Driving intention stage identification method and system based on driver cognition stage division

Abstract

The invention discloses a driving intention stage recognition method and system based on driver cognition stage division, and relates to the technical field of auxiliary driving, and the method comprises the steps: collecting driving data, and constructing a feature vector according to the driving data; labeling a driving intention for the feature vector; performing statistics or splicing on the feature vectors corresponding to the continuous driving data by using a time window, and constructing a corresponding time window sample; inputting the time window sample into an intention stage discrimination unit, and outputting discrimination probabilities of the driver in different intention stages by using the intention stage discrimination unit; and a stage state management module is used to identify the stage of the current driving intention of the driver according to the discrimination probability. According to the method, each stage of the driving intention is divided, the stage of the driving intention can be identified according to the discrimination probability, and the identification result of the stage can be used as a prepositive constraint or cognitive state judgment basis of driving intention identification, so that the reliability, the stability and the safety are improved.

Description

Technical Field

[0001] This application relates to the field of driver assistance technology, and in particular to a method and system for identifying driving intention stages based on the division of driver cognitive stages. Background Technology

[0002] Driving intention recognition technology is a crucial technology in intelligent driving systems, driver assistance systems, and intelligent traffic management systems. Its main objective is to identify the driver's impending operational intentions by analyzing vehicle operating status, traffic environment information, and driver behavioral characteristics, thereby supporting vehicle control decisions, risk warnings, and traffic organization. Currently, most existing driving intention recognition methods are based on machine learning or deep learning models. They use multi-source driving data (including vehicle motion parameters, relative relationships with surrounding objects, driving operation behaviors, visual or eye-tracking data, etc.) as model input and directly output specific driving intention categories, such as lane changing or turning.

[0003] Related research and engineering implementations typically assume that the driver's driving intention has already been formed within the data time window input to the model. However, in real-world traffic scenarios, especially in high-load driving environments such as buses and complex urban roads, drivers' behavior is not always in a state of clear intention execution. A large number of driving behaviors are in intermediate stages of intention generation, hesitation, abandonment, or interruption, and existing intention recognition methods generally do not distinguish between these cognitive stages. Furthermore, current technologies generally treat driving intention recognition as a single-stage classification problem, ignoring the significant cognitive differences between "whether an intention is generated" and "how to execute the intention." This approach can easily lead to misjudgments, premature triggering of control strategies, and even safety hazards in practical applications. Summary of the Invention

[0004] The main objective of this application is to propose a method and system for identifying driving intention stages based on the driver's cognitive stages, so as to accurately identify the stage of driving intention.

[0005] To achieve the above objectives, one aspect of this application proposes a method for identifying driving intention stages based on the division of driver cognitive stages, the method comprising the following steps: Collect driving data and construct a feature vector based on the driving data; The feature vector is labeled with the stage of driving intention; wherein the driving intention includes at least one of left lane change, right lane change, left turn, right turn, lane keeping and parking, and each type of driving intention goes through an intention generation stage, an intention decision stage and an intention execution stage in time; By using time windows, the feature vectors corresponding to continuous driving data are statistically analyzed or spliced ​​to construct corresponding time window samples; The time window sample is input into the intent stage discrimination unit, and the intent stage discrimination unit outputs the discrimination probability that the driver is in different intent stages; The stage state management module uses the discrimination probability to identify the stage in which the driver's current driving intention is located.

[0006] In some embodiments, the collection of driving data includes the following steps: The driving data is collected by onboard sensors and driver monitoring equipment; wherein, the driving data includes vehicle operation data, driving operation data and driver visual behavior data.

[0007] In some embodiments, constructing the feature vector based on the driving data includes the following steps: Based on the driving data, vectors of spatial features, motion features, driving operation features, and visual features are constructed as the feature vectors.

[0008] In some embodiments, the stage of labeling the feature vector with driving intent includes the following steps: The execution events of the driving intention are determined based on changes in vehicle state and driver behavior, and the generation and decision-making stages of the driving intention are divided forward based on the execution events; wherein the execution events include at least one of the following: the vehicle's lateral position exceeds the boundary, the vehicle's steering trajectory changes significantly, and the vehicle speed drops to zero.

