Vehicle back door control method and vehicle

Abstract

The invention discloses a vehicle back door control method and a vehicle, and belongs to the technical field of vehicle control. According to the technical scheme provided by the embodiment of the invention, the action recognition strategy corresponding to the target vehicle is dynamically determined in combination with the vehicle state of the target vehicle and the environment condition of the position where the target vehicle is located under the condition that the user is recognized to approach the target vehicle, and the action recognition strategy is matched with the vehicle state and the environment condition. And under the condition that the action recognition strategy belongs to the first type of recognition strategy, the corresponding projection parameter is used for controlling the projection assembly to project the first pattern on the ground, and a user can control the back door through the first pattern. According to the technical scheme, the back door opening action within the first pattern indication range is recognized, the back door opening action is recognized through the action recognition strategy, the back door is controlled according to the recognition result, and therefore the requirement of a user for back door control in a specific scene is met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control, and more particularly, to a control method of a vehicle back door and a vehicle. BACKGROUND

[0002] The opening mode of a vehicle back door has evolved from traditional mechanical type to electric type, and then to induction type technology. At present, the "one kick" induction opening technology widely used in the industry detects user foot movements through sensors to trigger the opening of the back door.

[0003] Although this technology provides convenience in certain scenarios, it still has obvious limitations. For example, in rainy and snowy weather, the sensor is easily disturbed by the environment, causing false triggering and affecting user experience.

[0004] Therefore, there is an urgent need for a back door control method that adapts to more scenarios to improve the robustness and user experience of back door control. SUMMARY

[0005] The embodiments of the present application provide a control method of a vehicle back door and a vehicle, which can improve the robustness and user experience of back door control. The technical solutions are as follows: On the one hand, a control method of a vehicle back door is provided, the method comprising: In response to a user approaching a target vehicle, determining an action recognition strategy corresponding to the target vehicle based on vehicle state information of the target vehicle and environmental information of the location where the target vehicle is located; In the case where the action recognition strategy belongs to a first type of recognition strategy, controlling a projection component of the target vehicle to project a first pattern on the ground based on projection parameters corresponding to the action recognition strategy, the first type of recognition strategy being a strategy for recognizing a back door opening action using the first pattern; In response to a back door opening action within the first pattern indication range, recognizing the back door opening action using the action recognition strategy, and controlling the back door of the target vehicle according to the recognition result.

[0006] In one possible implementation, the action recognition strategy of the target vehicle is determined based on the vehicle state information of the target vehicle and the environmental information of the location where the target vehicle is located; Based on the vehicle state information, determining a vehicle inclination angle of the target vehicle; Based on the environmental information, determining a weather type and a road surface type of the location where the target vehicle is located; Based on the vehicle inclination angle, the weather type, and the road surface type, determining the action recognition strategy of the target vehicle.

[0007] In this implementation, the action recognition strategy of the target vehicle is determined in combination with the vehicle inclination, the weather type and the road surface type, the action recognition strategy matches the vehicle state of the target vehicle and the environmental state of the location, and the subsequent accurate control of the tailgate can be realized by using the action recognition strategy.

[0008] In a possible implementation, the action recognition strategy of the target vehicle is determined based on the vehicle inclination, the weather type and the road surface type, and the action recognition strategy of the target vehicle is determined based on the vehicle inclination or the road surface type. In a case where the weather type belongs to the first type of weather, the action recognition strategy of the target vehicle is determined based on the vehicle inclination or the road surface type. In a case where the weather type belongs to the second type of weather, the action recognition strategy of the target vehicle is determined as a first recognition strategy, the first recognition strategy belongs to a second type of recognition strategy, the second type of recognition strategy is a strategy of not performing the tailgate opening action recognition by using the first pattern, and the second type of weather is more severe than the first type of weather.

[0009] In this implementation, since the weather type has an important influence on the projection effect of the projection component, the determination of the action recognition strategy is performed according to whether the weather type belongs to the first type of weather or the second type of weather, whether the projection component associated manner is used for subsequent recognition is determined, and thus the adaptive determination of the recognition strategy is realized.

