CN111862585A

CN111862585A - System and method for traffic prediction

Info

Publication number: CN111862585A
Application number: CN201910665749.0A
Authority: CN
Inventors: 仇辉; 李海波
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2019-07-23
Filing date: 2019-07-23
Publication date: 2020-10-30
Anticipated expiration: 2039-07-23
Also published as: CN111862585B; WO2021012342A1

Abstract

The present application relates to systems and methods for predicting traffic parameters. The system may perform a method to obtain at least two first inputs relating to a driving condition corresponding to at least two candidate points in time; acquiring a first training prediction model; determining a target vector by using a first trained predictive model based on at least two first inputs; obtaining at least two second inputs relating to the driving condition corresponding to at least two future points in time; acquiring a second training prediction model; and predicting, by using the second trained predictive model, at least two future parameters related to the driving condition corresponding to at least two future points in time based on the target vector and the at least two second inputs.

Description

System and method for traffic prediction

Technical Field

The present application relates generally to systems and methods for traffic prediction and, more particularly, to systems and methods for predicting at least two future parameters related to traffic conditions.

Background

With the rapid development of traffic environment, efficient and accurate traffic information prediction becomes more and more important for people going out. Systems that provide traffic services (e.g., online-to-offline traffic services, navigation services, map services) may predict future traffic information based on current traffic information and/or historical traffic information according to a linear model or a tree model. However, it is difficult to fuse multi-channel data according to a linear model or a tree model, and to predict traffic information corresponding to a plurality of future time points, which may result in a relatively large deviation. Accordingly, it is desirable to provide systems and methods for fusing multi-channel data and accurately and efficiently predicting future traffic information.

Disclosure of Invention

A first aspect of the present application relates to a system for predicting traffic parameters. The system may include at least one storage medium comprising a set of instructions, and at least one processor in communication with the at least one storage medium. The set of instructions, when executed, may instruct at least one processor to: acquiring at least two first inputs related to the driving condition corresponding to at least two candidate time points; acquiring a first training prediction model; determining a target vector by using a first trained predictive model based on at least two first inputs; obtaining at least two second inputs relating to the driving condition corresponding to at least two future points in time; acquiring a second training prediction model; and predicting, by using the second trained predictive model, at least two future parameters related to the driving condition corresponding to at least two future points in time based on the target vector and the at least two second inputs.

In some embodiments, the driving condition may be associated with a road segment.

In some embodiments, each of the at least two first inputs may include a first parameter related to traffic conditions for a respective candidate point in time, at least two first historical parameters related to traffic conditions corresponding to at least two first historical points in time, respectively, and/or a first statistical parameter related to the at least two first historical parameters.

In some embodiments, the first parameter related to traffic conditions may include at least one of a traffic congestion level of the driving conditions, a traffic speed of the driving conditions, and/or a traffic flow of the driving conditions.

In some embodiments, the first statistical parameter may include a traffic congestion statistical parameter, a traffic speed statistical parameter, and/or a traffic flow statistical parameter.

In some embodiments, the traffic congestion statistical parameter may include at least one of a mode in the at least two historical traffic congestion levels and/or a congestion probability of the at least two historical traffic congestion levels.

In some embodiments, the traffic speed statistical parameter may include at least one of a mean of the at least two historical speeds, a median of the at least two historical speeds, a variance of the at least two historical speeds, a maximum of the at least two historical speeds, and/or a minimum of the at least two historical speeds.

In some embodiments, the traffic flow statistical parameter may include at least one of a mean of the at least two historical traffic flows, a median of the at least two historical traffic flows, a variance of the at least two historical traffic flows, a maximum of the at least two historical traffic flows, and/or a minimum of the at least two historical traffic flows.

In some embodiments, each of the at least two second inputs may include a second parameter related to traffic conditions at a previous time point of the respective future time point, at least two second historical parameters related to traffic conditions respectively corresponding to the at least two second historical time points, and/or a second statistical parameter related to the at least two second historical traffic parameters.

In some embodiments, each of the at least two second inputs may further comprise a reference parameter comprising weather information at a respective future point in time.

In some embodiments, the first trained predictive model may be a first part of the trained predictive model and the second trained predictive model may be a second part of the trained predictive model.

In some embodiments, the trained predictive model may be a sequence-to-sequence model. The first part of the trained prediction model may be an encoder and the second part of the trained prediction model may be a decoder.

In some embodiments, the training prediction model may be determined based on a training process. The training process may include obtaining at least two first sample inputs corresponding to at least two first sample time points, respectively; obtaining an initial predictive model comprising an initial first portion and an initial second portion; determining an initial vector by using the initial first portion based on the at least two first sample inputs; obtaining at least two second sample inputs corresponding to at least two second sample time points, respectively; predicting at least two sample parameters related to traffic conditions respectively corresponding to at least two second sample time points by using the initial second portion based on the initial vector and the at least two second sample inputs; acquiring at least two actual parameters related to the traffic condition respectively corresponding to the at least two second sample time points; determining a value of a loss function of the initial predictive model based on the at least two sample parameters related to the traffic condition and the at least two actual parameters related to the traffic condition; in response to determining that the value of the loss function is less than the loss threshold, the initial predictive model is designated as a training predictive model.

In some embodiments, in response to determining that the value of the loss function is greater than or equal to the loss threshold, the training process may further include updating the initial first portion or the initial second portion.

A second aspect of the present application relates to a method for predicting traffic parameters. The method can comprise the following steps: acquiring at least two first inputs related to the driving condition corresponding to at least two candidate time points; acquiring a first training prediction model; determining a target vector by using a first trained predictive model based on at least two first inputs; obtaining at least two second inputs relating to the driving condition corresponding to at least two future points in time; acquiring a second training prediction model; and predicting, by using the second trained predictive model, at least two future parameters related to the driving condition corresponding to at least two future points in time based on the target vector and the at least two second inputs.

A third aspect of the present application relates to a system for predicting traffic parameters. The system may include a first acquisition module, a vector determination module, a second acquisition module, and a prediction module. The first obtaining module may be configured to obtain at least two first inputs relating to the driving condition corresponding to at least two candidate time points, and obtain a first trained predictive model. The vector determination module may be configured to determine a target vector by using a first trained predictive model based on at least two first inputs. The second obtaining module may be configured to obtain at least two second inputs relating to the driving condition corresponding to at least two future points in time, and to obtain a second trained predictive model. The prediction module may be configured to predict at least two future parameters related to the driving condition corresponding to at least two future points in time by using the second trained predictive model based on the target vector and the at least two second inputs.

A fourth aspect of the present application relates to a non-transitory computer-readable medium. The non-transitory computer readable medium may include executable instructions. When the executable instructions are executed by at least one processor, the executable instructions may instruct the at least one processor to perform a method. The method can comprise the following steps: acquiring at least two first inputs related to the driving condition corresponding to at least two candidate time points; acquiring a first training prediction model; determining a target vector by using a first trained predictive model based on at least two first inputs; obtaining at least two second inputs relating to the driving condition corresponding to at least two future points in time; acquiring a second training prediction model; and predicting, by using the second trained predictive model, at least two future parameters related to the driving condition corresponding to at least two future points in time based on the target vector and the at least two second inputs.

Additional features of the present application will be set forth in part in the description which follows. Additional features of some aspects of the present application will be apparent to those of ordinary skill in the art in view of the following description and accompanying drawings, or in view of the production or operation of the embodiments. The features of the present application may be realized and attained by practice or use of the methods, instrumentalities and combinations of the various aspects of the specific embodiments described below.

