Vehicle entering intention detection method and system

Abstract

The invention provides an entering intention detection method and system for a vehicle, and the method comprises the steps: obtaining the sensor data of the vehicle in response to the detection that an intelligent key is in an effective unlocking range relative to the vehicle, and enabling the sensor data to be associated with a user carrying the intelligent key; detecting and tracking the user based on the acquired sensor data to predict an entry intent of the user; and determining whether to control the vehicle unlocking based on the user's entry intent. In addition, the invention also provides a vehicle supporting the detection of the entry intention.

Description

Technical Field

[0001] The present invention relates to the field of intelligent cockpits, and more particularly, to a method and system for detecting a vehicle entry intention for keyless entry into a vehicle. Background Art

[0002] In recent years, with the development of vehicle intelligence, the keyless entry system for automobiles has also been developing rapidly. From traditional mechanical keys to electronic keys, from remote control keys to keyless entry systems (PEPS), automobile keys have continuously undergone changes and innovations, and the overall safety and convenience performance of automobiles have also been continuously enhanced.

[0003] The keyless entry system (PEPS) with traditional physical keys has been mature and stable and has been widely applied to multiple vehicle models. Users do not need to manually operate the key. They can automatically unlock the vehicle just by carrying the key close to the vehicle and automatically lock the vehicle when they move away from the vehicle. With the popularization of digital car keys, the digital key keyless entry technology based on Bluetooth is being widely studied and applied by automobile manufacturers to some vehicle models. Car owners only need to carry their mobile phones. They do not need to open the mobile phone App or mini-program. They can control unlocking or locking when approaching or moving away from the vehicle, greatly improving the vehicle control experience.

[0004] However, there may still be some problems with the current keyless entry technology. For example, when a user passes by the vehicle after locking the vehicle and does not want to unlock the vehicle, the vehicle may be accidentally unlocked due to the activation of the smart key (because the user is within the unlocking range of the vehicle).

[0005] Therefore, in order to avoid accidental unlocking of the vehicle and further improve the user experience, it is desirable to analyze the user's entry intention to determine whether the user really wants to enter the vehicle. Therefore, it is desirable to provide a vehicle entry intention detection solution for keyless entry into a vehicle. Summary of the Invention

[0006] The summary of the invention is provided to introduce some concepts that will be further described in the following detailed description in a simplified form. The summary of the invention is not intended to identify the key features or essential features of the claimed subject matter, nor is it intended to be used to help determine the scope of the claimed subject matter.

[0007] In view of the above problems, according to a first aspect of the present invention, there is provided a method for detecting a vehicle entry intention, the method comprising: in response to detecting that a smart key is within an effective unlocking range relative to the vehicle, obtaining sensor data of the vehicle, the sensor data being associated with a user carrying the smart key; detecting and tracking the user based on the obtained sensor data to predict the user's entry intention; and determining whether to control the vehicle to unlock based on the user's entry intention.

[0008] In the technical solution of the embodiment of the present invention, by integrating vehicle sensor detection technology and combining user detection and tracking to identify and analyze user intentions, it is possible to avoid accidental unlocking of the vehicle due to the user being within the unlocking range of the vehicle (for example, in the case of simply passing by the vehicle in a narrow passage in a parking lot without intending to unlock the vehicle), making the vehicle more intelligent and further enhancing the user's vehicle control experience.

[0009] According to an embodiment of the present invention, the sensor data includes one or more of an image of the user, the distance between the user and the vehicle, the speed of the user, or the orientation of the user relative to the vehicle collected at regular time intervals.

[0010] According to a further embodiment of the present invention, the sensor data is collected via one or more of a visual sensor, a millimeter-wave radar sensor, an ultrasonic sensor, and a lidar sensor of the vehicle.

[0011] According to a further embodiment of the present invention, detecting and tracking the user based on the acquired sensor data to predict the user's entry intention further includes: preprocessing the acquired sensor data; and feeding the preprocessed sensor data into an entry intention prediction model to predict the user's entry intention, where the entry intention prediction model is trained based on machine learning.

[0012] According to a further embodiment of the present invention, the method further includes: uploading the predicted entry intention of the user, the actual entry intention of the user, and the acquired sensor data to the cloud for iterative optimization of the entry intention prediction model.

