Vehicle control method and apparatus, and terminal device and storage medium

Abstract

Disclosed in the present application are a vehicle control method and apparatus, and a terminal device and a storage medium. The method comprises: collecting a user location trajectory of a preset user outside a vehicle; performing intention recognition on the basis of a pre-trained intention recognition model and on the basis of the user location trajectory, so as to acquire an intention recognition result, wherein the intention recognition model is obtained by means of model training based on a pre-acquired historical user trajectory set and historical user behavior set; and controlling the vehicle on the basis of the intention recognition result.

Description

Vehicle control method, device, terminal device and storage medium

[0001] Related applications

[0002] This application claims priority to Chinese patent application No. 202410261758.4 filed on March 7, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of vehicle networking technology, and in particular to a vehicle control method, apparatus, terminal device, and storage medium. Background Art

[0004] While promoting the upgrade of automobile products, the Internet of Vehicles allows cars to evolve from transportation tools to smart terminals with the ability to interact and provide services. It usually uses Bluetooth, wireless radio frequency or infrared to seamlessly control vehicle unlocking and trunk opening.

[0005] However, current vehicle control is limited to unlocking through signal sensing and identity recognition within the range, and there is no senseless control based on user needs, resulting in a poor user experience.

[0006] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Technical issues

[0007] The main purpose of this application is to provide a vehicle control method, apparatus, terminal device and storage medium, aiming to solve the problem that vehicle control does not control according to user intention, resulting in poor user experience. Technical Solutions

[0008] To achieve the above objectives, the present application provides a vehicle control method, the vehicle control method comprising:

[0009] Collect user location trajectories of preset users outside the vehicle;

[0010] Based on a pre-trained intent recognition model, intent recognition is performed according to the user location trajectory to obtain an intent recognition result, wherein the intent recognition model is trained based on a pre-acquired historical user trajectory set and a historical user behavior set;

[0011] The vehicle is controlled according to the intention recognition result.

[0012] In one embodiment, the step of performing intent recognition based on the user location trajectory based on a pre-trained intent recognition model and obtaining an intent recognition result further includes:

[0013] Build an initial intent recognition model based on the preset intent recognition algorithm;

[0014] The initial intent recognition model is trained based on the previously acquired historical user trajectory set and historical user behavior set to obtain an intent recognition model.

[0015] In one embodiment, the step of training the initial intent recognition model based on the pre-acquired historical user trajectory set and historical user behavior set to obtain the intent recognition model includes:

[0016] Constructing a behavior area map set based on the historical user trajectory set and the historical user behavior set;

[0017] The behavior area atlas is input into the initial intention recognition model for model training to obtain an intention recognition model.

[0018] In one embodiment, the step of constructing a behavior area atlas based on the historical user trajectory set and the historical user behavior set includes:

[0019] Obtaining a user behavior type based on the historical user behavior set;

[0020] Classifying the historical user trajectory set according to the user behavior type to obtain a behavior trajectory set;

[0021] Performing region calibration according to the behavior trajectory set to obtain a behavior region set;

[0022] The behavior area set is mapped to a preset vehicle perception map to obtain a behavior area map set.

[0023] In one embodiment, the step of performing intent recognition based on the user location trajectory based on a pre-trained intent recognition model and obtaining an intent recognition result includes:

[0024] Based on a preset behavior area map, obtaining a user location set according to the user location trajectory;

[0025] The user location set is input into the intent recognition model to perform intent recognition and obtain an intent recognition result.

[0026] In one embodiment, the step of inputting the user location set into the intent recognition model to perform intent recognition and obtaining an intent recognition result includes:

[0027] The user location set is input into the intent recognition model for the following processing:

[0028] Obtaining a current location point and a previous location point according to the user location set;

[0029] Obtaining a current behavior type and a previous behavior type according to the behavior types corresponding to the current location point and the previous location point in a preset behavior area map;

[0030] If the current behavior type is consistent with the previous behavior type and the stay time at the current location point exceeds a preset time threshold, the current behavior recognition type is obtained as an intention recognition result.

[0031] In one embodiment, the vehicle control method further includes:

[0032] Perform user identification through a preset sensor and obtain a user identification result;

[0033] If the user identification result is the first preset user, the vehicle control includes at least one or more of turning on a vehicle search indicator light, unlocking a door, and opening a trunk;

[0034] If the user identification result is a second preset user, the vehicle control at least includes unlocking the passenger door.