[0010] In a possible implementation, the action recognition strategy of the target vehicle is determined based on the vehicle inclination or the road surface type, and the action recognition strategy of the target vehicle is determined based on the vehicle inclination or the road surface type. In a case where the vehicle inclination is greater than a first preset inclination, the action recognition strategy of the target vehicle is determined as a second recognition strategy, and the second recognition strategy is a repeated action recognition strategy. In a case where the vehicle inclination is less than a second preset inclination, the action recognition strategy of the target vehicle is determined as a third recognition strategy, and the third recognition strategy is a preset recognition strategy. In a case where the road surface type belongs to a preset road surface type, the action recognition strategy of the target vehicle is determined as a fourth recognition strategy, and the fourth recognition strategy is a time-delayed action recognition strategy. The tailgate recognition strategies corresponding to the second recognition strategy, the third recognition strategy and the fourth recognition strategy all belong to the first type of recognition strategy.

[0011] In this implementation, further strategy determination is realized in combination with the vehicle inclination or the road surface type, so that the determined action recognition strategy is more matched with the vehicle state or the environmental state. In this implementation, further strategy determination is realized in combination with the vehicle inclination or the road surface type, so that the determined action recognition strategy is more matched with the vehicle state or the environmental state.

[0012] In a possible implementation, before the projection component of the target vehicle projects the first pattern on the ground based on the projection parameter corresponding to the action recognition strategy, the method further includes: In a case where the action recognition strategy of the target vehicle is the second recognition strategy, the projection parameter is determined based on the vehicle inclination, the road surface type, and the distance between the projection component and the ground; In a case where the action recognition strategy of the target vehicle is the third recognition strategy, a preset projection parameter is determined as the projection parameter; In a case where the action recognition strategy of the target vehicle is the fourth recognition strategy, a target projection size and a target projection brightness of a projection pattern are determined based on the road surface type, and the projection parameter is determined based on the target projection size and the target projection brightness.

[0013] In this implementation, in a case where the action recognition strategy is the second recognition strategy, the projection parameter is determined based on the vehicle inclination, the road surface type, and the distance between the projection component and the ground, the projection parameter is adapted to the vehicle state and the environmental state of the target vehicle, and a projection performed by the projection component based on the projection parameter can achieve a better projection effect. In a case where the action recognition strategy is the preset recognition strategy, a preset projection parameter corresponding to the preset recognition strategy is determined as the projection parameter, and a better projection effect can be achieved. In a case where the action recognition strategy is the fourth recognition strategy, the target projection size and the target projection brightness are determined based on the road surface type, and the projection parameter is subsequently determined based on the target projection size and the target projection brightness, and the matching degree between the projection parameter and the fourth recognition strategy is relatively high.

[0014] In a possible implementation, the projection component of the target vehicle projects the first pattern on the ground based on the projection parameter corresponding to the action recognition strategy, and the method includes: The component control parameter of the projection component is determined based on the projection parameter; The projection component is controlled to project the first pattern on the ground by using the component control parameter.

[0015] In a possible implementation, the rear door opening action is recognized based on the action recognition strategy, and the rear door of the target vehicle is controlled according to the recognition result, and the method includes: In a case where the action recognition strategy is the second recognition strategy, the execution times of the rear door opening action in the first pattern indication range are recognized, the rear door is controlled to be opened in a case where the execution times are greater than or equal to a preset number of times, and the rear door is not controlled to be opened in a case where the execution times are less than the preset number of times. In a case where the action recognition strategy is a third recognition strategy, it is determined whether the back door opening action is located in a preset position range of the first pattern, and in a case where the back door opening action is located in the preset position range, the back door is controlled to be opened, and in a case where the back door opening action is not located in the preset position range, the back door opening action is not responded to. In a case where the action recognition strategy is a fourth recognition strategy, it is determined for how long the back door opening action lasts within the first pattern indication range, and in a case where the duration is greater than or equal to a preset duration, the back door is controlled to be opened, and in a case where the duration is less than the preset duration, the back door opening action is not responded to.

[0016] In this implementation, in a case where the action recognition strategy is a second recognition strategy, whether to respond to the back door opening action is determined by counting the number of times of execution of the back door opening action, and the accuracy of action response is relatively high. Since the vehicle inclination corresponding to the second recognition strategy is relatively large, through repeated action recognition, the accuracy of action recognition can be improved, and the user is reminded to pay attention to the situation that the objects in the target vehicle fall or the situation that the objects move forward is prevented. In a case where the action recognition strategy is a third recognition strategy, whether to respond to the back door opening action is determined by judging whether the back door opening action is located in a preset position range of the first pattern, and the accuracy of action response is relatively high. In a case where the action recognition strategy is a fourth recognition strategy, whether to respond to the back door opening action is determined by judging the duration of the back door opening action, the influence of the road surface with a relatively low flatness is eliminated as much as possible, and the accuracy of action response is improved.