Drawings

The present application will be further described by way of exemplary embodiments. These exemplary embodiments will be described in detail by means of the accompanying drawings. These embodiments are non-limiting exemplary embodiments in which like reference numerals represent similar structures throughout the several views of the drawings, and wherein:

FIG. 1 is a schematic diagram of an exemplary traffic prediction system, shown in accordance with some embodiments of the present application;

FIG. 2 is a schematic diagram of exemplary hardware and/or software components of an exemplary computing device shown in accordance with some embodiments of the present application;

FIG. 3 is a schematic diagram of exemplary hardware and/or software components of an exemplary mobile device shown in accordance with some embodiments of the present application;

FIG. 4 is a block diagram of an exemplary processing engine shown in accordance with some embodiments of the present application;

FIG. 5 is a flow chart illustrating an exemplary process for predicting future parameters related to driving conditions according to some embodiments of the present application;

FIG. 6 is a schematic diagram of an exemplary first input shown in accordance with some embodiments of the present application;

FIG. 7 is a schematic diagram of an exemplary second input shown in accordance with some embodiments of the present application;

FIG. 8 is a flow diagram of an exemplary training process for determining a predictive model, shown in accordance with some embodiments of the present application; and

FIG. 9 is a schematic diagram of an exemplary structure of a predictive model shown in accordance with some embodiments of the present application.

Detailed Description

The following description is presented to enable one of ordinary skill in the art to make and use the application and is provided in the context of a particular application and its requirements. It will be apparent to those skilled in the art that various modifications to the disclosed embodiments are possible, and that the general principles defined in this application may be applied to other embodiments and applications without departing from the spirit and scope of the application. Thus, the present application is not limited to the described embodiments, but should be accorded the widest scope consistent with the claims.

The terminology used in the description presented herein is for the purpose of describing particular example embodiments only and is not intended to limit the scope of the present application. As used herein, the singular forms "a", "an" and "the" may include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this application, specify the presence of stated features, integers, steps, operations, components, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, and/or groups thereof.

These and other features, aspects, and characteristics of the present application, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent from the following description of the present application when read in conjunction with the accompanying drawings, which are incorporated in and constitute a part of this specification. It is to be understood, however, that the drawings are designed solely for the purposes of illustration and description and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale.

Flow charts are used herein to illustrate operations performed by systems according to some embodiments of the present application. It should be understood that the operations in the flow diagrams may be performed out of order. Rather, various steps may be processed in reverse order or simultaneously. Also, one or more other operations may be added to the flowcharts. One or more operations may also be deleted from the flowchart.

Further, while the systems and methods disclosed herein are primarily directed to on-demand transport services, it should also be understood that this is merely one exemplary embodiment. The systems and methods of the present application may be applied to any other type of on-demand service. For example, the systems and methods of the present application may be applied to transportation systems in different environments, including terrestrial, marine, aerospace, and the like, or any combination thereof. The vehicles of the transportation system may include taxis, private cars, tailplanes, buses, trains, railcars, subways, ships, airplanes, airships, hot air balloons, unmanned vehicles, and the like, or any combination thereof. The transport system may also include any transport system that manages and/or distributes, for example, systems that send and/or receive couriers. Application scenarios of the system or method of the present application may include web pages, browser plug-ins, clients, client systems, internal analytics systems, artificial intelligence robots, and the like, or any combination thereof.

The terms "passenger," "requestor," "service requestor," and "customer," and the like in this application, are used interchangeably to refer to an individual, entity, or tool that can request or subscribe to a service. Likewise, the terms "driver," "provider," "service provider," "server," "service party," and the like, in this application are also used interchangeably to refer to an individual, tool, or other entity that provides a service or assists in providing a service. The term "user" in this application may refer to an individual, entity, or tool that may request a service, subscribe to a service, provide a service, or facilitate providing a service. For example, the user may be a passenger, a driver, an operator, etc., or any combination thereof. In this application, the terms "passenger" and "passenger terminal" are used interchangeably, and the terms "driver" and "driver terminal" are used interchangeably.

The terms "service," "request," and "service request" are used interchangeably herein to refer to a request that may be initiated by a passenger, a requester, a service requester, a customer, a driver, a provider, a service provider, a supplier, etc., or any combination thereof. The service request may be accepted by any of a passenger, a requester, a service requester, a customer, a driver, a provider, a service provider, or a provider. The service request may be billed or free of charge.

The positioning techniques used in the present application may be based on the Global Positioning System (GPS), global navigation satellite system (GLONASS), COMPASS navigation system (COMPASS), galileo positioning system, quasi-zenith satellite system (QZSS), wireless fidelity (WiFi) positioning techniques, and the like, or any combination thereof. One or more of the above-described positioning techniques may be used interchangeably in this application.

One aspect of the present application relates to systems and methods for predicting traffic parameters. The system may obtain at least two first inputs relating to the driving condition corresponding to at least two candidate time points, and determine a target vector by using a first trained predictive model based on the at least two first inputs. The system may also obtain at least two second inputs relating to the driving condition corresponding to at least two future points in time. Further, the system may predict at least two future parameters related to the trip condition corresponding to at least two future points in time by using the second trained predictive model based on the target vector and the at least two second inputs. According to the system and method of the present application, at least two future parameters related to the driving condition corresponding to at least two future time points can be predicted, and the at least two future parameters related to the driving condition corresponding to the at least two future time points are obtained based on at least two first inputs corresponding to at least two candidate time points and at least two second inputs corresponding to the at least two future time points, which can fuse multi-channel data and improve the efficiency and accuracy of traffic prediction.

Fig. 1 is a schematic diagram of an exemplary traffic prediction system, shown in accordance with some embodiments of the present application. The traffic prediction system may predict future parameters related to the traffic condition based on current parameters related to the traffic condition and/or historical parameters related to the traffic condition. The traffic prediction system may be applied to various application scenarios, such as an on-demand transportation service scenario, a navigation service scenario, a map service scenario, and the like. For illustrative purposes, the present application takes an on-demand transportation service scenario as an example, and accordingly, the traffic prediction system 100 may be an online transportation service platform for transportation services (e.g., taxi service, driver service, express service, carpool service, bus service, etc.). In some embodiments, the traffic prediction system 100 may include a server 110, a network 120, a requester terminal 130, a provider terminal 140, and a memory 150.

The server 110 may be a single server or a group of servers. The set of servers can be centralized or distributed (e.g., the servers 110 can be a distributed system). In some embodiments, the server 110 may be local or remote. For example, server 110 may access information and/or data stored within requester terminal 130, provider terminal 140, and/or memory 150 via network 120. As another example, server 110 may be connected to requester terminal 130, provider terminal 140, and/or memory 150 to access stored information and/or data. In some embodiments, the server 110 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof. In some embodiments, server 110 may execute on a computing device 200 described in FIG. 2 herein that includes one or more components.

In some embodiments, the server 110 may include a processing engine 112. Processing engine 112 may process information and/or data to perform one or more functions described herein. For example, the processing engine 112 may obtain at least two first inputs relating to the driving condition corresponding to at least two candidate points in time, respectively, and at least two second inputs relating to the driving condition corresponding to at least two future points in time, respectively. The processing engine 112 may further predict, based on the at least two first inputs and the at least two second inputs, at least two future parameters related to the trip condition that correspond to the at least two future points in time, respectively, using the trained predictive model. In some embodiments, processing engine 112 may include one or more processing engines (e.g., a single chip processing engine or a multi-chip processing engine). By way of example only, the processing engine 112 may include a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), an application specific instruction set processor (ASIP), an image processing unit (GPU), a physical arithmetic processing unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like, or any combination thereof.