[0035] The present application also provides a vehicle control device, comprising:

[0036] A collection module, used to collect user location trajectories of preset users outside the vehicle;

[0037] An identification module is configured to identify the intent of the user based on the user's location trajectory based on a pre-trained intent identification model, and obtain an intent identification result. The intent identification model is trained based on a pre-acquired historical user trajectory set and historical user behavior set.

[0038] A control module is used to control the vehicle according to the intention recognition result.

[0039] An embodiment of the present application also proposes a terminal device, which includes a memory, a processor, and a vehicle control program stored in the memory and executable on the processor. When the vehicle control program is executed by the processor, the steps of the vehicle control method described above are implemented.

[0040] An embodiment of the present application further provides a computer-readable storage medium, on which a vehicle control program is stored. When the vehicle control program is executed by a processor, the steps of the vehicle control method described above are implemented. Beneficial effects

[0041] The vehicle control method, apparatus, terminal device and storage medium proposed in the embodiments of the present application collect the user location trajectory of a preset user outside the vehicle; based on a pre-trained intention recognition model, perform intention recognition according to the user location trajectory to obtain an intention recognition result, the intention recognition model is trained according to a pre-acquired historical user trajectory set and historical user behavior set; and perform vehicle control according to the intention recognition result. The present application collects the user location trajectory of a preset user outside the vehicle, performs intention recognition according to the user location trajectory through an intention recognition model, and then uses the intention recognition result to control the vehicle, thereby achieving senseless vehicle control based on user intentions, solving the problem of poor user experience caused by vehicle control not being based on user intentions, and improving the accuracy and convenience of senseless vehicle control. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] FIG1 is a schematic diagram of the functional modules of a terminal device to which the vehicle control device of the present application belongs;

[0043] FIG2 is a flow chart of a first exemplary embodiment of a vehicle control method of the present application;

[0044] FIG3 is a flow chart of a second exemplary embodiment of a vehicle control method of the present application;

[0045] FIG4 is a schematic diagram of an example of a historical behavior trajectory of the vehicle control method of the present application;

[0046] FIG5 is a schematic diagram showing an example of the classification of behavior areas of the vehicle control method of the present application;

[0047] FIG6 is a schematic diagram of an example of a behavior area map of the vehicle control method of the present application;

[0048] FIG7 is a flow chart of a third exemplary embodiment of a vehicle control method of the present application;

[0049] FIG8 is a flow chart of a fourth exemplary embodiment of a vehicle control method of the present application.

[0050] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. Modes for Carrying Out the Invention

[0051] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] The main solution of the embodiment of the present application is: collecting the user location trajectory of the preset user outside the vehicle; based on the pre-trained intention recognition model, performing intention recognition according to the user location trajectory to obtain the intention recognition result, the intention recognition model is trained according to the pre-acquired historical user trajectory set and historical user behavior set; and controlling the vehicle according to the intention recognition result. The present application realizes senseless vehicle control according to the user's intention by collecting the user location trajectory of the preset user outside the vehicle, performing intention recognition according to the user location trajectory through the intention recognition model, and then using the intention recognition result to control the vehicle, thereby solving the problem of poor user experience caused by vehicle control not being controlled according to user intention, and improving the accuracy and convenience of senseless vehicle control.

[0053] The embodiments of the present application take into account that the relevant technical solutions use devices or keys with Bluetooth, wireless radio frequency and other functions to perform contactless vehicle control, but the existing technology of contactless vehicle control is limited to unlocking the door when the user approaches the vehicle and the vehicle senses unlocking signals such as Bluetooth, wireless radio frequency, etc., and does not perform control in accordance with user needs.

[0054] Based on this, the embodiment of the present application proposes a solution that, through machine learning and data analysis, more deeply combines vehicle control and user needs, provides users with more personalized services and experiences, intelligently identifies user needs and automatically performs corresponding vehicle control operations, thereby further improving user satisfaction and ease of use.

[0055] Specifically, referring to Figure 1, Figure 1 is a schematic diagram of the functional modules of the terminal device to which the vehicle control device of this application belongs. The vehicle control device can be a device independent of the terminal device that can control the vehicle, and can be hosted on the terminal device in the form of hardware or software. The terminal device can be a smart mobile device with vehicle control functions, such as a mobile phone or tablet computer, or a fixed terminal device or server with vehicle control functions.

[0056] In this embodiment, the terminal device to which the vehicle control apparatus belongs includes at least an output module 110 , a processor 120 , a memory 130 and a communication module 140 .

[0057] The memory 130 stores an operating system and vehicle control programs. The vehicle control device can store received and processed data in the memory 130. The output module 110 can be a display screen, a speaker, etc. The communication module 140 can include a Wi-Fi module, a mobile communication module, and a Bluetooth module, etc., and communicates with external devices or servers through the communication module 140.