[0013] According to a further embodiment of the present invention, the detection of the smart key is based on one or more of radio frequency identification (RFID), near field communication (NFC), low power Bluetooth (BLE), and ultra-wideband (UWB) technology.

[0014] According to a further embodiment of the present invention, in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, acquiring the sensor data of the vehicle further includes: in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, acquiring the vehicle state, where the vehicle state includes the door state or the start state; and when the acquired vehicle state is that the doors are locked and the vehicle is not started, acquiring the sensor data of the vehicle.

[0015] According to a second aspect of the present invention, there is provided an entering intention detection system for a vehicle, the system comprising: a data acquisition module configured to acquire sensor data of the vehicle in response to detecting that a smart key is within an effective unlocking range relative to the vehicle, the sensor data being associated with a user carrying the smart key; an intention prediction module configured to detect and track the user based on the acquired sensor data to predict the user's entering intention; and an unlocking control module configured to determine whether to control the vehicle to unlock based on the user's entering intention.

[0016] According to an embodiment of the present invention, the sensor data includes one or more of an image of the user acquired at regular time intervals, the distance between the user and the vehicle, the speed of the user, or the orientation of the user relative to the vehicle.

[0017] According to a further embodiment of the present invention, the intention prediction module is further configured to: preprocess the acquired sensor data; and send the preprocessed sensor data into an entering intention prediction model to predict the user's entering intention, wherein the entering intention prediction model is trained based on machine learning.

[0018] According to a further embodiment of the present invention, the intention prediction module is further configured to: upload the predicted entering intention of the user, the actual entering intention of the user, and the acquired sensor data to the cloud for iterative optimization of the entering intention prediction model.

[0019] According to a further embodiment of the present invention, the detection of the smart key is based on one or more of radio frequency identification (RFID), near field communication (NFC), low power Bluetooth (BLE), and ultra-wideband (UWB) technology.

[0020] According to a further embodiment of the present invention, the data acquisition module is further configured to: acquire the vehicle state in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, wherein the vehicle state includes the door state or the start state; and acquire the sensor data of the vehicle when the acquired vehicle state is that the doors are locked and the vehicle is not started.

[0021] According to a third aspect of the present invention, there is provided a vehicle supporting entry intention detection, the vehicle comprising: a sensor system configured to collect sensor data of the vehicle at regular time intervals in response to detecting that a smart key is within an effective unlocking range relative to the vehicle, the sensor data being associated with a user carrying the smart key; an entry intention detection system as described in any one of the foregoing aspects; and a vehicle controller configured to perform an unlocking operation in response to determining to control the vehicle to unlock.

[0022] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium storing instructions that, when executed, cause a vehicle to perform the method as described in any one of the foregoing aspects.

[0023] In view of the problems existing in the prior art, the present invention provides a solution for supporting vehicle entry intention detection, which has at least the following advantages: by integrating vehicle sensor detection technology on the basis of the passive entry technology and judging whether the user wishes to enter the vehicle or just pass by the vehicle (e.g., through user detection and tracking) based on the processing of sensor data (e.g., using an AI model), false unlocking of the vehicle can be avoided, and the user's passive entry experience can be further improved.

[0024] These and other features and advantages will become apparent by reading the following detailed description and referring to the associated drawings. It should be understood that the foregoing general description and the following detailed description are illustrative only and do not limit the various aspects claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to understand in detail the manner in which the above-described features of the present invention are used, the above briefly summarized content may be described more specifically with reference to the embodiments, some of which are shown in the drawings. It should be noted, however, that the drawings only show some typical aspects of the present invention and should not be considered to limit its scope, as the description may allow other equally effective aspects.

[0026] Figure 1 is a diagram showing a passive entry scenario of the prior art.

[0027] Figure 2 is an example architecture diagram of an entry intention detection system for a vehicle according to an embodiment of the present invention.

[0028] Figures 3A - 3B is a diagram showing an example scenario of entry intention detection for passive entry according to an embodiment of the present invention.

[0029] Figure 4 is an example flowchart of a method for entry intention detection for a vehicle according to an embodiment of the present invention.

[0030] Figure 5 It is an exemplary architecture diagram of a deep neural network for user detection and tracking according to an embodiment of the present invention.