[0017] In a possible implementation, the method further includes: In response to detecting a vehicle key of the target vehicle, it is determined that a user approaches the target vehicle; Alternatively, in response to detecting a target mobile terminal, it is determined that a user approaches the target vehicle, the target mobile terminal being a mobile terminal having a binding relationship with the target vehicle; Alternatively, in response to identifying a target user from the environment information, it is determined that a user approaches the target vehicle, the target user being a user having a binding relationship with the target vehicle.

[0018] In a possible implementation, the method further includes: In a case where the action recognition strategy belongs to a second type of recognition strategy, a projection component of the target vehicle is controlled to project a second pattern on the ground, and the second pattern is used to prompt that the back door of the target vehicle cannot be opened by a pattern projected by the projection component.

[0019] In this implementation, in the case that the back door cannot be controlled through the first type of identification strategy, the projection component is controlled to project a second pattern to prompt the user to control the back door in other ways, so as to improve the efficiency of human-computer interaction.

[0020] In an aspect, a control apparatus of a back door of a vehicle is provided, and the apparatus comprises: a strategy determination module configured to determine an action identification strategy corresponding to a target vehicle based on vehicle state information of the target vehicle and environment information of a location where the target vehicle is located in response to user approaching the target vehicle; a projection control module configured to control a projection component of the target vehicle to project a first pattern on the ground based on projection parameters corresponding to the action identification strategy in the case that the action identification strategy belongs to a first type of identification strategy, the first type of identification strategy being a strategy of identifying a back door opening action by using the first pattern; a back door control module configured to identify the back door opening action by using the action identification strategy in response to a back door opening action within an indication range of the first pattern, and control a back door of the target vehicle according to an identification result.

[0021] In an aspect, a vehicle is provided, and the vehicle comprises one or more processors and one or more memories, the one or more memories storing at least one program code, the program code being loaded and executed by the one or more processors to implement operations performed by a control method of a back door of the vehicle.

[0022] In an aspect, a computer-readable storage medium is provided, and the computer-readable storage medium stores at least one program code, the program code being loaded and executed by a processor to implement operations performed by a control method of a back door of a vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a schematic diagram of an implementation environment of a control method of a back door of a vehicle provided by an embodiment of the present application; Figure 2 is a flowchart of a control method of a back door of a vehicle provided by an embodiment of the present application; Figure 3 is a flowchart of another control method of a back door of a vehicle provided by an embodiment of the present application; Figure 4 is a schematic diagram of a target vehicle projecting a pattern on the ground provided by an embodiment of the present application; Figure 5 is a structural schematic diagram of a control apparatus of a back door of a vehicle provided by an embodiment of the present application; Figure 6FIG. 1 is a structural schematic diagram of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the present application will be described in detail below with reference to the drawings. In the description of the embodiments of the present application, unless otherwise specified, " / " represents the meaning of or, for example, A / B can represent A or B: "and / or" in the text only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which means that there are three cases of A alone, A and B together, and B alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0025] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features reflected. Therefore, the features defined with "first" and "second" can explicitly or implicitly include one or more features.

[0026] In order to describe the technical solutions provided by the embodiments of the present application, some terms related to the embodiments of the present application will be introduced.

[0027] DLP (Digital Light Processing) projector: A high-precision optical projection device based on a micro-mirror array. Its working principle is to control the flipping state of millions of micro-mirrors to reflect the light source and generate high-contrast, high-resolution images or patterns. In the control system of the rear door of the car, the DLP projector is used to project interactive AR patterns (such as circular targets, prompt areas, etc.) on the ground, and supports electric focusing (-15° to +30° angle adjustment) to adapt to different terrain and slope conditions.

[0028] Augmented Reality (AR): A technology that superimposes virtual information (such as graphics, text) onto the real environment and realizes interaction. In the embodiments of the present application, AR realizes intelligent opening control of the rear door by combining ground projection patterns (such as targets, prompt areas) with user actions (such as stepping), which improves user experience and operation convenience.

[0029] Rear door: Also known as tailgate or trunk door, it is a openable and closable structure at the rear of the vehicle, used for accessing luggage or cargo. The rear door in the embodiments of the present application supports electric opening and closing, and realizes intelligent operation through multi-mode control strategy (such as projection interaction, digital key triggering).

[0030] Back door controller: It is part of the vehicle's electronic control unit (ECU) responsible for receiving instructions from the body controller and performing actions such as unlocking, opening, and closing the back door. It works in coordination with sensors (such as inclination sensors) and actuators (such as electric support rods) to ensure operational safety and reliability.