[0058] When the vehicle control program in the memory 130 is executed by the processor, the following steps are implemented:

[0059] Collect user location trajectories of preset users outside the vehicle;

[0060] Based on a pre-trained intent recognition model, intent recognition is performed according to the user location trajectory to obtain an intent recognition result, wherein the intent recognition model is trained based on a pre-acquired historical user trajectory set and a historical user behavior set;

[0061] The vehicle is controlled according to the intention recognition result.

[0062] Furthermore, when the vehicle control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0063] Build an initial intent recognition model based on the preset intent recognition algorithm;

[0064] The initial intent recognition model is trained based on the previously acquired historical user trajectory set and historical user behavior set to obtain an intent recognition model.

[0065] Furthermore, when the vehicle control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0066] Constructing a behavior area map set based on the historical user trajectory set and the historical user behavior set;

[0067] The behavior area atlas is input into the initial intention recognition model for model training to obtain an intention recognition model.

[0068] Furthermore, when the vehicle control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0069] Obtaining a user behavior type based on the historical user behavior set;

[0070] Classifying the historical user trajectory set according to the user behavior type to obtain a behavior trajectory set;

[0071] Performing region calibration according to the behavior trajectory set to obtain a behavior region set;

[0072] The behavior area set is mapped to a preset vehicle perception map to obtain a behavior area map set.

[0073] Furthermore, when the vehicle control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0074] Based on a preset behavior area map, obtaining a user location set according to the user location trajectory;

[0075] The user location set is input into the intent recognition model to perform intent recognition and obtain an intent recognition result.

[0076] Furthermore, when the vehicle control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0077] The user location set is input into the intent recognition model for the following processing:

[0078] Obtaining a current location point and a previous location point according to the user location set;

[0079] Obtaining a current behavior type and a previous behavior type according to the behavior types corresponding to the current location point and the previous location point in a preset behavior area map;

[0080] If the current behavior type is consistent with the previous behavior type and the stay time at the current location point exceeds a preset time threshold, the current behavior recognition type is obtained as an intention recognition result.

[0081] Furthermore, when the vehicle control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0082] Perform user identification through a preset sensor and obtain a user identification result;

[0083] If the user identification result is the first preset user, the vehicle control includes at least one or more of turning on a vehicle search indicator light, unlocking a door, and opening a trunk;

[0084] If the user identification result is a second preset user, the vehicle control at least includes unlocking the passenger door.

[0085] This embodiment adopts the above scheme, specifically by collecting the user location trajectory of the preset user outside the vehicle; based on the pre-trained intention recognition model, performing intention recognition according to the user location trajectory, and obtaining the intention recognition result, the intention recognition model is trained according to the pre-acquired historical user trajectory set and historical user behavior set; and controlling the vehicle according to the intention recognition result. This application collects the user location trajectory of the preset user outside the vehicle, performs intention recognition according to the user location trajectory through the intention recognition model, and then uses the intention recognition result to control the vehicle, thereby realizing senseless vehicle control according to the user's intention, solving the problem of poor user experience caused by vehicle control not being controlled according to the user's intention, and improving the accuracy and convenience of senseless vehicle control.

[0086] Based on the above terminal device architecture but not limited to the above architecture, an embodiment of the method of the present application is proposed.

[0087] 2, which is a flow chart of a first exemplary embodiment of a vehicle control method of the present application. The vehicle control method includes:

[0088] Step S10: collecting the user location trajectory of the preset user outside the vehicle.

[0089] The execution subject of the method of this embodiment can be a vehicle control device, or a vehicle control terminal device or server. This embodiment takes a vehicle control device as an example, and the vehicle control device can be integrated into a terminal device with data processing function.

[0090] Collecting the location trajectory of pre-set users outside the vehicle. This can be accomplished through the vehicle's built-in sensors, which can collect data on the user's location trajectory outside the vehicle from devices such as smartphones, other wearable devices, or vehicle keys that have signal transceiver capabilities (including but not limited to Bluetooth, infrared, and wireless radio frequency). For example, the vehicle's Bluetooth sensors can be used to obtain real-time location information of authorized users' devices outside the vehicle, generating a user location trajectory. Alternatively, the vehicle network can be used to collect positioning information of authorized users outside the vehicle, generating a user location trajectory.

[0091] Step S20: Based on a pre-trained intention recognition model, perform intention recognition according to the user location trajectory to obtain an intention recognition result. The intention recognition model is trained based on a pre-acquired historical user trajectory set and a historical user behavior set.