Network 120 may facilitate the exchange of information and/or data. In some embodiments, one or more components of the traffic prediction system 100 (e.g., the server 110, the requester terminal 130, the provider terminal 140, and the memory 150) may send information and/or data to other components of the traffic prediction system 100 via the network 120. For example, the processing engine 112 may retrieve the at least two first inputs and the at least two second inputs from the memory 150 via the network 120. In some embodiments, the network 120 may be any form of wired or wireless network, or any combination thereof. By way of example only, network 120 may include a cable network, a wireline network, a fiber optic network, a telecommunications network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth network, a zigbee network, a Near Field Communication (NFC) network, the like, or any combination thereof. In some embodiments, network 120 may include one or more network access points. For example, the network 120 may include wired or wireless network access points, such as base stations and/or internet exchange points 120-1, 120-2, … …, which may be connected to the network 120 by one or more components of the traffic prediction system 100 to exchange data and/or information.

In some embodiments, the requester may be a user of requester terminal 130. In some embodiments, the user of requester terminal 130 may be a person other than the requester. For example, user a of requester terminal 130 may send a service request to user B through requester terminal 130 or receive services and/or information or instructions from server 110. In some embodiments, the provider may be a user of the provider terminal 140. In some embodiments, the user of provider terminal 140 may be a person other than the provider. For example, user C of provider terminal 140 may receive a service request for user D through provider terminal 140 and/or information or instructions from server 110. In some embodiments, "requester" and "requester terminal" are used interchangeably, and "provider" and "provider terminal" are used interchangeably.

In some embodiments, the requester terminal 130 may include a mobile device 130-1, a tablet computer 130-2, a handheld computer 130-3, a vehicle mounted device 130-4, and the like, or any combination thereof. In some embodiments, mobile device 130-1 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any thereof And (4) combining. In some embodiments, the smart home devices may include smart lighting devices, smart appliance control devices, smart monitoring devices, smart televisions, smart cameras, interphones, and the like, or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, a smart footwear, smart glasses, a smart helmet, a smart watch, a smart garment, a smart backpack, a smart accessory, or the like, or any combination thereof. In some embodiments, the smart mobile device may include a smart phone, a Personal Digital Assistant (PDA), a gaming device, a navigation device, a point of sale (POS), or the like, or any combination thereof. In some embodiments, the virtual reality device and/or the augmented reality device may include a virtual reality helmet, virtual reality glasses, virtual reality eyeshields, augmented reality helmets, augmented reality glasses, augmented reality eyeshields, and the like, or any combination thereof. For example, the virtual reality device and/or augmented reality device may include a Google Glass^TM、RiftCon^TM、Fragments^TM、Gear VR^TMAnd the like. In some embodiments, the in-vehicle device 130-4 may include an in-vehicle computer, an in-vehicle television, or the like. In some embodiments, requester terminal 130 may be a device having location technology for determining a location of the requester and/or requester terminal 130.

In some embodiments, provider terminal 140 may be a similar or the same device as requester terminal 130. In some embodiments, provider terminal 140 may be a device with location technology for locating the location of the provider and/or provider terminal 140. In some embodiments, provider terminal 140 may periodically send GPS information to server 110. In some embodiments, requester terminal 130 and/or provider terminal 140 may communicate with another locating device to determine the location of the requester, requester terminal 130, provider, and/or provider terminal 140. In some embodiments, the requester terminal 130 and/or the provider terminal 140 may transmit location information to the server 110.

Memory 150 may store data and/or instructions. In some embodiments, memory 150 may store data obtained from requester terminal 130 and/or provider terminal 140. In some embodiments, memory 150 may store data and/or instructions used by server 110 to perform or use to perform the exemplary methods described in this application. In some embodiments, memory 150 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), and the like, or any combination thereof. Exemplary mass storage devices may include magnetic disks, optical disks, solid state disks, and the like. Exemplary removable memories may include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tape, and the like. Exemplary volatile read and write memory can include Random Access Memory (RAM). Exemplary RAM may include Dynamic Random Access Memory (DRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), Static Random Access Memory (SRAM), thyristor random access memory (T-RAM), and zero capacitance random access memory (Z-RAM), among others. Exemplary read-only memories may include mask read-only memory (MROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), digital versatile disc read-only memory (dvd-ROM), and the like. In some embodiments, the memory 150 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof.

In some embodiments, the memory 150 may be connected to the network 120 to communicate with one or more components of the traffic prediction system 100 (e.g., the server 110, the requester terminal 130, the provider terminal 140). One or more components of the traffic prediction system 100 may access data or instructions stored in the memory 150 via the network 120. In some embodiments, the memory 150 may be directly connected to or in communication with one or more components of the traffic prediction system 100 (e.g., the server 110, the requester terminal 130, the provider terminal 140). In some embodiments, the memory 150 may be part of the server 110.

In some embodiments, one or more components of the traffic prediction system 100 (e.g., the server 110, the requester terminal 130, the provider terminal 140) may access the memory 150. In some embodiments, one or more components of the traffic prediction system 100 may read and/or modify information related to the requester, the provider, and/or the public when one or more conditions are satisfied. For example, after a service is completed, server 110 may read and/or modify information for one or more users. For another example, when the provider terminal 140 receives a service request from the requester terminal 130, the provider terminal 140 may access information related to the requester but cannot modify the information related to the requester.

In some embodiments, the exchange of information by one or more components of the traffic prediction system 100 may be accomplished by way of a request for service. The object of the service request may be any product. In some embodiments, the product may be a tangible product or an intangible product. Tangible products may include food, pharmaceuticals, commodities, chemical products, appliances, clothing, automobiles, homes, luxury goods, and the like, or any combination thereof. Intangible products may include service products, financial products, knowledge products, internet products, and the like, or any combination thereof. The internet products may include personal host products, website products, mobile internet products, commercial host products, embedded products, and the like, or any combination thereof. The mobile internet product may be used for software, programs, systems, etc. of the mobile terminal or any combination thereof. The mobile terminal may include a tablet computer, a handheld computer, a mobile phone, a Personal Digital Assistant (PDA), a smart watch, a POS device, an in-vehicle computer, a vehicle television, a wearable device, and the like, or any combination thereof. For example, the product may be any software and/or application used on a computer or mobile phone. The software and/or applications may be related to social interaction, shopping, transportation, entertainment, learning, investment, etc., or any combination thereof. In some embodiments, transportation-related system software and/or applications may include travel software and/or applications, vehicle scheduling software and/or applications, mapping software and/or applications, and/or the like. In vehicle scheduling software and/or applications, a vehicle may include a horse, a carriage, a human powered vehicle (e.g., unicycle, bicycle, tricycle), an automobile (e.g., taxi, bus, private car), and the like or any combination thereof.