[0031] Figure 6 It is an exemplary vehicle that supports entry intention detection according to an embodiment of the present invention. Detailed Description of the Invention

[0032] The present invention will be described in detail below with reference to the accompanying drawings, and the features of the present invention will be further manifested in the following specific description. The term "vehicle" used throughout the specification refers to any type of automobile, including but not limited to sedans, vans, trucks, buses, etc. For simplicity, the present invention is described with respect to "automobiles". The term "A or B" used in the specification means "A and B" and "A or B", and does not mean that A and B are exclusive, unless otherwise specified.

[0033] Figure 1 FIG. 100 showing a keyless entry scenario of the prior art. As Figure 1 shown, the vehicle is equipped with a passive entry passive start (PEPS) system (e.g., a Bluetooth low energy (BLE) PEPS system). When the user carrying a smart key (e.g., which can be integrated in a smart device (e.g., a smart phone, a smart bracelet, etc.)) enters the effective unlocking range of the vehicle (e.g., within 1.5 m from the vehicle, as shown in the left figure), the vehicle door can be unlocked passively after key positioning and two-way authentication. And after the user walks out of the effective locking range of the vehicle (e.g., 2 m from the vehicle, as shown in the right figure), the vehicle door can be locked passively.

[0034] For example, for a BLE PEPS system, it locates the position of the Bluetooth smart key based on the received signal strength indication (RSSI) of BLE. Specifically, when the BLE master module arranged on the vehicle detects the Bluetooth smart key, it actively wakes up the BLE slave module. The BLE slave module communicates with the Bluetooth smart key to obtain the RSSI value corresponding to the Bluetooth smart key, and transmits the obtained RSSI value to the BLE master module for calculating the position of the Bluetooth smart key.

[0035] However, there are some problems with the current keyless entry technology. For example, when the user parks the car in a small parking lot and then wants to walk out of the parking lot through a narrow passage, the vehicle may be unlocked accidentally when passing by the vehicle due to being too close (e.g., because the smart key is within the effective unlocking range of the vehicle), which will result in a poor passive entry experience for the user.

[0036] To solve the above problems, the embodiments of this specification envision a vehicle entry intention detection scheme for keyless entry, which detects and tracks users by combining sensor data collected by the vehicle to predict the user's entry intention, thereby avoiding misunlocking the vehicle when the user does not wish to enter the vehicle and further enhancing the user experience.

[0037] Figure 2 FIG. 4 shows an exemplary architecture diagram of an entry intention detection system 200 for a vehicle according to an embodiment of the present invention. As Figure 2 shown, the system 200 may at least include a data acquisition module 202, an intention prediction module 204, and an unlocking control module 206.

[0038] The data acquisition module 202 may acquire sensor data of the vehicle in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, where the sensor data is associated with the user carrying the smart key.

[0039] In one embodiment, the sensor data may include one or more of an image of the user, the distance of the user from the vehicle, the speed of the user, or the orientation of the user from the vehicle acquired at regular time intervals.

[0040] In one embodiment, the sensor data may be acquired via one or more of a visual sensor, a millimeter-wave radar sensor, an ultrasonic sensor, or a lidar sensor of the vehicle.

[0041] In a further embodiment, the data acquisition module 202 may also acquire the vehicle state in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, where the vehicle state includes the door state or the start state, and may acquire the sensor data of the vehicle when the acquired vehicle state is that the doors are locked and the vehicle is not started.

[0042] In other words, it is necessary to determine the vehicle state before the vehicle collects sensor data for subsequent user detection and tracking. If the doors are already in the open state at that time, there is no need to collect.

[0043] In one embodiment, the detection of the smart key may be based on one or more of radio frequency identification (RFID), near field communication (NFC), low-power Bluetooth (BLE), or ultra-wideband (UWB) technology. The smart key may be a physical key with a smart chip or a virtual key integrated in a smart device (e.g., a smart phone, a smart bracelet, etc.).

[0044] The intention prediction module 204 may detect and track the user based on the acquired sensor data to predict the user's entry intention.

[0045] In one embodiment, the intent prediction module 204 may preprocess the acquired sensor data and feed the preprocessed sensor data into an entry intent prediction model to predict the user's entry intent, where the entry intent prediction model may be trained based on machine learning.