[0009] In some embodiments, the intent stage discrimination unit is a supervised learning model obtained according to machine learning methods or deep learning methods; The step of inputting the time window sample into the intent stage discrimination unit and using the intent stage discrimination unit to output the discrimination probability of the driver being in different intent stages includes the following steps: The supervised learning model is used to output the discrimination probability of the driver being in different stages of the driving intention for each time window sample; wherein the discrimination probability includes at least the discrimination probability of the driving intention generation stage and the discrimination probability of the driving intention decision stage.

[0010] In some embodiments, the step of using the stage state management module to identify the stage of the driver's current driving intention based on the discrimination probability includes the following steps: The stage state management module uses the discrimination probability to perform intention stage stability judgment, stage switching control, trigger judgment, and stage rollback and interruption judgment to identify the stage in which the driver's current driving intention is located. The intention stage stability judgment is used to determine whether the intention decision stage is stably formed within a continuous time window. The stage switching control is used to switch the current stage from the intent generation stage to the intent decision stage when the stability condition is met. The trigger judgment is used to call the intent execution judgment module to monitor whether the vehicle operation behavior meets the preset execution trigger conditions after determining that the driver is in the intent decision stage; when the corresponding driving behavior event is detected, it is determined that the driver has entered the intent execution stage. The stage rollback and interruption judgment is used to determine that the current driving intention has been interrupted if, after entering the intention decision stage but without detecting the execution of the triggering condition, the discrimination probability rolls back to the intention generation stage in the subsequent time window.

[0011] In some embodiments, the method further includes the following steps: The time window sample is updated based on the stage of the driver's current driving intention as identified, in order to update the subsequent discrimination probability.

[0012] To achieve the above objectives, another aspect of this application proposes a driving intention stage recognition system based on driver cognitive stage segmentation, the system comprising: The data preparation module is used to collect driving data and construct feature vectors based on the driving data. The stage labeling module is used to label the stage of driving intention on the feature vector; wherein, the driving intention includes at least one of left lane change, right lane change, left turn, right turn, lane keeping and parking, and each type of driving intention goes through an intention generation stage, an intention decision stage and an intention execution stage in time; The time window construction module is used to statistically analyze or concatenate the feature vectors corresponding to continuous driving data using a time window to construct corresponding time window samples. The probability discrimination module is used to input the time window sample into the intent stage discrimination unit and use the intent stage discrimination unit to output the discrimination probability that the driver is in different intent stages; The intent stage recognition module is used to identify the current stage of the driver's driving intent based on the discrimination probability using the stage state management module.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0015] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides a method and system for identifying driving intention stages based on driver cognitive stage segmentation. The solution involves collecting driving data and constructing feature vectors from the data; labeling the feature vectors with the stages of driving intention; where driving intentions include at least one of left lane change, right lane change, left turn, right turn, lane keeping, and parking, each type of driving intention experiencing an intention generation stage, an intention decision stage, and an intention execution stage in time; using time windows to statistically analyze or concatenate the feature vectors corresponding to continuous driving data to construct corresponding time window samples; inputting the time window samples into an intention stage discrimination unit, which outputs the discrimination probability of the driver being in different intention stages; and using a stage state management module to identify the current stage of the driver's driving intention based on the discrimination probability. This application segments driving intentions into various stages and can identify which stage the driver's driving intention is in based on the discrimination probability. The stage identification result can then be used as a pre-constraint for driving intention recognition or a basis for judging cognitive state, thereby improving the reliability, stability, and safety of the driving intention recognition system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the driving intention stage recognition method based on driver cognitive stage division provided in this application embodiment; Figure 2 An example flowchart of a driving intention stage recognition method based on driver cognitive stage division provided in an embodiment of this application; Figure 3 A flowchart of stage status management provided for embodiments of this application; Figure 4A schematic diagram of the structure of a driving intention stage recognition system based on driver cognitive stage division provided in an embodiment of this application; Figure 5 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] Before providing a detailed description of the embodiments of this application, some related technologies involved in the embodiments of this application will be described first, as follows: Existing driver intent recognition technology has at least the following shortcomings: (1) Lack of differentiation of driving intention cognition stages. Existing methods usually do not distinguish whether the driver is in the intention generation stage or the intention decision stage, and cannot determine whether the driving intention has been formed, making it difficult to identify intention interruption or hesitation.

