application that enables the vehicle operator to accept requests for transporting passengers; and
the personalized route is utilized in connection with transporting a passenger from the origin location to the destination location.
26. The system of claim 25 , wherein:
the one or more operator route preferences include a revenue preference of the vehicle operator;
the ride hailing application executes a surge pricing function that is configured to dynamically adjust prices for passenger rides in a geographic area based on a current supply of automobile vehicle operators or a current demand for the automobile vehicle operators; and
based on the revenue preference of the vehicle operator, the personalized vehicle route is selected based, at least in part, by identifying the geographic regions where the prices for passenger rides are adjusted by the surge pricing function.