[0046] The intent prediction model may include a user detector and a user tracker. For example, when the acquired sensor data is an image sequence containing a user, the user can be detected and tracked to identify the user's position (or location), movement direction, etc. to facilitate subsequent identification and analysis of the user's entry intent.

[0047] In some examples, a machine learning-based classifier (e.g., a deep neural network, as further described below in Figure 5 ) can be used as the user detector to perform object detection (i.e., user detection) on each image in the image sequence. The images can be preprocessed before being provided to the user detector. For example, the height and / or width of the images can be adjusted or scaled. The user detector can then perform user detection on the preprocessed images.

[0048] One or more user trackers (such as a Kalman filter, an extended Kalman filter, a particle filter, or a combination thereof) can then be used to perform subsequent and continuous tracking of the user in one or more subsequent images (after the image on which user detection is performed) in the image sequence. Subsequently, the user's entry intent (wanting to enter the vehicle or just passing by the vehicle) can be determined, for example, by analyzing the user's travel path relative to the vehicle (by detecting the user's movement direction, position, etc.).

[0049] Of course, it can be understood that any other suitable sensor data can also be collected, and any other suitable techniques or models known to those skilled in the art can be used to perform vehicle entry intent prediction for keyless entry.

[0050] In a further embodiment, the intent prediction module may upload the predicted entry intent of the user, the actual entry intent of the user, and the acquired sensor data to the cloud for iterative optimization of the entry intent prediction model.

[0051] The unlocking control module 106 can then determine whether to control the vehicle to unlock based on the user's entry intent.

[0052] Those skilled in the art can understand that the entry intent prediction system of the present invention can be implemented in either hardware form or software form, and the modules can be combined or combined in any suitable manner.

[0053] Figures 3A - 3BA diagram showing example scenarios 300 and 302 for entry intention detection for keyless entry into a vehicle according to an embodiment of the present invention.

[0054] In scenario 300, the vehicle is parked by the roadside and the doors are locked. A user is passing by the vehicle while carrying a smartphone (integrated with a smart key). Since the user is within the effective unlocking range of the vehicle (e.g., within the Bluetooth communication range), the acquisition of vehicle sensor data is activated.

[0055] The vehicle then collects sensor data at regular time intervals (including, for example, an image of the user, the user's speed, the distance of the user from the vehicle, the orientation, etc.), and after preprocessing these data, sends them to an entry intention prediction model (e.g., it can be a machine learning-based AI model) to obtain an entry intention regarding whether the user wishes to enter the vehicle. In this scenario, since the user's movement path / direction is parallel to or gradually away from the vehicle, which indicates a relatively high possibility that the user is just passing by the vehicle, it is estimated that the user's entry intention is not to enter the vehicle. Therefore, no unlocking instruction is issued and the vehicle remains locked when the user passes by, thus avoiding accidental unlocking of the vehicle.

[0056] Similarly, in scenario 302, the vehicle is parked by the roadside and the doors are locked. A user is passing by the vehicle while carrying a smartphone integrated with a smart key. Since the user is within the effective unlocking range of the vehicle, the acquisition of vehicle sensor data is activated.

[0057] The vehicle then collects sensor data at regular time intervals, and after preprocessing these data, sends them to an entry intention prediction model to obtain an entry intention regarding whether the user wishes to enter the vehicle. In this scenario, since the user's movement path / direction is towards the vehicle, which indicates a relatively high possibility that the user wishes to enter the vehicle, it is estimated that the user's entry intention is to enter the vehicle. Therefore, an unlocking instruction is issued to unlock the doors without a key, thus further enhancing the user's keyless entry experience.

[0058] Figure 4 A diagram showing an example flowchart of an entry intention detection method 400 for a vehicle according to an embodiment of the present invention. Method 400 starts at step 402. The data acquisition module 102 can obtain the sensor data of the vehicle in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, and these sensor data are associated with the user carrying the smart key.

[0059] In some examples, the sensor data may include one or more of an image of the user, the distance of the user from the vehicle, the user's speed, or the orientation of the user from the vehicle collected at regular time intervals.

[0060] In some examples, the detection of the smart key can be based on one or more of RFID, NFC, BLE, or UWB technologies.