[0022] (2) Directly classifying intent based on the characteristics of the decision-making stage can easily lead to false triggering. When the driver has not yet formed a clear intent, the model may still output a specific intent category, reducing the reliability of the system.

[0023] (3) There is a lack of a unified pre-constraint mechanism among different driving intent recognition models. Various intent models operate independently and fail to screen and constrain the intent recognition process at the system level.

[0024] (4) The model lacks effective characterization of individual differences among drivers and changes in cognitive strategies. In particular, the model's generalization ability is insufficient in multi-driver scenarios such as buses.

[0025] Therefore, this application aims to solve the following technical problem: how to construct a scheme that can identify whether a bus driver is currently in the intention generation stage or the intention decision stage, use the identification result of this stage as a prerequisite constraint or cognitive state judgment basis for bus driver intention recognition, and further identify the intention interruption state, thereby improving the reliability, stability and safety of the driver intention recognition system.

[0026] This application provides a method and system for identifying driving intention stages based on driver cognitive stage segmentation, relating to the field of assisted driving technology. The method and system provided in this application can be applied to a terminal, a server, or software running on a terminal or server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited thereto; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the server can also be a node server in a blockchain network; the software can be an application implementing the driving intention stage identification method based on driver cognitive stage segmentation, but is not limited to the above forms.

[0027] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0028] Reference Figure 1 This application provides a method for identifying driving intention stages based on the division of driver cognitive stages. This method may include, but is not limited to, steps S100 to S140, as follows: S100: Collect driving data and construct a feature vector based on the driving data; S110: The stage of labeling driving intentions on the feature vector; wherein, the driving intentions include at least one of left lane change, right lane change, left turn, right turn, lane keeping and parking, and each type of driving intention goes through an intention generation stage, an intention decision stage and an intention execution stage in time; S120: Utilize time windows to statistically analyze or concatenate the feature vectors corresponding to continuous driving data to construct corresponding time window samples; S130: Input the time window sample into the intent stage discrimination unit, and use the intent stage discrimination unit to output the discrimination probability that the driver is in different intent stages; S140: The stage of the driver's current driving intention is identified by the stage state management module based on the discrimination probability.

[0029] Optionally, the collection of driving data includes the following steps: The driving data is collected by onboard sensors and driver monitoring equipment; wherein, the driving data includes vehicle operation data, driving operation data and driver visual behavior data.

[0030] Optionally, constructing the feature vector based on the driving data includes the following steps: Based on the driving data, vectors of spatial features, motion features, driving operation features, and visual features are constructed as the feature vectors.

[0031] Optionally, the stage of labeling the feature vector with driving intent includes the following steps: The execution events of the driving intention are determined based on changes in vehicle state and driver behavior, and the generation and decision-making stages of the driving intention are divided forward based on the execution events; wherein the execution events include at least one of the following: the vehicle's lateral position exceeds the boundary, the vehicle's steering trajectory changes significantly, and the vehicle speed drops to zero.

[0032] Optionally, the intent stage discrimination unit is a supervised learning model obtained based on machine learning methods or deep learning methods; The step of inputting the time window sample into the intent stage discrimination unit and using the intent stage discrimination unit to output the discrimination probability of the driver being in different intent stages includes the following steps: The supervised learning model is used to output the discrimination probability of the driver being in different stages of the driving intention for each time window sample; wherein the discrimination probability includes at least the discrimination probability of the driving intention generation stage and the discrimination probability of the driving intention decision stage.

[0033] Optionally, the step-state management module identifies the current stage of the driver's driving intention based on the discrimination probability, including the following steps: The stage state management module uses the discrimination probability to perform intention stage stability judgment, stage switching control, trigger judgment, and stage rollback and interruption judgment to identify the stage in which the driver's current driving intention is located. The intention stage stability judgment is used to determine whether the intention decision stage is stably formed within a continuous time window. The stage switching control is used to switch the current stage from the intent generation stage to the intent decision stage when the stability condition is met. The trigger judgment is used to call the intent execution judgment module to monitor whether the vehicle operation behavior meets the preset execution trigger conditions after determining that the driver is in the intent decision stage; when the corresponding driving behavior event is detected, it is determined that the driver has entered the intent execution stage. The stage rollback and interruption judgment is used to determine that the current driving intention has been interrupted if, after entering the intention decision stage but without detecting the execution of the triggering condition, the discrimination probability rolls back to the intention generation stage in the subsequent time window.