[0092] Intent recognition is performed using the user's location trajectory outside the vehicle as the basis for intent recognition, using a pre-trained intent recognition model to generate the results. This pre-trained intent recognition model, based on a machine learning algorithm, uses statistical relationships between user behavior patterns and characteristics in historical datasets to determine the purpose and intent of the current user's behavior. The model's output serves as input to the vehicle control system, helping it better understand the user's intentions and thus provide a more intelligent and personalized automotive service experience. The intent recognition model analyzes this trajectory data and identifies the user's likely behavioral intentions, such as whether the user is preparing to return to the vehicle, whether the vehicle needs to be unlocked, or whether the trunk needs to be opened.

[0093] Step S30: Control the vehicle according to the intention recognition result.

[0094] Based on the results of intent recognition, the system performs corresponding non-sensing vehicle control operations, such as turning on the vehicle search indicator light, unlocking the vehicle, opening the trunk, etc., to meet the user's needs, including:

[0095] 1. Turn on the vehicle search indicator light: If the user's intention to return to the vehicle is recognized, the vehicle's vehicle search indicator light can be turned on through remote control to help the user quickly find their vehicle.

[0096] 2. Unlock the vehicle: If it is recognized that the user needs to unlock the vehicle, the vehicle can be unlocked by remote control, allowing the user to easily enter the vehicle.

[0097] 3. Open the trunk: If it is recognized that the user needs to open the trunk, the trunk of the vehicle can be opened remotely to facilitate the user to store or take out items.

[0098] Through contactless vehicle control operations, users can control the vehicle without manual operation and enjoy a more intelligent and convenient car experience.

[0099] This embodiment adopts the above scheme, specifically by collecting the user location trajectory of the preset user outside the vehicle; based on the pre-trained intention recognition model, performing intention recognition according to the user location trajectory, and obtaining the intention recognition result, the intention recognition model is trained according to the pre-acquired historical user trajectory set and historical user behavior set; and controlling the vehicle according to the intention recognition result. This application collects the user location trajectory of the preset user outside the vehicle, performs intention recognition according to the user location trajectory through the intention recognition model, and then uses the intention recognition result to control the vehicle, thereby realizing senseless vehicle control according to the user's intention, solving the problem of poor user experience caused by vehicle control not being controlled according to the user's intention, and improving the accuracy and convenience of senseless vehicle control.

[0100] 3 , which is a flow chart of a second exemplary embodiment of a vehicle control method of the present application.

[0101] Based on the first embodiment, a second embodiment of the present application is proposed. The difference between the second embodiment of the present application and the first embodiment is that:

[0102] In this embodiment, the step of performing intent recognition based on the user location trajectory based on a pre-trained intent recognition model and obtaining the intent recognition result further includes:

[0103] Step S201: constructing an initial intent recognition model according to a preset intent recognition algorithm;

[0104] Step S202: Based on the pre-acquired historical user trajectory set and historical user behavior set, the initial intent recognition model is trained to obtain an intent recognition model.

[0105] Specifically, an initial intent recognition model is first constructed based on an intent recognition algorithm. This algorithm can be a machine learning-based classification algorithm (e.g., decision tree, support vector machine, neural network, etc.) or a deep learning-based model (e.g., recurrent neural network, convolutional neural network, etc.). The initial model is constructed based on the algorithm's requirements, requiring it to be able to use data such as user location trajectory as model input to perform intent recognition and generate model outputs that can be used for vehicle control.

[0106] Finally, we use the existing datasets—historical user trajectory data and historical user behavior data—as training datasets and input them into the intent recognition model for training. This training process helps the model learn the statistical relationships between user behavior patterns and behavioral characteristics from historical data, enabling it to better identify user intent.

[0107] Furthermore, as an implementation method, the step of training the initial intent recognition model based on the pre-acquired historical user trajectory set and historical user behavior set to obtain the intent recognition model includes:

[0108] Step S2021: constructing a behavior area map set based on the historical user trajectory set and the historical user behavior set;

[0109] Step S2022: inputting the behavior area atlas into the initial intention recognition model for model training to obtain an intention recognition model.

[0110] Specifically, first, a behavioral area atlas is constructed based on historical user trajectory sets and historical user behavior sets. The construction of the behavioral area atlas requires processing historical user trajectory data, dividing the user's historical location trajectory into different areas, and associating each area with a specific user behavior, such as searching for a vehicle, unlocking a vehicle, and opening the trunk. This allows the construction of a behavioral area atlas, where each area corresponds to a specific behavior type.

