Systems and methods for recommending an estimated time of arrival

Abstract

The present disclosure relates to a method and system for estimating an estimated time of arrival for an intended trip. The system includes at least one storage medium configured to store a machine learning model trained using historical trip data. The system further includes a processing engine configured to receive road-related data associated with an intended route of the intended trip. The intended route includes a plurality of road sections. The processing engine is further configured to determine global features of the intended route based on the road-related data. At least one global feature reflects a relationship between at least two road sections among the plurality of road sections. The processing engine is also configured to determine the estimated time of arrival based on the machine learning model and the global features of the intended route and provide the estimated time of arrival for display.

What is claimed is: 
     
         1 . A system for estimating an estimated time of arrival for an intended trip, comprising:
 at least one storage medium configured to store a machine learning model trained using historical trip data; and   a processing engine configured to:
 receive road-related data associated with an intended route of the intended trip, the intended route including a plurality of road sections; 
 determine global features of the intended route based on the road-related data, wherein at least one global feature reflects a relationship between at least two road sections among the plurality of road sections; 
 determine the estimated time of arrival based on the machine learning model and the global features of the intended route; and 
 provide the estimated time of arrival for display. 
   
     
     
         2 . The system of  claim 1 , wherein the processing engine is further configured to dynamically update the estimated time of arrival based on a traffic condition of the plurality of road sections. 
     
     
         3 . The system of  claim 1 , wherein the historical trip data used to train the machine learning model includes historical routes including at least a subset of the plurality of road sections and corresponding historical durations. 
     
     
         4 . The system of  claim 1 , wherein the road sections are divided based on at least one of levels of road, traffic lights, or geographical conditions. 
     
     
         5 . The system of  claim 1 , wherein the intended trip is requested by a user via a service request, and the intended route is determined based on the service request. 
     
     
         6 . The system of  claim 5 , wherein the processing engine is configured to obtain the intended route from a passenger terminal on which the user makes the service request. 
     
     
         7 . The system of  claim 5 , wherein the processing engine is configured to obtain the intended route from a driver terminal on which a driver confirms the service request. 
     
     
         8 . The system of  claim 1 , wherein the intended route is a first intended route, and the processing engine is further configured to:
 receive road-related data associated with a second intended route, wherein the first and second intended routes correspond to a map of a same geographic region;   determine global features based on the road-related data of the first and second intended routes, wherein at least one global feature reflects a relationship between a road section in the first intended route and a road section in the second intended route; and   determine the estimated time of arrival based on the machine learning model and the global features.   
     
     
         9 . The system of  claim 1 , wherein to determine the estimated time of arrival, the processing engine is further configured to:
 estimate a duration of the intended trip based on the machine learning model and the global features of the intended route; and   determine the estimated time of arrival based on a starting time of the intended trip and the estimated duration.   
     
     
         10 . The system of  claim 1 , wherein the global feature reflecting a relationship between at least two road sections is represented by a sparse matrix. 
     
     
         11 . The system of  claim 1 , wherein the machine learning model includes at least one of a regression model, a decision tree model, an artificial neural network model, a support vector regression model, or a restricted Boltzmann Machine model. 
     
     
         12 . The system of  claim 1 , wherein the machine learning model includes a plurality of sub-models each trained for a predetermined scenario, wherein the processing engine is further configured to:
 determine a scenario of the intended route;   select a sub-model trained for the scenario trained for the scenario of the intended route; and   determine the estimated time of arrival based on the sub-model and the global features of the intended route.   
     
     
         13 . A method for estimating an estimated time of arrival for an intended trip, comprising:
 receiving road-related data associated with an intended route of the intended trip, the intended route including a plurality of road sections;   determining, by a processing engine, global features of the intended route based on the road-related data, wherein at least one global feature reflects a relationship between at least two road sections among the plurality of road sections;   determining, by the processor engine, the estimated time of arrival based on a machine learning model and the global features of the intended route, wherein the machine learning model is trained using historical trip data; and   providing the estimated time of arrival for display.   
     
     
         14 . The method of  claim 13 , further comprising dynamically updating the estimated time of arrival based on a traffic condition of the plurality of road sections. 
     
     
         15 . The method of  claim 13 , wherein the historical trip data used to train the machine learning model includes historical routes including at least a subset of the plurality of road sections and corresponding historical durations. 
     
     
         16 . The method of  claim 13 , wherein the intended trip is requested by a user via a service request, and the intended route is determined based on the service request. 
     
     
         17 . The method of  claim 13 , wherein the global feature reflecting a relationship between at least two road sections is represented by a sparse matrix. 
     
     
         18 . The method of  claim 13 , wherein the machine learning model includes at least one of a regression model, a decision tree model, an artificial neural network model, a support vector regression model, or a restricted Boltzmann Machine model. 
     
     
         19 . The method of  claim 13 , wherein the machine learning model includes a plurality of sub-models each trained for a predetermined scenario, wherein determining the estimated time of arrival further comprises:
 determining a scenario of the intended route;   selecting a sub-model trained for the scenario trained for the scenario of the intended route; and   determining the estimated time of arrival based on the sub-model and the global features of the intended route.   
     
     
         20 . A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:
 receiving road-related data associated with an intended route of the intended trip, the intended route including a plurality of road sections;   determining global features of the intended route based on the road-related data, wherein at least one global feature reflects a relationship between at least two road sections among the plurality of road sections;   determining the estimated time of arrival based on a machine learning model and the global features of the intended route, wherein the machine learning model is trained using historical trip data; and   providing the estimated time of arrival for display.