Those of ordinary skill in the art will appreciate that when an element of the roadway identification system 100 executes, the element may execute via electrical and/or electromagnetic signals. For example, when the requester terminal 130 processes a task, such as making a decision, identifying or selecting an object, the requester terminal 130 may operate logic circuits in its processor to process such task. When the requester terminal 130 issues a service request to the server 110, the processor of the service requester terminal 130 may generate an electrical signal encoding the service request. The processor of the requester terminal 130 may then send the electrical signal to an output port. If the requester terminal 130 communicates with the server 110 via a wired network, the output port may be physically connected to a cable, which may also send electrical signals to the input port of the server 110. If the requester terminal 130 communicates with the server 110 via a wireless network, the output port of the requester terminal 130 may be one or more antennas that may convert electrical signals to electromagnetic signals. Similarly, provider terminal 140 may process tasks by operation of logic circuits in its processor and receive instructions and/or service requests from server 110 via electrical or electromagnetic signals. In an electronic device (e.g., requester terminal 130, provider terminal 140, and/or server 110) when its processor processes instructions, issues instructions, and/or performs actions, the instructions and/or actions are performed via electrical signals. For example, when the processor retrieves or stores data from a storage medium (e.g., memory 150), it may send electrical signals to a read/write device of the storage medium, which may read or write structured data in the storage medium. The structured data may be transmitted to the processor in the form of electrical signals via a bus of the electronic device. Herein, an electrical signal may refer to one electrical signal, a series of electrical signals, and/or at least two discrete electrical signals.

It should be noted that the application scenario shown in fig. 1 is provided for illustrative purposes only and is not intended to limit the scope of the present application. Various changes and modifications will occur to those skilled in the art based on the description herein. For example, the traffic prediction system 100 may be used as a navigation system. The navigation system may include a user terminal (e.g., provider terminal 140) and a server (e.g., server 110). The navigation system may provide a navigation service to a user when the user plans to drive the vehicle to a destination, and during the navigation service, the navigation system may periodically acquire GPS information of the vehicle from a GPS device integrated in a user terminal. The navigation system may acquire GPS information related to at least two vehicles and determine traffic information based on the GPS information. Further, the navigation system may predict future traffic information based on current traffic information and/or historical traffic information according to the processes and/or methods described herein.

FIG. 2 is a schematic diagram of exemplary hardware and/or software components of an exemplary computing device shown in accordance with some embodiments of the present application. In some embodiments, the server 110, the requester terminal 130, and/or the provider terminal 140 may be implemented on the computing device 200. For example, the processing engine 112 may be implemented on the computing device 200 and perform the functions of the processing engine 112 disclosed herein.

The computing device 200 may be used to implement any of the components of the traffic prediction system 100 as described herein. For example, the processing engine 112 may be implemented on the computing device 200 by its hardware, software programs, firmware, or a combination thereof. For convenience, while only one such computer is shown, the computer functions described herein may be implemented in a distributed fashion across multiple similar platforms to spread the processing load.

For example, computing device 200 may include a communication port 250 to connect to a network to enable data communication. Computing device 200 may also include a processor (e.g., processor 220) in the form of one or more processors (e.g., logic circuits) for executing program instructions. For example, a processor includes, among other things, interface circuitry and processing circuitry. Interface circuitry may be configured to receive electrical signals from bus 210, where the electrical signals encode structured data and/or instructions for processing by the processing circuitry. The processing circuitry may perform logical computations and then determine the conclusion, result, and/or instruction encoding as electrical signals. The interface circuit may then send the electrical signals from the processing circuit via bus 210.

Computing device 200 may also include different forms of program storage and data storage, including, for example, a disk 270, Read Only Memory (ROM)230, or Random Access Memory (RAM)240 for storing various data files processed and/or transmitted by computing device 200. The exemplary computing device 200 may also include program instructions stored in the ROM 230, RAM 240, and/or other forms of non-transitory storage that are executed by the processor 220. The methods and/or processes of the present application may be embodied in the form of program instructions. Computing device 200 also includes I/O components 260 that support input/output between the computer and other components. Computing device 200 may also receive programming and data via network communications.

For illustration only, only one CPU and/or processor is shown in FIG. 2. It is also contemplated that multiple CPUs and/or processors may be used; thus, operations and/or method steps described herein as being performed by one CPU and/or processor may also be performed by multiple CPUs and/or processors, either jointly or separately. For example, if in the present application the CPUs and/or processors of computing device 200 perform steps a and B, it should be understood that steps a and B may also be performed by two different CPUs and/or processors of computing device 200, either collectively or independently (e.g., a first processor performing step a, a second processor performing step B, or a first and second processor collectively performing steps a and B).

Fig. 3 is a schematic diagram of exemplary hardware and/or software components of an exemplary mobile device shown in accordance with some embodiments of the present application. In some embodiments, the requester terminal 130 and/or the provider terminal 140 may be implemented on a mobile device 300.

As shown in FIG. 3, mobile device 300 may include a communication platform 310, a display 320, a Graphics Processing Unit (GPU)330, a Central Processing Unit (CPU)340, I/O350, memory 360, and storage 390. In some embodiments, any other suitable component, including but not limited to a system bus or a controller (not shown), may also be included in mobile device 300. In some embodiments, the operating system 370 is mobile (e.g., iOS) ^TM、Android^TM、Windows Phone^TM) And one or more application programs 380 may be loaded from storage 390 into memory 360 for execution by CPU 340. The application 380 may include a browser or any other suitable mobile application for receiving and presenting information related to the traffic prediction system 100. User interaction with the information flow may be accomplished through I/O350 and provided to one or more components of traffic prediction system 100 via network 120.

To implement the various modules, units, and their functions described herein, a computer hardware platform may be used as the hardware platform for one or more of the components described herein. A computer with user interface components may be used to implement a Personal Computer (PC) or any other type of workstation or terminal device. If programmed properly, the computer may also act as a server.

FIG. 4 is a block diagram of an exemplary processing engine shown in accordance with some embodiments of the present application. The processing engine 112 may include a first acquisition module 410, a vector determination module 420, a second acquisition module 430, a prediction module 440, and a training module 450.

The first obtaining module 410 may be configured to obtain at least two first inputs related to the driving condition corresponding to at least two candidate time points, respectively. As used herein, a driving condition may relate to a road segment through which a vehicle may pass. In some embodiments, as shown in fig. 6, each of the at least two first inputs may include a first parameter related to traffic conditions at the corresponding candidate time point, at least two first historical parameters related to traffic conditions respectively corresponding to the at least two first historical time points, a first statistical parameter related to the at least two first historical parameters, and the like. In some embodiments, the first obtaining module 410 may also be configured to obtain a first trained predictive model. The first trained predictive model may be configured to extract feature information of the at least two first inputs and fuse the at least two first inputs based on the feature information.

The vector determination module 420 may be configured to determine a target vector (which may also be referred to as a "state vector") based on at least two first inputs by using a first trained predictive model. As used herein, a target vector may be an expression indicative of a fused result of at least two first inputs, including a relationship between any two of the at least two first inputs.

The second obtaining module 430 may be configured to obtain at least two second inputs related to the driving condition corresponding to at least two future points in time, respectively. In some embodiments, as shown in fig. 7, each of the at least two second inputs may include a second parameter related to traffic conditions at a previous time point of the corresponding future time point (which may be a predicted future parameter of the previous time point as described in operation 560; for the first future time point, the previous time point refers to the current time point), at least two second historical parameters related to traffic conditions respectively corresponding to the at least two second historical time points, a second statistical parameter related to the at least two second historical parameters, and so on. In some embodiments, the second acquisition module 430 may also be configured to acquire a second trained predictive model, which may be configured to predict future traffic information.

The prediction module 440 may be configured to predict at least two future parameters related to the trip condition corresponding to at least two future points in time, respectively, by using the second trained predictive model based on the target vector and the at least two second inputs.

The training module 440 may be configured to determine the first training predictive model and/or the second training predictive model. More description of the model training process may be found elsewhere in this application (e.g., FIG. 8 and its description).

The modules in the processing engine 112 may be connected or in communication with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, etc., or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. Two or more modules may be combined into a single module and any one module may be split into two or more units.