[0111] Finally, the behavior area atlas is input into the initial intent recognition model for model training, resulting in an intent recognition model. The behavior area atlas serves as training data and is then input into the initial intent recognition model for model training. Different algorithms and model structures can be used for training based on different behavior types to achieve optimal intent recognition. This training results in an accurate and reliable intent recognition model that can be applied to real-time user data to accurately identify and predict user behavioral intent, improving the intelligence and accuracy of vehicle control.

[0112] Furthermore, as an embodiment, the step of constructing a behavior area atlas according to the historical user trajectory set and the historical user behavior set includes:

[0113] Step S20211: Obtaining a user behavior type based on the historical user behavior set;

[0114] Step S20212: Classifying the historical user trajectory set according to the user behavior type to obtain a behavior trajectory set;

[0115] Step S20213: performing region calibration according to the behavior trajectory set to obtain a behavior region set;

[0116] Step S20214: Map the behavior area set to a preset vehicle perception map to obtain a behavior area map set.

[0117] Specifically, first, the historical user behavior data is processed to identify the specific behavior types performed by each user, such as searching for a vehicle, unlocking a vehicle, opening the trunk, etc.

[0118] Then, the historical user trajectory set is classified according to the user behavior type to obtain the behavior trajectory set. In this process, the historical user trajectory data needs to be processed and the user's movement trajectory is classified according to different behavior types to obtain the corresponding behavior trajectory set.

[0119] Then, region calibration is performed based on the behavior trajectory set to obtain a behavior region set. Region calibration is performed on the behavior trajectory set to divide the trajectory data under each behavior type into different regions, and each region is associated with the corresponding behavior type to obtain a behavior region set with corresponding regions and behaviors.

[0120] Finally, the behavior region set is mapped onto the preset vehicle perception map to obtain a behavior region atlas. This requires mapping the behavior region set onto a map where the vehicle can perceive the user's location, resulting in a behavior region atlas. During the mapping process, the scale and accuracy of the vehicle perception map must be considered to ensure that the behavior region map accurately reflects the vehicle's actual location and behavior trajectory.

[0121] More specifically, the system will record the dwell time and operations (such as finding the car, unlocking, opening the trunk, etc.) at each point based on the position change curve of the user's key at each point next to the car.

[0122] As shown in Figure 4, Figure 4 is a schematic diagram of an example of the historical behavior trajectory of the vehicle control method of the present application. The curve in the figure is the user's behavior trajectory curve, where the position of AC can be classified as the intention to find the vehicle, and the control to be executed by the vehicle is to turn on the vehicle search indicator light. The position of DI can be classified as the intention to unlock the vehicle, and the control to be executed by the vehicle is to unlock the vehicle door.

[0123] As shown in Figure 5, a sample diagram of the behavior region classification of the vehicle control method of the present application is shown. Based on the behavior trajectory set, region calibration is performed. The Z0 region corresponding to AC can be calibrated as the behavior region for vehicle search intent, the Z1 region corresponding to DI (excluding G) can be calibrated as the behavior region for unlocking intent, and the Z2 region can be calibrated as the behavior region for trunk opening intent.

[0124] As shown in Figure 6, which is a sample diagram of the behavior area map of the vehicle control method of the present application, after multiple behavior area calibration processes, the final behavior area set is mapped to the preset vehicle perception map to obtain the final behavior area map set.

[0125] This embodiment utilizes the above solution to construct an initial intent recognition model based on a preset intent recognition algorithm. This initial intent recognition model is then trained based on a previously acquired set of historical user trajectories and behaviors to generate an intent recognition model. The trained intent recognition model can be applied to real-time user data to accurately identify and predict user behavior intent, providing a better user experience and vehicle control services.

[0126] 7 , which is a flowchart of a third exemplary embodiment of a vehicle control method of the present application.

[0127] Based on the first embodiment, a third embodiment of the present application is proposed. The difference between the third embodiment of the present application and the first embodiment is that:

[0128] In this embodiment, the steps of performing intent recognition based on the user location trajectory based on the pre-trained intent recognition model and obtaining the intent recognition result include:

[0129] Step S203: Based on a preset behavior area map, a user location set is obtained according to the user location trajectory;

[0130] Step S204: input the user location set into the intention recognition model to perform intention recognition and obtain an intention recognition result.

[0131] Specifically, first, the user's actual location trajectory data is used in combination with a pre-built behavior area map to map the user's location trajectory to the corresponding behavior area, and the user's location information in different behavior areas is obtained to form a user location set.