[0031] UWB (Ultra-Wideband) digital key: It is an identity authentication and positioning system based on ultra-wideband wireless communication technology. Through centimeter-level precision (error <10 cm) three-dimensional space positioning, the UWB digital key can identify the user's approach to the vehicle and authorize the legal user to trigger the back door opening operation.

[0032] TOF (Time of Flight): It is a technology that calculates distance by measuring the time difference between the emission and reflection of light waves (usually near-infrared light). In the embodiments of the present application, TOF sensors are used to detect the interaction distance change between the user's feet and the ground projection pattern, thereby determining whether the user has performed an opening action (such as tapping the target).

[0033] In related technologies, the opening method of the vehicle back door has evolved from traditional mechanical to electric, and then to inductive technology. The technical solutions currently being researched and applied in the industry mainly include the following categories: Foot kick sensor scheme: This is currently a common inductive opening method. By installing sensors (such as capacitive, infrared sensors) in the rear bumper to detect the waveform characteristics of the user's foot kick action, the back door is triggered to open. However, it has a high risk of false triggering, unstable recognition rate, and requires the user to operate close to the tail, which needs to be improved in terms of convenience and safety.

[0034] Millimeter wave radar scheme: This scheme reuses the millimeter wave radar originally deployed for blind area detection systems (BSD) on vehicles. It detects user foot kick actions within a predetermined range and matches them with action templates in a self-learning library to determine the operation intent. Its advantage is that the millimeter wave radar has strong penetration and is less affected by environmental interference (such as rain, fog, and dust), but the technology is complex and highly dependent on algorithms.

[0035] Visual recognition scheme: It uses a camera (such as a rearview camera) to capture images of the vehicle's rear, and uses image recognition technology (such as recognizing specific body parts and gestures) to authenticate user identity and determine their opening intent. For example, some patents mention that the back door can be controlled to open by recognizing user gestures. However, its algorithm complexity is high, it is greatly affected by light conditions (performance may decrease at night or in strong light), and it may involve user privacy issues.

[0036] Most of these technical solutions require the user to perform specific limb movements (kicking, waving hands), which is still inconvenient for the user when holding heavy objects, and even poses certain safety risks (such as kicking empty or tripping over obstacles).

[0037] Based on this, the technical scheme provided by the embodiments of the present application is proposed, which dynamically determines the action recognition strategy in combination with vehicle state information and environment information, thereby realizing flexible control of the vehicle back door and meeting the control needs of users in different scenarios.

[0038] The implementation environment of the embodiments of the present application is introduced below, and the implementation environment of the control method of the vehicle back door provided by the embodiments of the present application includes a vehicle controller 101, a back door controller 102, and a projection controller 103. Figure 1

[0039] The vehicle controller 101 is a controller with data acquisition and data processing capabilities, which can control the vehicle globally. In the embodiments of the present application, the vehicle controller 101 can acquire vehicle state information collected by a vehicle state sensor and environment information collected by an environment sensor, and process the vehicle state information and environment light information. The vehicle controller 101 is connected with the back door controller 102 through a CAN bus, and the results processed by the vehicle controller 101 can be executed through the back door controller 102. In addition, the vehicle controller 101 is connected with the projection controller 103 through a CAN bus, and can control the projection component to project on the ground by controlling the projection controller 103.

[0040] The back door controller 102 is used to control the back door, including controlling the opening and closing of the back door, and further can finely control the opening angle of the back door.

[0041] The projection controller 103 is used to control the projection component of the vehicle, and the projection component can project a pattern on the ground. In the embodiments of the present application, the projection component is located at the tail of the vehicle, that is, the projection component can project a pattern at the rear of the vehicle. Since the user needs to approach the vehicle from the rear when using the trunk, the pattern is projected to the rear of the vehicle to facilitate the user to control the back door by using the pattern on the ground.

[0042] After introducing the implementation environment of the embodiments of the present application, the application scenarios of the technical scheme provided by the embodiments of the present application are introduced below.

[0043] ​The technical scheme provided by the embodiment of the application can be applied to a vehicle configured with an electric back door and a projection assembly. By using the technical scheme provided by the embodiment of the application, the action recognition strategy can be automatically determined according to the vehicle state and the environmental condition, and the projection assembly is controlled to project a pattern on the ground in a manner corresponding to the action recognition strategy. Subsequently, the back door opening action is recognized by using the action recognition strategy, and flexible control of the back door can be achieved.