[0034] Optionally, the method further includes the following steps: The time window sample is updated based on the stage of the driver's current driving intention as identified, in order to update the subsequent discrimination probability.

[0035] The following sections will provide a detailed description and explanation of some optional embodiments of this application, using specific application examples.

[0036] This embodiment proposes a driving intention stage recognition method and system based on driver cognitive stage division, which is used to identify the driving intention stage of the driver in real time during continuous driving, and realize closed-loop recognition of the generation, decision-making, execution and interruption of driving intention through stage stability judgment and execution event detection mechanism.

[0037] (I) Overall technical approach.

[0038] In complex traffic scenarios such as buses, the driving process is continuous, and driving intentions also exhibit continuous evolution. This embodiment assumes that there is no unintentional state during the driving process and uses six types of driving intentions—left lane change, right lane change, left turn, right turn, lane keeping, and parking—to fully describe the driving process. Each type of driving intention goes through an intention generation stage, an intention decision stage, and an intention execution stage in time.

[0039] To address the problem that existing driving intent recognition methods struggle to depict the formation and interruption of driving intent, this embodiment constructs an intent stage discrimination model and combines a time window stability judgment rule with an intent execution detection mechanism to perform hierarchical recognition of driving intent stages, thereby achieving dynamic switching and closed-loop management of driving intent stages.

[0040] (II) Method and Flow refer to Figure 2 The driving intent stage recognition method in this embodiment includes the following steps: Step S1: Driving data collection and feature construction.

[0041] The system first collects driving data through onboard sensors and driver monitoring equipment, including vehicle operation data, driving operation data, and driver visual behavior data. Based on the collected data, a unified feature vector is constructed as the model input, and the features include at least: Spatial characteristics: such as the relative positional relationship between the current lane and the target lane; Motion characteristics: such as vehicle speed, acceleration, and other operational status information; Driving operation characteristics: such as steering wheel angle, steering change trend, etc.; Visual characteristics: such as the distribution of the driver's gaze in front, to the left and right, and in the rearview mirror area.

[0042] By using a unified feature construction method, the samples from the intent generation stage and the intent decision stage are kept consistent in terms of feature dimensions, providing a foundation for the training and inference of the stage discrimination model.

[0043] Step S2: Marking the driving intention stage.

[0044] Based on changes in vehicle status and driver behavior, driving intention execution events are determined. Using these execution events as a benchmark, the intention generation stage and the intention decision stage are further divided. These intention execution events include, but are not limited to: the vehicle's lateral position crossing the boundary; a significant change in the vehicle's steering trajectory; and the vehicle's speed dropping to zero.

[0045] Step S3: Time window sample construction.

[0046] Continuous driving data is organized using a sliding time window approach. Within each time window, features are statistically analyzed or spliced ​​to construct corresponding time window samples.

[0047] Step S4: Intent stage discrimination unit.

[0048] The constructed time window samples are input into the intent stage discrimination unit. This discrimination unit, based on a supervised learning model, outputs the discrimination confidence level of the driver being in different intent stages for each time window sample, including at least: the discrimination probability of the driving intent generation stage; and the discrimination probability of the driving intent decision stage. The discrimination results are output in probabilistic form, providing basic input for subsequent stage state management.

[0049] The intent stage discrimination unit can be implemented using machine learning or deep learning methods, including but not limited to feature-based classification models or time-series modeling models. The key is to learn the behavioral differences of drivers at different cognitive stages, thereby achieving real-time discrimination of driving intent stages.

[0050] Step S5: Stage Status Management Module.

[0051] like Figure 3 As shown, a stage state management module is set up to determine and update the stage of the driver's current driving intention based on the discrimination probability output by the intention stage discrimination unit and in conjunction with the time window stability rule. The stage state management module includes: (1) Intention stage stability judgment, used to determine whether the intention decision stage is stably formed within a continuous time window; (2) Stage switching control, used to switch the current stage from the generation stage to the decision stage when the stability condition is met; (3) Execute the trigger judgment.