For example, the first acquisition module 410 and the second acquisition module 430 may be combined into a single module, which may be configured to acquire at least two first inputs, a first trained predictive model, at least two second inputs, and a second trained predictive model. As another example, the training module 450 may be optional, and the first trained predictive model and/or the second trained predictive model may be retrieved from a storage device (e.g., the memory 150) disclosed elsewhere in this application. As another example, the processing engine 112 may include a storage module (not shown) that may be configured to store at least two first inputs, a first trained predictive model, at least two second inputs, a second trained predictive model, a target vector, at least two future parameters, and/or the like.

FIG. 5 is a flow diagram illustrating an exemplary process for predicting future parameters related to traffic conditions according to some embodiments of the present application. The process 500 may be performed by the traffic prediction system 100. For example, process 500 may be implemented as a set of instructions stored in memory ROM 230 or RAM 240. Processor 220 and/or the modules in fig. 4 may execute the set of instructions and, when executing the instructions, processor 220 and/or the modules may be configured to perform process 500. The operation of the process shown below is for illustration purposes only. In some embodiments, process 500 may utilize one or more additional operations not described, and/or be accomplished without one or more of the operations discussed. Additionally, the order in which the operations of process 500 are illustrated in FIG. 5 and described below is not limiting.

At 510, the processing engine 112 (e.g., the first obtaining module 410) (e.g., the interface circuit of the processor 220) may obtain at least two first inputs related to the driving condition corresponding to at least two candidate time points, respectively. As used herein, a driving condition may relate to a road segment through which a vehicle may pass.

In some embodiments, the at least two candidate points in time comprise a current point in time. In some embodiments, the number of at least two candidate time points and/or the time interval between any adjacent two of the at least two candidate time points may be a default setting of the traffic prediction system 100 or may be adjustable under different circumstances. In some embodiments, at least two candidate time points may be arranged chronologically. For example, assuming that the current time point is "10 am", the number of at least two candidate time points is 10, and the time interval between any adjacent two of the at least two candidate time points is 1 minute, the at least two candidate time points may be represented as the following set:

T₁＝{9:51,9:52,9:53,9:54,9:55,9:56,9:57,9:58,9:59,10:00} (1)

Accordingly, the at least two first inputs respectively corresponding to the at least two candidate time points may be represented as follows:

F₁＝{f_t|t∈{1,2....n}}, (2)

wherein, F₁The representation comprises a set of at least two first inputs, f_tRepresents a tth first input corresponding to a tth candidate time point, and n represents the number of at least two candidate time points (i.e., the number of at least two first inputs).

In some embodiments, as shown in fig. 6, each of the at least two first inputs may include a first parameter related to traffic conditions at the corresponding candidate time point, at least two first historical parameters related to traffic conditions respectively corresponding to the at least two first historical time points, a first statistical parameter related to the at least two first historical parameters, and the like.

In some embodiments, each of the at least two first historical points in time may correspond to a candidate point in time. For example, assuming that the particular candidate time point is "10 am on weekdays," the at least two first historical time points may be at least two corresponding historical time points (i.e., 10:00 am on weekdays) within a predetermined time period (e.g., last week, last month, last three months).

In some embodiments, the first parameter related to traffic conditions may include a traffic congestion level of the travel conditions, a traffic speed of the travel conditions, a traffic flow (which may be represented by a number of vehicles) of the travel conditions, and the like, or any combination thereof. As used herein, a traffic congestion level of a driving condition may refer to at least two traffic congestion levels for at least two locations within a road segment, an average traffic congestion level of the at least two traffic congestion levels, a sum of the at least two traffic congestion levels, and the like. The traffic speed of the driving condition may refer to at least two traffic speeds of at least two locations within the section of road, an average traffic speed of the at least two traffic speeds, a sum of the at least two traffic speeds, and the like. The traffic flow of the driving condition may refer to at least two traffic flows of at least two locations within the link, an average traffic flow of the at least two traffic flows, a sum of the at least two traffic flows, and the like.

In some embodiments, the traffic congestion level of the driving condition may be represented as at least two levels based on the traffic flow of the driving condition, for example, "heavy congestion", "normal congestion", "light congestion", "clear traffic" are shown in table 1 below.

TABLE 1 exemplary Congestion levels

Congestion level	Flow of traffic	Rank value
			Severe blockage	F＜a	4
Normal blockage	A≤F＜b	3
			Light congestion	B≤F＜c	2
Traffic is unobstructed	F≥c	1

As shown in table 1, each of the parameters "a", "b", and "c" refers to a traffic flow threshold value, and F refers to a traffic flow of a specific location point within a link. The traffic flow threshold may be a default setting for the traffic prediction system 100 or may be adjustable under different circumstances (e.g., the traffic flow threshold may be different for different cities).

In some embodiments, as described above, each of the at least two first historical parameters related to traffic conditions may include a historical traffic congestion level for the travel conditions, a historical traffic speed for the travel conditions, a historical traffic flow for the travel conditions, and the like, or any combination thereof. In some embodiments, for at least two first historical parameters related to traffic conditions, the processing engine 112 may determine a first aggregate historical parameter based on the at least two first historical parameters. For example, the processing engine 112 may determine a sum or a weighted sum of at least two first historical parameters as the first synthesized historical parameter, wherein the closer the first historical time point is to the current time point, the greater the weight of the first historical parameter corresponding to the first historical time point may be.

In some embodiments, the first statistical parameter may include a traffic congestion statistical parameter, a traffic speed statistical parameter, a traffic flow statistical parameter, or the like, or any combination thereof. The traffic congestion statistical parameters may include a mode of the at least two historical traffic congestion levels, a congestion probability of the at least two historical traffic congestion levels, and the like. The traffic speed statistical parameter may include a mean of the at least two historical speeds, a median of the at least two historical speeds, a variance of the at least two historical speeds, a maximum of the at least two historical speeds, a minimum of the at least two historical speeds, and the like. The traffic flow statistical parameters may include a mean of the at least two historical traffic flows, a median of the at least two historical traffic flows, a variance of the at least two historical traffic flows, a maximum of the at least two historical traffic flows, a minimum of the at least two historical traffic flows, and the like.

As used herein, the congestion probability of at least two historical traffic congestion levels refers to a ratio of a particular historical traffic congestion level of the at least two historical traffic congestion levels. For example, assume that at least two historical traffic congestion levels are shown in table 2 below.

TABLE 2 exemplary historical traffic congestion levels

Historical time points	Congestion level	Rank value
				1	Light congestion	2
2	Light congestion	2
			3	Light congestion	2
4	Traffic is unobstructed	1
			5	Light congestion	2
6	Light congestion	2
			7	Traffic is unobstructed	1
8	Light congestion	2
			9	Light congestion	2
10	Traffic is unobstructed	1

As shown in table 2, it can be seen that the ratio of "light congestion" is 0.7 and the ratio of "clear traffic" is 0.3. Therefore, the congestion probability of "light congestion" is 0.7 and the congestion probability of "clear traffic" is 0.3.

In 520, the processing engine 112 (e.g., the first obtaining module 410) (e.g., the processing circuitry of the processor 220) may obtain a first trained predictive model. The first trained predictive model may be configured to extract feature information of the at least two first inputs and fuse the at least two first inputs based on the feature information. The processing engine 112 may retrieve the first trained predictive model from a storage device (e.g., memory 150) disclosed elsewhere in this application.

In some embodiments, the first training model may be a Recurrent Neural Network (RNN) model, a Long Short Term Memory (LSTM) model, a gated recurrent unit model (GRU), or the like. In some embodiments, the first trained predictive model may be a first part of a trained predictive model (e.g., a sequence to sequence model). The sequence-to-sequence model may include one or more RNN units, one or more LSTM units, one or more GRU units, etc. More description on training the predictive model may be found elsewhere in this application (e.g., fig. 9 and its description).