[0044] After introducing the implementation environment and application scenario of the embodiment of the application, the technical scheme provided by the embodiment of the application is introduced below, referring to Figure 2 Taking the vehicle controller as an example of the execution subject, the method comprises the following steps.

[0045] 201. In response to user approaching a target vehicle, determining an action recognition strategy corresponding to the target vehicle based on vehicle state information of the target vehicle and environmental information of a location where the target vehicle is located.

[0046] The target vehicle is a vehicle to be controlled by the back door. The vehicle state information is a data set describing the static and dynamic conditions of the vehicle, and is used to reflect the state of the target vehicle. The environmental information is the perception data of the external environment where the vehicle is located, and can reflect the environmental condition of the location where the target vehicle is located. The action recognition strategy is a complete and configurable detection and verification rule adopted for opening the back door.

[0047] 202. In the case that the action recognition strategy belongs to a first type of recognition strategy, controlling a projection assembly of the target vehicle to project a first pattern on the ground based on projection parameters corresponding to the action recognition strategy. The first type of recognition strategy is a strategy for recognizing the back door opening action by using the first pattern.

[0048] The projection parameters are parameters used to control the projection assembly. The projection assembly is a device capable of projecting a pattern on the ground, such as a hardware module capable of realizing the ground projection function. The first pattern is a projection pattern projected on the ground by the projection assembly, used to guide the user to perform the back door opening action. The first type of recognition strategy refers to a recognition strategy that relies on the ground projection pattern to recognize the back door opening action. The first type of recognition strategy can include multiple action recognition strategies.

[0049] 203. In response to a back door opening action within the indication range of the first pattern, recognizing the back door opening action by using the action recognition strategy, and controlling the back door of the target vehicle according to the recognition result.

[0050] The back door opening action refers to a specific limb action performed by the user for opening the back door. In the first type of recognition strategy, the back door opening action refers to the action performed by the foot of the user on the first pattern on the ground. The recognition result is the final judgment obtained after detection and algorithm analysis of the back door opening action. The result is usually a binary judgment: it is an effective opening signal or it is not an effective opening signal. Only when the result is the former, the vehicle controller will issue an unlock instruction to the back door controller, thereby controlling the back door to open.

[0051] Through the technical solutions provided by the embodiments of the present application, in the case of recognizing that the user approaches the target vehicle, the action recognition strategy corresponding to the target vehicle is dynamically determined in combination with the vehicle state of the target vehicle and the environmental situation of the location where the target vehicle is located. The action recognition strategy is matched with the vehicle state and the environmental situation. In the case that the action recognition strategy belongs to the first type of recognition strategy, the corresponding projection parameters are used to control the projection component to project the first pattern on the ground, and the user can use the first pattern to control the back door. The back door opening action within the indication range of the first pattern is recognized, the back door opening action is recognized by using the action recognition strategy, and the back door is controlled according to the recognition result, thereby meeting the user's demand for controlling the back door in a specific scenario.

[0052] It should be noted that the above steps 201-203 are a simple description of the control method of the vehicle back door provided by the embodiments of the present application. In the following, the control method of the vehicle back door provided by the embodiments of the present application will be described in more detail in combination with some examples, see Figure 3 Taking the whole vehicle controller as an example, the method comprises the following steps.

[0053] 301、The whole vehicle controller determines whether there is a user approaching the target vehicle.

[0054] The target vehicle is a vehicle to be controlled by the back door, and the target vehicle is an electric vehicle, a fuel vehicle, or a hybrid vehicle, which is not limited in the embodiment of the application. The target vehicle has an electric back door and a projection assembly. The projection assembly is installed at the tail of the target vehicle. Since the target vehicle can have various forms (sedan, SUV, MPV, etc.), the projection assembly can be installed on the back door of the target vehicle or below the back door. The specific setting is made by personnel according to the actual situation. Alternatively, the projection assembly can be integrated into the brake light of the target vehicle, so that the projection assembly has higher integration and better working environment. The setting mode of the projection assembly is designed by the technician in combination with the actual situation of the target vehicle, which is not limited in the embodiment of the application. In the above step 301, the user refers to the user of the target vehicle, and the user of the target vehicle refers to the user who has the use right of the target vehicle, and any user approaching the target vehicle is not determined as the user approaching the target vehicle. For example, the user carrying the vehicle key of the target vehicle can be regarded as the user of the target vehicle, and the user carrying the mobile terminal having a binding relationship with the target vehicle can also be regarded as the user of the target vehicle.