[0052] After determining that the driver is in the intention decision-making stage, the intention execution determination module is invoked to monitor whether the vehicle operation behavior meets the preset execution trigger conditions; when the corresponding driving behavior event is detected, it is determined that the driver has entered the intention execution stage. (4) Stage rollback and interruption judgment.

[0053] If the driving intent is interrupted after entering the intent decision stage but no execution trigger condition is detected, and the stage judgment result reverts to the generation stage within the subsequent time window, then the current driving intent is determined to be interrupted.

[0054] After completing a stage determination in each time window, the stage status management module feeds back the results to the time window sample construction step, thus forming a continuous and cyclical stage identification process.

[0055] Through the above technical solution, this embodiment realizes dynamic recognition and state management of the driving intention stage. It can not only identify whether the driver has entered the intention decision stage, but also identify intention interruption and stage rollback, effectively improving the stability and practicality of the driving intention recognition system in complex traffic scenarios.

[0056] For example, this embodiment provides a method for identifying driving intention stages based on the division of driver cognitive stages, which is applicable to driving behavior analysis and auxiliary decision-making scenarios in complex traffic environments such as public transportation vehicles.

[0057] (1) Data acquisition and preprocessing.

[0058] Driving data is collected using sensors installed inside and outside the vehicle. The features collected in this example include: the distance from the current vehicle to the previous bus stop, the distance to the next bus stop, the vehicle's real-time speed and speed changes, the time spent scanning the front of the vehicle, the left rearview mirror, the right rearview mirror, and the percentage of time spent scanning the rearview mirrors. The collected data is segmented according to different stages of intent and standardized to form a sample sequence for model input.

[0059] (2) Driving intention stage division.

[0060] The evolution of a single driving intention can be divided into the following stages: Intention generation stage: The driver has not yet performed any explicit vehicle control actions, but has already demonstrated environmental perception and cognitive readiness behaviors related to a specific driving task; Intent decision-making stage: After confirming the necessary environment, the driver gradually forms a clear control decision, and the vehicle control behavior is about to occur; Intent execution phase: The driver has performed the corresponding vehicle control action and executed the driving intent.

[0061] The start time of the intent execution phase can be determined by an event that significantly changes the vehicle's state, including but not limited to the vehicle crossing lane boundaries in lateral position, a significant change in the vehicle's steering trajectory, and a significant change in vehicle speed. The intent generation phase and the intent decision phase occur before the execution time, and their time range is determined based on changes in the driver's behavioral characteristics.

[0062] (3) Identification methods for the intention generation stage.

[0063] Before the execution of the intention, the intention generation stage is determined by analyzing the temporal changes in the driver's perceptual behavioral characteristics. In a specific implementation, the perceptual behavioral characteristics include the driver's visual attention distribution characteristics. The driver's line of sight attention is divided into several typical areas, including but not limited to the forward field of vision area, the left and right side field of vision areas, the left and right rearview mirror areas, and the instrument panel area. When it is detected that the driver's attention to the perceptual areas related to the target driving task significantly increases while the attention to irrelevant areas significantly decreases within a certain time period, the starting point of that time period is determined as the starting point of the intention generation stage.

[0064] (4) Identification methods in the intention decision-making stage.

[0065] Following the intent generation phase and preceding the intent execution phase, the intent decision-making phase is determined through further analysis of the driver's decision-making behavior characteristics. These characteristics include, but are not limited to: the duration of the driver's environmental confirmation; trends in vehicle operating status; and changes in the stability of the driver's visual attention in key perception areas. When the driver has completed the primary environmental confirmation and exhibited relevant decision-making characteristics, but has not yet triggered a vehicle execution event, the corresponding time period is defined as the intent decision-making phase.

[0066] (5) Model construction and stage identification.

[0067] Driving Intent Stage Recognition Model Training: Data samples from the intent generation and intent decision stages are input into the intent stage discrimination model for training. The intent stage discrimination unit outputs the probability value that the current time window belongs to the intent generation or intent decision stage. The output probability value is then input into the subsequent intent stage state management module to identify the formation, maintenance, interruption, and execution status of driving intentions.