[0061] Further, the data acquisition module 102 can acquire the vehicle state in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, where the vehicle state includes the door state or the start state, and can acquire the sensor data of the vehicle when the acquired vehicle state is that the doors are locked and the vehicle is not started.

[0062] In step 404, the intention prediction module 104 can detect and track the user based on the acquired sensor data to predict the user's entry intention.

[0063] In one embodiment, the intention prediction module 104 can preprocess the acquired sensor data and send the preprocessed sensor data into an entry intention prediction model to predict the user's entry intention, where the entry intention prediction model can be trained based on machine learning.

[0064] As mentioned above, the intention prediction model can include a user detector and a user tracker (e.g., which can be trained based on machine learning) for detecting the user's position / movement direction or path for subsequent analysis and recognition of the user's intention to enter the vehicle. For example, when it is determined that the user's movement path is parallel to or away from the vehicle (as Figure 3A ), it is determined that the user is more likely to just pass by the vehicle, and when it is determined that the user's movement path is towards the vehicle (as Figure 3B ), it is determined that the user is more likely to want to unlock and enter the vehicle.

[0065] Further preferably, the intention prediction module 104 can upload the predicted entry intention of the user, the actual entry intention of the user, and the acquired sensor data to the cloud for further iterative optimization of the entry intention prediction model.

[0066] In step 406, the unlocking control module 106 can determine whether to control the vehicle to unlock based on the user's entry intention.

[0067] Specifically, when the user's entry intention is not to enter the vehicle, the vehicle remains in the locked state, and when the user's entry intention is to enter the vehicle, an unlocking instruction is issued to control the doors to unlock.

[0068] Figure 5 An example architecture diagram of a deep neural network 500 for user detection and tracking according to an embodiment of the present invention is shown.

[0069] The deep neural network 500 includes an input layer 502 configured to receive input data, such as a preprocessed image on which user detection is to be performed. In one example, the input layer 502 may include data representing the pixels of the input image or video frame. The neural network 500 includes a plurality of hidden layers 504a, 504b to 504n. The hidden layers 504a, 504b to 504n include "n" hidden layers, where "n" is an integer greater than or equal to 1. The number of hidden layers can be made to include as many layers as required for a given application. The neural network 500 further includes an output layer 506 that provides an output generated by the processing performed by the hidden layers 504a, 504b to 504n. In one example, the output layer 506 may provide a classification of the objects in the image or input video frame. The classification may include a category identifying the type of object (e.g., a person or other object).

[0070] The neural network 500 may include any suitable deep network. One example includes a convolutional neural network (CNN), which includes an input layer and an output layer, with a plurality of hidden layers therebetween. The hidden layers of the CNN may include, for example, a series of convolutional layers, non-linear layers, pooling layers (for downsampling), and fully connected layers. Additionally, the neural network 500 may also include and is not limited to any other deep network other than the CNN, such as an autoencoder, a deep belief network (DBN), a recurrent neural network (RNN), etc.

[0071] Figure 6 An exemplary vehicle 600 that supports entry intent detection according to an embodiment of the present invention is shown. The vehicle 600 may include various software and hardware components connected via a bus 602. For example, the vehicle 600 may include one or more processors 604 and a memory 606. The memory 606 may include RAM, ROM, or a combination thereof. The memory 606 may store computer-executable instructions that, when executed by the one or more processors 604, cause the one or more processors 604 to perform the various functions described herein (e.g., with respect to Figure 4 the functions described).

[0072] The one or more processors 604 may include a CPU, which may be a multi-core CPU in some examples. The instructions executed at the CPU may be loaded, for example, from a program memory associated with the CPU or may be loaded from the memory 606. The one or more processors 604 may further include additional processing components customized for specific functions, such as a graphics processing unit (GPU), a digital signal processor (DSP), a neural processing unit (NPU), a multimedia processing unit. In some examples, the one or more processors may be based on the ARM or RISC-V instruction set.