At 530, the processing engine 112 (e.g., the vector determination module 420) (e.g., the processing circuitry of the processor 220) may determine a target vector using the first trained predictive model based on the at least two first inputs. As used herein, a target vector may be an expression indicative of a fused result of at least two first inputs, including a relationship between any two of the at least two first inputs.

In some embodiments, the at least two first inputs may be input into the first training model in a temporal order, and the intermediate result corresponding to the previous candidate time point may be used as part of the input corresponding to the next adjacent candidate time point (the intermediate result may be assigned a weight), thereby extracting a context dependency between the at least two first inputs. More description may be found elsewhere in this application (e.g., fig. 9 and its description).

At 540, the processing engine 112 (e.g., the second obtaining module 430) (e.g., the interface circuit of the processor 220) may obtain at least two second inputs related to the driving condition corresponding to at least two future points in time, respectively.

In some embodiments, similar to the at least two candidate points in time, the number of the at least two future points in time and/or the time interval between any adjacent two of the at least two future points in time may be a default setting of the traffic prediction system 100 or may be adjustable under different circumstances. In some embodiments, the at least two future points in time may be arranged in chronological order. For example, assuming that the first future point in time is "10: 01 am", the number count of the at least two future points in time is 10, and the time interval between any adjacent two of the at least two future points in time is 1 minute, the at least two future points in time may be represented as the following set:

T₂＝{10:01,10:02,10:03,10:04,10:05,10:06,10:07,10:08,10:09,10:10} (3)

Accordingly, the at least two second inputs respectively corresponding to the at least two future points in time may be represented as follows:

F₂＝{f_t+i|t∈{1,2....n},i∈{1,2....m}} (4)

wherein, F₂The representation comprises a set of at least two second inputs, f_t+iRepresents the ith second input (which corresponds to the (t + i) th point in time), and m represents the number of at least two future points in time (i.e., the number of at least two second inputs).

In some embodiments, as shown in fig. 7, each of the at least two second inputs may include a second parameter related to traffic conditions at a previous time point of the corresponding future time point (which may be a predicted future parameter of the previous time point as described in operation 560; for the first future time point, the previous time point refers to the current time point), at least two second historical parameters related to traffic conditions respectively corresponding to the at least two second historical time points, a second statistical parameter related to the at least two second historical parameters, and so on. In some embodiments, each of the at least two second historical points in time may correspond to a future point in time, similar to the at least two first historical points in time. For example, assuming that the specific future point in time is "10: 02 am on weekdays", the at least two second historical points in time may be at least two corresponding historical points in time (i.e., 10:02 am on weekdays) within a predetermined time period (e.g., last week, last month, last three months).

In some embodiments, similar to the first parameter related to traffic conditions, the second parameter related to traffic conditions may include a traffic congestion level of the driving conditions, a traffic speed of the driving conditions, a traffic flow of the driving conditions, or the like, or any combination thereof.

In some embodiments, as described above, each of the at least two second historical parameters related to traffic conditions may include a historical traffic congestion level for the travel conditions, a historical traffic speed for the travel conditions, a historical traffic flow for the travel conditions, and the like, or any combination thereof. In some embodiments, also similar to the at least two first historical parameters related to traffic conditions, for the at least two second historical parameters related to traffic conditions, the processing engine 112 may determine a second aggregated historical parameter based on the at least two second historical parameters. For example, the processing engine 112 may determine a sum or a weighted sum of at least two second historical parameters as the second integrated historical parameter, wherein the closer the second historical time point is to the current time point, the greater the weight of the second historical parameter corresponding to the second historical time point may be.

In some embodiments, also similar to the first statistical parameter, the second statistical parameter may include a traffic congestion statistical parameter, a traffic speed statistical parameter, a traffic flow statistical parameter, or the like, or any combination thereof.

In some embodiments, each of the at least two second inputs may further include reference parameters (e.g., weather forecast information) for a respective future point in time.

At 550, the processing engine 112 (e.g., the second obtaining module 430) (e.g., interface circuitry of the processor 220) may obtain a second trained predictive model, which may be configured to predict future traffic information. The processing engine 112 may retrieve the second trained predictive model from a storage device (e.g., memory 150) disclosed elsewhere in this application.

In some embodiments, the second trained predictive model may be a Recurrent Neural Network (RNN) model, a Long Short Term Memory (LSTM) model, a gated recursive unit model (GRU), or the like. In some embodiments, the second trained predictive model may be a second part of the trained predictive model (e.g., a sequence to sequence model). The sequence-to-sequence model may include one or more RNN units, one or more LSTM units, one or more GRU units, etc. More description of the trained predictive model may be found elsewhere in this application (e.g., fig. 9 and its description).

At 560, the processing engine 112 (e.g., the prediction module 440) (e.g., the processing circuitry of the processor 220) may predict, based on the target vector and the at least two second inputs, at least two future parameters related to the trip condition that correspond to the at least two future points in time, respectively, using the second trained predictive model.

In some embodiments, the at least two second inputs may be input into the second training model in a time sequence, and the intermediate result corresponding to the previous future point in time may be used as part of the input corresponding to the next adjacent future point in time (the intermediate result may be assigned a weight), thereby extracting a context dependency between the at least two second inputs. More description may be found elsewhere in this application (e.g., fig. 9 and its description).

As described in connection with operation 540, the at least two future parameters related to the driving condition respectively corresponding to the at least two future points in time may be expressed as follows:

P＝{p_t+i|t∈{1,2....n},i∈{1,2....m}} (5)

wherein the P representation comprises a set of at least two future parameters and P_t+iRepresenting the ith future parameter (which corresponds to the (t + i) th point in time).

In some embodiments, after predicting the at least two future parameters related to the driving condition, the processing engine 112 may transmit information (e.g., travel prompts) indicating the at least two future parameters to the requester terminal 130 and/or the provider terminal 140. In some embodiments, processing engine 112 may predetermine a scheduling policy based on the predicted future parameters. In some embodiments, the processing engine 112 may use at least two future parameters as a reference in estimating service information (e.g., recommended routes, Estimated Time of Arrival (ETA)) related to the service request.

It should be noted that the foregoing is provided for illustrative purposes only and is not intended to limit the scope of the present application. Many variations and modifications may be made to the teachings of the present application by those of ordinary skill in the art in light of the present disclosure. However, variations and modifications may be made without departing from the scope of the present application. For example, one or more other optional operations (e.g., a store operation) may be added elsewhere in process 500. In a storage operation, the processing engine 112 may store information and/or data (e.g., at least two first inputs, a target vector, at least two second inputs, at least two predicted future parameters) related to the traffic prediction system 100 in a storage device (e.g., the memory 150) disclosed elsewhere in this application.

FIG. 8 is a flow diagram of an exemplary training process for determining a predictive model, shown in accordance with some embodiments of the present application. The process 800 may be performed by the traffic prediction system 100. For example, process 800 may be implemented as a set of instructions stored in memory ROM230 or RAM 240. Processor 220 and/or the modules in fig. 4 may execute the set of instructions and, when executing the instructions, processor 220 and/or the modules may be configured to perform process 800. The operation of the process shown below is for illustration purposes only. In some embodiments, process 800 may utilize one or more additional operations not described, and/or be accomplished without one or more of the operations discussed. Additionally, the order in which the operations of process 800 are illustrated in FIG. 8 and described below is not limiting.