[0068] (6) Experimental results.

[0069] Experimental results show that, using only the features of the generation and decision-making stages, the accuracy rate of intent stage identification can reach 84%, proving the feasibility of driving intent stage segmentation. It can also effectively identify situations where driving intent is interrupted, verifying the effectiveness of the method in this embodiment in driving intent stage identification and interruption determination.

[0070] Compared with the prior art, this embodiment has the following advantages: (1) Introduce driver cognitive stage division to realize the logical closed loop of intention recognition process.

[0071] This embodiment divides the process of forming driving intention into the intention generation stage, the intention decision stage, and the intention execution stage. It also realizes the dynamic switching and rollback of the stages through the stage state management mechanism, thereby constructing a closed loop of intention stage recognition covering the entire driving process. This effectively avoids the limitation of existing technologies that only recognize based on the execution behavior before and after.

[0072] (2) It can identify interruptions in driving intentions and improve the reliability of driving assistance systems.

[0073] This embodiment introduces a stage stability determination and stage rollback mechanism in the intention decision stage, which can identify situations where the driver's driving intention has been interrupted but has not entered the execution stage, thereby preventing downstream driving assistance or automatic control systems from being triggered at the wrong time and significantly improving system safety and reliability.

[0074] (3) It is applicable to a variety of vehicle models and driving scenarios and has good engineering scalability.

[0075] This embodiment does not rely on a specific vehicle platform or a single sensor configuration. It can flexibly select driver status characteristics, vehicle operation characteristics, and environmental characteristics according to different application scenarios, and is applicable to various driving scenarios such as buses and passenger cars.

[0076] The stage recognition model in this embodiment can be implemented using different algorithms, the time window samples can be constructed with a fixed length or an adaptive length, and the feature form can also be replaced according to the application scenario, without affecting the core idea of ​​this embodiment.

[0077] Other notes: 1. In this embodiment, "stage" refers to the driver's cognitive intention stage, rather than the vehicle's motion state; 2. The solution in this embodiment can operate as an independent driver monitoring system, or as a front-end module for a driving intention recognition or driving assistance system; 3. The scenarios and parameters in this embodiment are merely examples and do not constitute a limitation on the scope of protection of this embodiment.

[0078] Reference Figure 4 This application also provides a driving intention stage recognition system based on driver cognitive stage segmentation, which can implement the above-mentioned driving intention stage recognition method based on driver cognitive stage segmentation. The system includes: The data preparation module is used to collect driving data and construct feature vectors based on the driving data. The stage labeling module is used to label the stage of driving intention on the feature vector; wherein, the driving intention includes at least one of left lane change, right lane change, left turn, right turn, lane keeping and parking, and each type of driving intention goes through an intention generation stage, an intention decision stage and an intention execution stage in time; The time window construction module is used to statistically analyze or concatenate the feature vectors corresponding to continuous driving data using a time window to construct corresponding time window samples. The probability discrimination module is used to input the time window sample into the intent stage discrimination unit and use the intent stage discrimination unit to output the discrimination probability that the driver is in different intent stages; The intent stage recognition module is used to identify the current stage of the driver's driving intent based on the discrimination probability using the stage state management module.

[0079] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0080] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method of this application. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0081] It is understood that the content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the methods of this application, and the beneficial effects achieved are also the same as those achieved by the methods of this application.

[0082] Figure 5 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 101 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 102 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 102 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 102 and is called and executed by the processor 101. Input / output interface 103 is used to implement information input and output; The communication interface 104 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 105 transmits information between various components of the device (e.g., processor 101, memory 102, input / output interface 103, and communication interface 104); The processor 101, memory 102, input / output interface 103 and communication interface 104 are connected to each other within the device via bus 105.

[0083] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of this application.