[0132] Finally, a pre-trained intent recognition model is used, taking the user's location set as input. Through the model's reasoning process, the user's behavioral intent is identified, resulting in an intent recognition result. The intent recognition model can determine the user's intent, i.e., the specific type of behavior the user may perform (such as finding a vehicle, unlocking the vehicle, opening the trunk, etc.), based on the user's location information in different behavioral areas, to obtain an intent recognition result. The specific recognition process of the intent recognition model can be to form a continuous location trajectory curve based on the location points in the user's location set, and then use a specific recognition algorithm (such as curve fitting, least squares method, etc.) to determine whether the curve conforms to the characteristic location curve of a specific behavioral intention, thereby obtaining an intent recognition result. Furthermore, the specific recognition process of the intent recognition model can also be to further determine the connection between different location points based on the attribute characteristics of the location points in the user's location set, including but not limited to time, direction, and other attributes, thereby identifying the user's intent and obtaining an intent recognition result.

[0133] Furthermore, as an implementation manner, the step of inputting the user location set into the intent recognition model for intent recognition and obtaining the intent recognition result includes:

[0134] The user location set is input into the intent recognition model for the following processing:

[0135] Step S2041: obtaining the current location point and the previous location point according to the user location set;

[0136] Step S2042: obtaining the current behavior type and the previous behavior type according to the behavior types corresponding to the current location point and the previous location point in the preset behavior area map;

[0137] Step S2043: If the current behavior type is consistent with the previous behavior type and the stay time at the current location point exceeds a preset time threshold, the current behavior recognition type is obtained as an intention recognition result.

[0138] Specifically, first, based on the time sequence of the user's location set, the two nearest adjacent locations are obtained: the current location and the previous location. The user location set is a location data set arranged in time order, and the model can identify user intent based on the two most recent location points.

[0139] Then, based on the behavior types corresponding to the current and previous locations in the preset behavior area map, the current and previous behavior types are obtained. The current and previous locations need to be mapped to the preset behavior area map to obtain the behavior types corresponding to different locations. By comparing the current and previous behavior types, it is possible to determine whether the user's behavioral intention has changed.

[0140] Finally, if the current behavior type matches the previous behavior type and the dwell time at the current location exceeds the preset time threshold, the current behavior recognition type is determined as intent recognition. This involves calculating the user's dwell time at the current location based on the timestamp of the current location and the timestamp of the previous location, and comparing it with the preset time threshold. If the user's dwell time at the current location exceeds the preset time threshold and the current behavior type matches the previous behavior type, the user's behavioral intention is considered recognized. At this point, the current behavior type is the intent recognition result.

[0141] More specifically, the location points of the user's location trajectory are combined with the residence time to increase the accuracy of regional area judgment.

[0142] As shown in Figure 4, a user spends time t at point E without performing any operation, but then performs an operation at point F. This indicates that the user at point E intends to perform an operation at point F. By repeatedly mapping the area, we can determine whether the point belongs to the area where the operation occurred. If it is an isolated point, it is discarded; if it is surrounded by points in the area where point F was operating, it belongs to the same area. Compared to traditional calibration areas, this solution maps areas based on actual user behavior, making it more accurate and practical.

[0143] This embodiment utilizes the above solution to obtain a user location set based on the user's location trajectory based on a preset behavior area map; then inputs the user location set into the intent recognition model for intent recognition, obtaining an intent recognition result. Intent recognition based on the user's location trajectory can help understand the user's specific behavioral intentions, which can be used to guide vehicle control decisions and execution, providing a more intelligent and personalized interactive experience.

[0144] 8 , which is a flowchart of a fourth exemplary embodiment of a vehicle control method of the present application.

[0145] Based on the first embodiment, a fourth embodiment of the present application is proposed. The fourth embodiment of the present application differs from the first embodiment in that:

[0146] In this embodiment, the vehicle control method further includes:

[0147] Step S40: performing user identification through a preset sensor to obtain a user identification result;

[0148] Step S50: If the user identification result is the first preset user, the vehicle control includes at least one or more of turning on a vehicle search indicator light, unlocking the doors, and opening the trunk;

[0149] Step S60: If the user identification result is the second preset user, the vehicle control at least includes unlocking the passenger door.

[0150] Specifically, user identification is first performed using pre-set sensors to obtain a user identification result. This can be done using pre-set sensors, such as facial recognition or voiceprint recognition, or through network authorization and identification using Bluetooth, radio frequency, or other signals. By comparing the user's biometrics or other identity information, the user's identity can be determined and a user identification result obtained, enabling vehicle control based on the user's identity.