[0073] An NPU is generally configured as a dedicated circuit for implementing the control and arithmetic logic for executing machine learning algorithms, such as algorithms for processing artificial neural networks (ANNs), deep neural networks (DNNs), random forests (RFs), etc. An NPU may sometimes be alternatively referred to as a neural signal processor (NSP), a tensor processing unit (TPU), a neural network processor (NNP), an intelligent processing unit (IPU), or a vision processing unit (VPU). The NPU can be configured to accelerate the execution of common machine learning tasks, such as image classification and various other prediction tasks. In some examples, multiple NPUs may be instantiated on a single chip, such as a system-on-chip (SoC), while in other examples, multiple NPUs may be part of a dedicated neural network accelerator. The NPU can be optimized for training or inference, or in some cases may be configured to balance the performance between training and inference. For an NPU capable of performing both training and inference, these two tasks may generally still be executed independently. An NPU designed to accelerate training is generally configured to accelerate the optimization of a new model, which is a highly computationally intensive operation involving inputting an existing dataset (often labeled or tagged), iterating over the dataset, and then adjusting model parameters, such as weights and biases, to improve model performance. Generally, optimizing based on incorrect predictions involves backpropagating through the layers of the model and determining gradients to reduce the prediction error. An NPU designed to accelerate inference is generally configured to operate on a trained model. Such an NPU can thus be configured to input new data segments and quickly process the data segments through the already trained model to generate a model output (e.g., an inference).

[0074] In some cases, the memory 606 may particularly include a BIOS that can control basic hardware or software operations, such as interactions with peripheral components or devices. The processor 604 may include intelligent hardware devices (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, discrete gate or transistor logic components, discrete hardware components, or any combination thereof).

[0075] Vehicle 600 may include a sensor system 608 that may be used to collect sensor data of the vehicle at regular time intervals in response to detecting that a smart key is within an effective unlocking range relative to the vehicle, and this sensor data is associated with a user carrying the smart key. The sensor system 608 may include one or more cameras (e.g., a front camera, a rear camera, side mirrors mounted on the sides of the vehicle), and the cameras can provide object detection and distance estimation. For example, a sequence of images collected by the cameras may be fed into system 100 to detect and track the user for predicting the user's intention to enter the vehicle. The sensor system 608 may also include millimeter-wave radar, ultrasonic radar, lidar LIDAR, etc. For example, lidar can use pulsed lasers to measure the range to an object. Although the (s) cameras can provide object detection, lidar can provide a means to more deterministically detect the distance (and orientation) of an object. Lidar measurements can also be used to estimate travel speed, vector direction, relative position, stopping distance, etc. by providing accurate distance measurements and incremental distance measurements.

[0076] In addition, vehicle 600 may further include the above-mentioned entry intention detection system 100, which may perform the following functions: in response to detecting that a smart key is within an effective unlocking range relative to the vehicle, obtain the sensor data of the vehicle, and this sensor data is associated with a user carrying the smart key; detect and track the user based on the obtained sensor data to predict the user's entry intention; and determine whether to control the vehicle to unlock based on the user's entry intention.

[0077] In addition, vehicle 600 may further include a vehicle controller 610, which may perform an unlocking operation in response to determining to control the vehicle to unlock. In addition, vehicle 600 may further include one or more wireless transceivers 612 for transmitting and receiving data via various means, protocols, and standards. In some embodiments, the wireless transceiver 612 may be configured to transmit and receive data messages and elements via a short-range wireless communication protocol (e.g., etc.) and / or via a local area network and / or a wide area network, and / or via a cellular network, and / or via any suitable wireless network. Of course, it should be understood that these are merely examples of networks that vehicle 600 can utilize over a wireless link, and the claimed subject matter is not limited in this regard. In addition, vehicle 600 may further include a global navigation satellite system (GNSS) receiver 614. The GNSS receiver 614 may be configured to receive and digitally process signals from navigation satellites (and / or other vehicles) to provide the positioning, speed, and time of the receiver. The GNSS receiver 614 may include hardware and / or software components.

[0078] The functions described herein can be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions can be stored on or transmitted via a computer-readable medium as one or more instructions or code. Other examples and implementations fall within the scope of the present disclosure and the appended claims. For example, due to the nature of software, the functions described herein can be implemented using software, hardware, firmware, hardwiring, or any combination thereof executed by a processor. The features implementing the functions can also be physically located in various positions, including being distributed such that portions of the functions are implemented at different physical locations.

[0079] The foregoing has included examples of aspects of the claimed subject matter. Of course, it is not possible to describe every conceivable combination of components or methods for purposes of describing the claimed subject matter, but one of ordinary skill in the art will recognize that many further combinations and permutations of the claimed subject matter are possible. Accordingly, the claimed subject matter is intended to cover all such alterations, modifications, and variations that fall within the spirit and scope of the appended claims.