[0084] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0085] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0086] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0087] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0088] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0089] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0090] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0091] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0092] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0094] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for identifying driving intention stages based on driver cognitive stage segmentation, characterized in that, The method includes the following steps: Collect driving data and construct a feature vector based on the driving data; The feature vector is labeled with the stage of driving intention; wherein the driving intention includes at least one of left lane change, right lane change, left turn, right turn, lane keeping and parking, and each type of driving intention goes through an intention generation stage, an intention decision stage and an intention execution stage in time; By using time windows, the feature vectors corresponding to continuous driving data are statistically analyzed or spliced ​​to construct corresponding time window samples; The time window sample is input into the intent stage discrimination unit, and the intent stage discrimination unit outputs the discrimination probability that the driver is in different intent stages; The stage state management module uses the discrimination probability to identify the stage in which the driver's current driving intention is located.

2. The driving intention stage recognition method based on driver cognitive stage division according to claim 1, characterized in that, The collection of driving data includes the following steps: The driving data is collected by onboard sensors and driver monitoring equipment; wherein, the driving data includes vehicle operation data, driving operation data and driver visual behavior data.

3. The driving intention stage recognition method based on driver cognitive stage division according to claim 1, characterized in that, The process of constructing a feature vector based on the driving data includes the following steps: Based on the driving data, vectors of spatial features, motion features, driving operation features, and visual features are constructed as the feature vectors.

4. The driving intention stage recognition method based on driver cognitive stage division according to claim 1, characterized in that, The stage of labeling the feature vector with driving intent includes the following steps: The execution events of the driving intention are determined based on changes in vehicle state and driver behavior, and the generation and decision-making stages of the driving intention are divided forward based on the execution events; wherein the execution events include at least one of the following: the vehicle's lateral position exceeds the boundary, the vehicle's steering trajectory changes significantly, and the vehicle's speed drops to zero.

5. The driving intention stage recognition method based on driver cognitive stage segmentation according to claim 1, characterized in that, The intent stage discrimination unit is a supervised learning model obtained based on machine learning methods or deep learning methods; The step of inputting the time window sample into the intent stage discrimination unit and using the intent stage discrimination unit to output the discrimination probability of the driver being in different intent stages includes the following steps: The supervised learning model is used to output the discrimination probability of the driver being in different stages of driving intention for each time window sample; The discrimination probability mentioned therein includes at least the discrimination probability of the driving intention generation stage and the discrimination probability of the driving intention decision stage.

6. The driving intention stage recognition method based on driver cognitive stage division according to claim 1, characterized in that, The stage state management module identifies the current stage of the driver's driving intention based on the discrimination probability, including the following steps: The stage state management module uses the discrimination probability to perform intention stage stability judgment, stage switching control, trigger judgment, and stage rollback and interruption judgment to identify the stage in which the driver's current driving intention is located. The intention stage stability judgment is used to determine whether the intention decision stage is stably formed within a continuous time window. The stage switching control is used to switch the current stage from the intent generation stage to the intent decision stage when the stability condition is met. The trigger judgment is used to call the intent execution judgment module to monitor whether the vehicle operation behavior meets the preset execution trigger conditions after determining that the driver is in the intent decision stage; when the corresponding driving behavior event is detected, it is determined that the driver has entered the intent execution stage. The stage rollback and interruption judgment is used to determine that the current driving intention has been interrupted if, after entering the intention decision stage but without detecting the execution of the triggering condition, the discrimination probability rolls back to the intention generation stage in the subsequent time window.

7. The method for identifying driving intention stages based on driver cognitive stage division according to any one of claims 1 to 6, characterized in that, The method further includes the following steps: The time window sample is updated based on the stage of the driver's current driving intention as identified, in order to update the subsequent discrimination probability.

8. A driving intention stage recognition system based on driver cognitive stage segmentation, characterized in that, The system includes: The data preparation module is used to collect driving data and construct feature vectors based on the driving data. The stage labeling module is used to label the stage of driving intention on the feature vector; wherein, the driving intention includes at least one of left lane change, right lane change, left turn, right turn, lane keeping and parking, and each type of driving intention goes through an intention generation stage, an intention decision stage and an intention execution stage in time; The time window construction module is used to statistically analyze or concatenate the feature vectors corresponding to continuous driving data using a time window to construct corresponding time window samples. The probability discrimination module is used to input the time window sample into the intent stage discrimination unit and use the intent stage discrimination unit to output the discrimination probability that the driver is in different intent stages; The intent stage recognition module is used to identify the current stage of the driver's driving intent based on the discrimination probability using the stage state management module.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images