[0151] Then, if the user's identity is identified as the first preset user, vehicle control includes at least one or more of turning on the vehicle search light, unlocking the doors, and opening the trunk. The first preset user is a user with a pre-set identity, and their actual identity may be the vehicle owner's identity pre-set on the vehicle terminal or vehicle app. This is to confirm whether the user is the vehicle owner. If the user's identity is identified as the vehicle owner, vehicle control can continue based on the intention recognition results. For example, if the user is identified as the vehicle owner, vehicle control permissions such as turning on the vehicle search light, unlocking / opening the driver's door or all doors, starting the engine, and opening the trunk can be performed based on the user's intention. This means that the first preset user has advanced vehicle control permissions, which can be further differentiated based on the vehicle owner and other vehicle users to achieve personalized services.

[0152] Finally, if the user's identity is identified as the second preset user, vehicle control will at least include unlocking the passenger door. The second preset user is a user with another pre-defined identity, and their actual identity may be a passenger identity pre-set on the vehicle terminal or vehicle app. This is to confirm whether the user is a passenger. If the user's identity is identified as a passenger, vehicle control permissions such as unlocking / opening the passenger door or all doors, and opening the trunk can continue to be performed according to the user's intent. This means that the second preset user has lower-level vehicle control permissions, and users with authorized permissions can change the second preset user's vehicle control permissions through authorization settings. Furthermore, if user identity identification fails, it means that the user is not an authorized user of the vehicle and does not have vehicle control permissions.

[0153] This embodiment utilizes the above-described solution to specifically identify a user through a preset sensor and obtain a user identification result. If the user identification result is the first preset user, the vehicle control includes at least one of activating a vehicle search indicator light, unlocking the doors, and opening the trunk. If the user identification result is the second preset user, the vehicle control includes at least unlocking the passenger door. User identity verification is performed before vehicle control is performed, and different vehicle control operations can only be executed after confirming different user identities, thereby ensuring vehicle safety and the legitimacy of user access rights.

[0154] In addition, an embodiment of the present application further provides a vehicle control device, the vehicle control device comprising:

[0155] A collection module, used to collect user location trajectories of preset users outside the vehicle;

[0156] An identification module is configured to identify the intent of the user based on the user's location trajectory based on a pre-trained intent identification model, and obtain an intent identification result. The intent identification model is trained based on a pre-acquired historical user trajectory set and historical user behavior set.

[0157] A control module is used to control the vehicle according to the intention recognition result.

[0158] For the principle and implementation process of vehicle control implemented in this embodiment, please refer to the above embodiments and will not be repeated here.

[0159] In addition, an embodiment of the present application also proposes a terminal device, which includes a memory, a processor, and a vehicle control program stored in the memory and runnable on the processor. When the vehicle control program is executed by the processor, the steps of the vehicle control method described above are implemented.

[0160] Since the vehicle control program adopts all the technical solutions of all the aforementioned embodiments when executed by the processor, it has at least all the beneficial effects brought by all the technical solutions of all the aforementioned embodiments, which will not be described one by one here.

[0161] In addition, an embodiment of the present application further proposes a computer-readable storage medium, on which a vehicle control program is stored. When the vehicle control program is executed by a processor, the steps of the vehicle control method described above are implemented.

[0162] Since the vehicle control program adopts all the technical solutions of all the aforementioned embodiments when executed by the processor, it has at least all the beneficial effects brought by all the technical solutions of all the aforementioned embodiments, which will not be described one by one here.

[0163] Compared to existing technologies, the vehicle control method, apparatus, terminal device, and storage medium proposed in the embodiments of this application collect the user location trajectory of a preset user outside the vehicle; perform intent recognition based on the user location trajectory based on a pre-trained intent recognition model, obtaining an intent recognition result. The intent recognition model is trained based on a pre-acquired set of historical user trajectories and historical user behavior; and vehicle control is performed based on the intent recognition result. This solution, based on seamless vehicle control tailored to user intent, effectively addresses the problem of poor user experience caused by vehicle control not tailored to user intent, thereby improving the accuracy and convenience of seamless vehicle control.

[0164] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0165] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0166] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0167] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A vehicle control method, wherein: The vehicle control method comprises the following steps: Collect user location trajectories of preset users outside the vehicle; Based on a pre-trained intent recognition model, intent recognition is performed according to the user location trajectory to obtain an intent recognition result, wherein the intent recognition model is trained based on a pre-acquired historical user trajectory set and a historical user behavior set; The vehicle is controlled according to the intention recognition result.