Claims

1. An entering intention detection method for a vehicle, the method comprising: In response to detecting that the smart key is within an effective unlocking range relative to the vehicle, obtaining sensor data of the vehicle, the sensor data being associated with a user carrying the smart key; Detecting and tracking the user based on the obtained sensor data to predict the user's entering intention; And Determining whether to control the vehicle to unlock based on the user's entering intention.

2. The method according to claim 1, wherein The sensor data includes one or more of an image containing the user, the distance of the user from the vehicle, the speed of the user, or the orientation of the user from the vehicle, collected at regular time intervals.

3. The method according to claim 2, wherein The sensor data is collected via one or more of a vision sensor, a millimeter-wave radar sensor, an ultrasonic sensor, and a lidar sensor of the vehicle.

4. The method according to claim 1, characterized in that Detecting and tracking the user based on the obtained sensor data to predict the user's entering intention further includes: Preprocessing the obtained sensor data; and Feeding the preprocessed sensor data into an entering intention prediction model to predict the user's entering intention, where the entering intention prediction model is trained based on machine learning.

5. The method according to claim 4, wherein The method further includes: Uploading the predicted entering intention of the user, the actual entering intention of the user, and the obtained sensor data to the cloud for iterative optimization of the entering intention prediction model.

6. The method according to claim 1, characterized in that, The detection of the smart key is based on one or more of radio frequency identification (RFID), near field communication (NFC), low power Bluetooth (BLE), and ultra-wideband (UWB) technology.

7. The method according to claim 1, characterized in that, In response to detecting that the smart key is within an effective unlocking range relative to the vehicle, obtaining the sensor data of the vehicle further includes: In response to detecting that the smart key is within an effective unlocking range relative to the vehicle, obtaining the vehicle state, where the vehicle state includes a door state or a startup state; and When the obtained vehicle state is that the doors are locked and the vehicle is not started, obtaining the sensor data of the vehicle.

8. An entering intention detection system for a vehicle, the system comprising: A data acquisition module configured to obtain sensor data of the vehicle in response to detecting that the smart key is within an effective unlocking range relative to the vehicle, the sensor data being associated with a user carrying the smart key; An intention prediction module configured to detect and track the user based on the obtained sensor data to predict the user's entering intention; And An unlocking control module configured to determine whether to control the vehicle to unlock based on the user's entering intention.

9. The system according to claim 8, wherein The sensor data includes one or more of an image containing the user, the distance of the user from the vehicle, the speed of the user, or the orientation of the user from the vehicle, collected at regular time intervals.

10. The system according to claim 8, characterized in that, The intention prediction module is further configured to: Preprocess the obtained sensor data; and The preprocessed sensor data is sent into an entry intention prediction model to predict the user's entry intention, where the entry intention prediction model is trained based on machine learning.

11. The system according to claim 10, wherein, The intention prediction module is further configured to: Upload the predicted entry intention of the user, the actual entry intention of the user, and the acquired sensor data to the cloud for iterative optimization of the entry intention prediction model.

12. The system according to claim 8, wherein The detection of the smart key is based on one or more of radio frequency identification (RFID), near field communication (NFC), low power Bluetooth (BLE), and ultra-wideband (UWB) technology.

13. The system according to claim 8, characterized in that, The data acquisition module is further configured to: In response to detecting that the smart key is within the effective unlocking range relative to the vehicle, acquire the vehicle state, where the vehicle state includes the door state or the start state; And When the acquired vehicle state is that the doors are locked and the vehicle is not started, acquire the sensor data of the vehicle.

14. A vehicle supporting entry intention detection, the vehicle comprising: A sensor system configured to collect the sensor data of the vehicle at regular time intervals in response to detecting that the smart key is within the effective unlocking range relative to the vehicle, the sensor data being associated with a user carrying the smart key; The entry intention detection system according to any one of claims 8-13; And A vehicle controller configured to perform an unlocking operation in response to determining to control the vehicle to unlock.

15. A computer-readable storage medium storing instructions that, when executed, cause a vehicle to perform the method according to any one of claims 1-7.

Images