At 810, processing engine 112 (e.g., training module 450) (e.g., processing circuitry of processor 220) may obtain at least two first sample inputs corresponding to at least two first sample points in time, respectively. The processing engine 112 may retrieve at least two first sample inputs from the memory 150 via the network 120. As described in connection with operation 510, each of the at least two first sample inputs may include a first sample parameter associated with the traffic condition at the respective sample time point, at least two first sample historical parameters associated with the traffic condition corresponding to the at least two first sample historical time points, respectively, a first sample statistical parameter associated with the at least two first historical parameters, and/or the like.

At 820, the processing engine 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may obtain an initial predictive model that includes an initial first portion and an initial second portion. In some embodiments, the initial first portion and/or the initial second portion may include one or more RNN units, one or more LSTM units, one or more GRU units, or the like. In some embodiments, the initial prediction model may include one or more initial model parameters, such as the number of cells in the initial first portion, the number of cells in the initial second portion, the size of each cell, the number of layers in each cell, the number of gates per cell, a weight parameter between any two adjacent cells (which may be used to assign a weight to an intermediate result corresponding to a first of the two adjacent cells) (also referred to as a "weight parameter between any two adjacent points in time"), and so forth.

At 830, processing engine 112 (e.g., training module 450) (e.g., processing circuitry of processor 220) may determine an initial vector based on the at least two first sample inputs using the initial first portion. As described in connection with operation 530, the initial vector may be an expression indicating a fused result of the at least two first sample inputs, including a relationship between any two of the at least two first sample inputs.

In 840, processing engine 112 (e.g., training module 450) (e.g., processing circuitry of processor 220) may obtain at least two second sample inputs corresponding to at least two second sample time points, respectively. The processing engine 112 may retrieve at least two second sample inputs from the memory 150 via the network 120. As described in connection with operation 540, the at least two second sample time points refer to future time points with respect to the at least two first sample time points, and each of the at least two second sample inputs may include a second sample parameter related to traffic conditions at a previous time point of the corresponding second sample time point, at least two second sample history parameters related to traffic conditions respectively corresponding to the at least two second sample history time points, a second sample statistical parameter related to the at least two second sample history traffic parameters, and the like.

In 850, the processing engine 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may predict at least two sample parameters related to traffic conditions corresponding to at least two second sample time points, respectively, by using the initial second portion based on the initial vector and the at least two second sample inputs.

At 860, the processing engine 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may obtain at least two actual parameters related to traffic conditions corresponding to at least two second sample time points, respectively. The processing engine 112 may retrieve at least two actual parameters related to traffic conditions from the memory 150 via the network 120.

In 870, the processing engine 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may determine a value of a loss function of the initial predictive model based on the at least two sample parameters and the at least two actual parameters. In some embodiments, the loss function may be a Root Mean Square Error (RMSE).

At 880, processing engine 112 (e.g., training module 450) (e.g., processing circuitry of processor 220) may determine whether a value of a loss function of the initial predictive model is less than a loss threshold. In response to determining that the value of the loss function of the initial predictive model is less than the loss threshold, the processing engine 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may designate the initial predictive model as a training predictive model in 890, meaning that the training process is complete, and the training predictive model may be stored in a storage device (e.g., the memory 150) disclosed elsewhere in this application.

In response to determining that the value of the loss function of the initial predictive model is greater than or equal to the loss threshold, the processing engine 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may perform the process 800 to return to operation 820 to update the initial predictive model (e.g., update the initial first portion and/or the initial second portion). In some embodiments, processing engine 112 may update one or more initial model parameters.

Further, the processing engine 112 may determine whether the value of the loss function of the updated predictive model is less than a loss threshold. In response to determining that the value of the loss function of the updated predictive model is less than the loss threshold, the processing engine 112 may designate the updated predictive model as the training predictive model. On the other hand, in response to determining that the value of the loss function of the updated predictive model is greater than or equal to the loss threshold, the processing engine 112 may still perform process 800 to return to operation 820 to update the updated predictive model until the value of the loss function of the updated predictive model is less than the loss threshold.

It will be apparent to one of ordinary skill in the art that the operation of the training process is substantially similar to the operation of the practice process, and thus, some details have been omitted from process 800, which may be found elsewhere in the present application (e.g., process 500 and descriptions thereof).

It should be noted that the foregoing is provided for illustrative purposes only and is not intended to limit the scope of the present application. Various changes and modifications will occur to those skilled in the art based on the description herein. However, such changes and modifications do not depart from the scope of the present application. For example, the training module 450 may update the training predictive model at certain time intervals (e.g., monthly, every two months) based on at least two newly acquired samples. For another example, in addition to the value of the loss function, the processing engine 112 may use other conditions (e.g., number of iterations, accuracy rate) to determine whether the training process is complete.

FIG. 9 is a schematic diagram of an exemplary structure of a predictive model shown in accordance with some embodiments of the present application. For illustration purposes, the prediction model may be a sequence-to-sequence model, including an encoder and a decoder.

As shown, the encoder may include at least two GRU units. Corresponding to at least two candidate time points (e.g., t)₁、t₂、...、t_nWherein t is_nMay refer to a current time point) of at least two first inputs (e.g., X) related to a driving condition₁、X₂、...、X_n) May be input into at least two GRU units, respectively. It can be seen that the intermediate result corresponding to the previous candidate point in time can be used as part of the input corresponding to the next adjacent candidate point in time (the intermediate result can be assigned a weight). Further, a target vector may be generated based on the at least two first inputs.

Further, the decoder may include at least two GRU units, and the target vector may be input into a first GRU of the decoder. Further, toShould correspond to at least two future points in time (e.g., t)_n+1、t_n+2、...、t_n+m) At least two second inputs (e.g. Z) related to the driving condition₁、Z₂、...、Z_n) May be input to at least two GRU units, respectively. It can be seen that the intermediate result corresponding to the previous future point in time can be used as part of the input corresponding to the next adjacent future point in time (the intermediate result can be assigned a weight). Further, at least two future parameters (e.g., Y) related to the driving condition may be predicted₁、Y₂、...、Y_n)。

In some embodiments, the sequence-to-sequence model may include only one GRU unit, which may be shared by at least two inputs (i.e., at least two first inputs and at least two second inputs). In this case, the at least two first inputs and the at least two second inputs may be chronologically arranged and may be sequentially input to the GRU unit, and accordingly, the at least two future parameters related to the running condition may be sequentially predicted.

Having thus described the basic concepts, it will be apparent to those of ordinary skill in the art having read this application that the foregoing disclosure is to be construed as illustrative only and is not limiting of the application. Various modifications, improvements and adaptations of the present application may occur to those skilled in the art, although they are not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific language to describe embodiments of the application. For example, "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the application may be combined as appropriate.

Moreover, those of ordinary skill in the art will understand that aspects of the present application may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, articles, or materials, or any new and useful modification thereof. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as "block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may take the form of a computer program product embodied in one or more computer-readable media, with computer-readable program code embodied therein.

A computer readable signal medium may comprise a propagated data signal with computer program code embodied therewith, for example, on baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including electro-magnetic, optical, and the like, or any suitable combination. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable signal medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, etc., or any combination of the preceding.