2. The vehicle control method according to claim 1, wherein: The step of collecting the user location trajectory of the preset user outside the vehicle includes: The user location trajectory data outside the vehicle is collected by the vehicle's built-in sensors from a device with a signal receiving and transmitting function to form the user location trajectory.

3. The vehicle control method according to claim 1, wherein: The step of collecting the user location trajectory of the preset user outside the vehicle includes: The location information of authorized users outside the vehicle is collected through the Internet of Vehicles to form the user location trajectory.

4. The vehicle control method according to claim 1, wherein: Before the step of performing intent recognition based on the user location trajectory based on the pre-trained intent recognition model and obtaining the intent recognition result, the following steps are further included: Build an initial intent recognition model based on the preset intent recognition algorithm; The initial intent recognition model is trained based on the previously acquired historical user trajectory set and historical user behavior set to obtain an intent recognition model.

5. The vehicle control method according to claim 4, wherein: The step of training the initial intent recognition model based on the pre-acquired historical user trajectory set and historical user behavior set to obtain the intent recognition model includes: Constructing a behavior area map set based on the historical user trajectory set and the historical user behavior set; The behavior area atlas is input into the initial intention recognition model for model training to obtain an intention recognition model.

6. The vehicle control method according to claim 5, wherein: The step of constructing a behavior area atlas according to the historical user trajectory set and the historical user behavior set includes: Obtaining a user behavior type based on the historical user behavior set; Classifying the historical user trajectory set according to the user behavior type to obtain a behavior trajectory set; Performing region calibration according to the behavior trajectory set to obtain a behavior region set; The behavior area set is mapped to a preset vehicle perception map to obtain a behavior area map set.

7. The vehicle control method according to claim 1, wherein: The step of performing intent recognition based on the user location trajectory based on the pre-trained intent recognition model and obtaining the intent recognition result includes: Based on a preset behavior area map, obtaining a user location set according to the user location trajectory; The user location set is input into the intent recognition model to perform intent recognition and obtain an intent recognition result.

8. The vehicle control method according to claim 7, wherein: The step of obtaining a user location set according to the user location trajectory based on the preset behavior area map includes: Based on the preset behavior area map, the user location trajectory is mapped to the corresponding behavior area, and the user location information in different behavior areas is obtained to form the user location set.

9. The vehicle control method according to claim 7, wherein: The step of inputting the user location set into the intent recognition model to perform intent recognition and obtaining an intent recognition result comprises: A continuous location trajectory curve is formed according to the location points in the user location set, and a recognition algorithm is used to identify whether the location trajectory curve conforms to a characteristic location curve of a specific behavior intention, thereby obtaining an intention recognition result.

10. The vehicle control method according to claim 7, wherein: The step of inputting the user location set into the intent recognition model to perform intent recognition and obtaining an intent recognition result comprises: According to the attribute characteristics of the location points in the user location set, the connection between different location points is determined, intention recognition is performed, and an intention recognition result is obtained.

11. The vehicle control method according to claim 7, wherein: The step of inputting the user location set into the intent recognition model to perform intent recognition and obtaining an intent recognition result comprises: The user location set is input into the intent recognition model for the following processing: Obtaining a current location point and a previous location point according to the user location set; Obtaining a current behavior type and a previous behavior type according to the behavior types corresponding to the current location point and the previous location point in a preset behavior area map; If the current behavior type is consistent with the previous behavior type and the stay time at the current location point exceeds a preset time threshold, the current behavior recognition type is obtained as an intention recognition result.

12. The vehicle control method according to any one of claims 1 to 11, wherein: The vehicle control method further includes: Perform user identification through a preset sensor and obtain a user identification result; If the user identification result is the first preset user, the vehicle control includes at least one or more of turning on a vehicle search indicator light, unlocking a door, and opening a trunk; If the user identification result is a second preset user, the vehicle control at least includes unlocking the passenger door.

13. A vehicle control device, wherein: The device comprises: A collection module, used to collect user location trajectories of preset users outside the vehicle; an identification module, configured to identify intent based on the user's location trajectory based on a pre-trained intent identification model, and obtain an intent identification result, wherein the intent identification model is trained based on a pre-acquired historical user trajectory set and historical user behavior set; A control module is used to control the vehicle according to the intention recognition result.

14. A terminal device, wherein: The terminal device includes: a memory, a processor, and a vehicle control program stored in the memory and executable on the processor, wherein the vehicle control program is configured to implement the steps of the vehicle control method according to any one of claims 1 to 12.

15. A storage medium, wherein: The storage medium stores a vehicle control program, which, when executed by a processor, implements the steps of the vehicle control method according to any one of claims 1 to 12.

Images