Computer program code required for the operation of various portions of the present application may be written in any one or more programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, a conventional programming language such as C programming language, Visual Basic, Fortran2003, Perl, COBOL 2002, PHP, ABAP, a dynamic programming language such as Python, Ruby, and Groovy, or other programming languages, and the like. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the present application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the embodiments. This method of application, however, is not to be interpreted as reflecting an intention that the claimed subject matter to be scanned requires more features than are expressly recited in each claim. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Claims

1. A method for predicting traffic parameters, the method comprising:

acquiring at least two first inputs related to the driving condition corresponding to at least two candidate time points;

acquiring a first training prediction model;

determining a target vector by using the first trained predictive model based on the at least two first inputs;

obtaining at least two second inputs relating to the driving condition corresponding to at least two future points in time;

acquiring a second training prediction model; and

predicting, by the second trained predictive model, at least two future parameters related to the driving condition corresponding to the at least two future points in time based on the target vector and the at least two second inputs.

2. The method of claim 1, wherein the driving condition is associated with a road segment.

3. The method according to claim 1 or 2, wherein each of the at least two first inputs comprises a first parameter related to traffic conditions for the respective candidate point in time, at least two first historical parameters related to traffic conditions corresponding to at least two first historical points in time, respectively, and a first statistical parameter related to the at least two first historical parameters.

4. The method of claim 3, wherein the first parameter related to traffic conditions comprises at least one of a traffic congestion level of the driving conditions, a traffic speed of the driving conditions, or a traffic flow of the driving conditions.

5. The method of claim 3, wherein the first statistical parameter comprises a traffic congestion statistical parameter, a traffic speed statistical parameter, or a traffic flow statistical parameter.

6. The method of claim 5, wherein the first statistical parameter comprises the traffic congestion statistical parameter, and wherein the traffic congestion statistical parameter comprises at least one of a mode of at least two historical traffic congestion levels or a congestion probability of the at least two historical traffic congestion levels.

7. The method of claim 5, wherein the first statistical parameter comprises the traffic speed statistical parameter, and the traffic speed statistical parameter comprises at least one of a mean of at least two historical speeds, a median of the at least two historical speeds, a variance of the at least two historical speeds, a maximum of the at least two historical speeds, or a minimum of the at least two historical speeds.

8. The method of claim 5, wherein the first statistical parameter comprises the traffic flow statistical parameter, and the traffic flow statistical parameter comprises at least one of a mean of at least two historical traffic flows, a median of the at least two historical traffic flows, a variance of the at least two historical traffic flows, a maximum of the at least two historical traffic flows, or a minimum of the at least two historical traffic flows.

9. The method according to claim 1 or 2, characterized in that each of the at least two second inputs comprises a second parameter related to traffic conditions at a previous point in time of the respective future point in time, at least two second historical parameters related to traffic conditions corresponding to at least two second historical points in time, respectively, and a second statistical parameter related to the at least two second historical traffic parameters.

10. The method of claim 9, wherein each of the at least two second inputs further comprises a reference parameter comprising weather information for the respective future point in time.

11. The method of claim 1, wherein the first trained predictive model is a first part of a trained predictive model and the second trained predictive model is a second part of the trained predictive model.

12. The method of claim 11,

the trained predictive model is a sequence-to-sequence model,

the first part of the training of the prediction model is an encoder, and

the second part of the training of the prediction model is the decoder.

13. The method of claim 11 or 12, wherein the training prediction model is determined based on a training process comprising:

obtaining at least two first sample inputs corresponding to at least two first sample time points, respectively;

obtaining an initial predictive model comprising an initial first portion and an initial second portion;

determining an initial vector by using the initial first portion based on the at least two first sample inputs;

obtaining at least two second sample inputs corresponding to at least two second sample time points, respectively;

predicting at least two sample parameters related to traffic conditions respectively corresponding to the at least two second sample time points by using the initial second portion based on the initial vector and the at least two second sample inputs;

acquiring at least two actual parameters related to traffic conditions respectively corresponding to the at least two second sample time points;

Determining a value of a loss function of the initial predictive model based on the at least two traffic condition related sample parameters and the at least two traffic condition related actual parameters; and

in response to determining that the value of the loss function is less than a loss threshold, designating the initial predictive model as the training predictive model.

14. The method of claim 13, the training process further comprising:

updating the initial first portion or the initial second portion in response to determining that the value of the loss function is greater than or equal to the loss threshold.

15. A system for predicting traffic parameters, comprising:

a first acquisition module configured to:

acquiring at least two first inputs related to the driving condition corresponding to at least two candidate time points; and

obtaining a prediction of a first training;

a vector determination module configured to determine a target vector by using the first trained predictive model based on the at least two first inputs;

a second acquisition module configured to:

obtaining at least two second inputs relating to the driving condition corresponding to at least two future points in time; and

Acquiring a second training prediction model; and

a prediction module configured to predict, by the second trained predictive model, at least two future parameters related to the driving condition corresponding to the at least two future points in time based on the target vector and the at least two second inputs.

16. The system of claim 15, wherein the driving condition is associated with a road segment.

17. The system of claim 15 or 16, wherein each of the at least two first inputs comprises a first parameter related to traffic conditions at the respective candidate time point, at least two first historical parameters related to traffic conditions corresponding to at least two first historical time points, respectively, and a first statistical parameter related to the at least two first historical parameters.

18. The system of claim 17, wherein the first parameter relating to traffic conditions comprises at least one of a traffic congestion level of the driving conditions, a traffic speed of the driving conditions, or a traffic flow of the driving conditions.

19. The system of claim 17, wherein the first statistical parameter comprises a traffic congestion statistical parameter, a traffic speed statistical parameter, or a traffic flow statistical parameter.

20. The system of claim 19, wherein the first statistical parameter comprises the traffic congestion statistical parameter, and wherein the traffic congestion statistical parameter comprises at least one of a mode of at least two historical traffic congestion levels or a congestion probability of the at least two historical traffic congestion levels.

21. The system of claim 19, wherein the first statistical parameter comprises the traffic speed statistical parameter, and the traffic speed statistical parameter comprises at least one of a mean of at least two historical speeds, a median of the at least two historical speeds, a variance of the at least two historical speeds, a maximum of the at least two historical speeds, or a minimum of the at least two historical speeds.

22. The system of claim 19, wherein the first statistical parameter comprises the traffic flow statistical parameter, and the traffic flow statistical parameter comprises at least one of a mean of at least two historical traffic flows, a median of the at least two historical traffic flows, a variance of the at least two historical traffic flows, a maximum of the at least two historical traffic flows, or a minimum of the at least two historical traffic flows.

23. The system according to claim 15 or 16, wherein each of the at least two second inputs comprises a second parameter related to traffic conditions at a previous point in time of the respective future point in time, at least two second historical parameters related to traffic conditions corresponding to at least two second historical points in time, respectively, and a second statistical parameter related to the at least two second historical traffic parameters.

24. The system of claim 23, wherein each of the at least two second inputs further comprises a reference parameter comprising weather information for the respective future point in time.

25. The system of claim 15, wherein the first trained predictive model is a first part of a trained predictive model and the second trained predictive model is a second part of the trained predictive model.

26. The system of claim 25,

the trained predictive model is a sequence-to-sequence model,

the first part of the training of the prediction model is an encoder, and

the second part of the training of the prediction model is the decoder.

27. The system of claim 25 or 26, further comprising a training module configured to perform a training process to determine the training predictive model, the training process comprising:

28. The system of claim 27, the training process further comprising:

29. A system for predicting traffic parameters, comprising:

at least one storage medium comprising a set of instructions; and

at least one processor is in communication with the at least one storage medium, wherein the at least one processor, when executing the set of instructions, causes the system to perform the method of claims 1-14.

30. A non-transitory computer-readable medium comprising executable instructions, wherein when executed by at least one processor, the executable instructions instruct the at least one processor to perform the method of claims 1-14.