[0055] In a possible implementation, in response to detecting the vehicle key of the target vehicle, the vehicle controller determines that the user approaches the target vehicle.

[0056] The vehicle key is a UWB key or a Bluetooth key. The UWB key refers to an electronic key that establishes a communication connection with the target vehicle through a UWB communication mode. The Bluetooth key refers to an electronic key that establishes a communication connection with the target vehicle through a Bluetooth connection. The vehicle key and the target vehicle have a binding relationship. The vehicle controller controls the communication unit to actively search for the vehicle key after locking the vehicle. In the case that the vehicle key enters the search range of the target vehicle, the communication unit attempts to establish a communication connection with the vehicle key. In the case that the communication unit establishes a communication connection with the vehicle key, the vehicle controller determines that the vehicle key is detected. The detection of the vehicle key indicates that the user carrying the vehicle key moves to the vicinity of the target vehicle, so it can be directly determined that the user approaches the target vehicle.

[0057] In this implementation, whether the user approaches the target vehicle is determined by detecting the vehicle key. The determination method is simple and can directly reuse the original key search process, so the implementation cost is low.

[0058] Another embodiment of the above step 301 is described below.

[0059] In a possible implementation, in response to detecting the target mobile terminal, the vehicle controller determines that the user approaches the target vehicle. The target mobile terminal is a mobile terminal having a binding relationship with the target vehicle.

[0060] The target mobile terminal can establish a direct connection with the target vehicle through wireless communication or Bluetooth communication, or can establish an indirect connection with the target vehicle through a server. In the above embodiment, detecting the target mobile terminal means that the target vehicle and the target mobile terminal establish a direct connection, that is, a connection through wireless connection or Bluetooth connection. In the case where the target vehicle and the target mobile terminal establish a direct connection, it indicates that the user carrying the target mobile terminal moves to the vicinity of the target vehicle, and at this time, it can be directly judged that the user is close to the target vehicle.

[0061] In this embodiment, the judgment of whether the user is close to the target vehicle is displayed by detecting the mobile terminal having a binding relationship with the target vehicle, and the judgment process is relatively simple and has high accuracy.

[0062] Another embodiment of the above step 301 will be described below.

[0063] In a possible embodiment, in response to identifying the target user from the environment information, the vehicle controller determines that the user is close to the target vehicle, and the target user is a user having a binding relationship with the target vehicle.

[0064] The environment information includes an environment image, and identifying the target user from the environment information means identifying the target user from the environment image. The user image or user feature of the target user is stored in the vehicle controller, and of course, the storage and use of the user image and user feature need to be authorized by the target user. The target vehicle includes a plurality of cameras and a plurality of radars, and the plurality of cameras and the plurality of radars are distributed around the target vehicle. In the case where the target vehicle is locked, most of the plurality of cameras are in a standby state, and a small part of the plurality of cameras are in a working state. The plurality of radars operate in a low-power mode, and in the case where any radar detects a moving target, the camera corresponding to the radar is started and performs image acquisition to obtain an environment image, and the camera corresponding to the radar refers to the camera on the same side as the camera.

[0065] In this embodiment, the environment information is used to identify the target user, and in the case where the user does not carry the vehicle key or the target mobile terminal, the target vehicle can also identify that the user is close to the target vehicle, and the user identification method is more diverse.

[0066] The way of identifying the target user from the environment information in the above embodiment will be described below.

[0067] In some embodiments, the vehicle controller performs target detection on the environment image to obtain a target image region in the environment image, the target image region being an image region where the user is located. The vehicle controller performs feature extraction on the target image region to obtain a user feature corresponding to the target image region. The vehicle controller compares the user feature with a plurality of candidate user features, and in a case where the user feature matches any candidate user feature of the plurality of candidate user features, determines that the target user is recognized. In a case where the user feature does not match any candidate user feature of the plurality of candidate user features, it is determined that the target user is not recognized.

[0068] The plurality of candidate user features are user features of different target users, and the plurality of candidate user features are stored in advance.

[0069] 302、In response to the user approaching the target vehicle, the vehicle controller determines an action recognition strategy corresponding to the target vehicle based on vehicle state information of the target vehicle and environment information of a location where the target vehicle is located.

[0070] The vehicle state information is a set of data describing static and dynamic conditions of the vehicle itself, and is used to reflect the state of the target vehicle. The environment information is perception data of an external environment where the vehicle is located, and can reflect the environmental situation of the location where the target vehicle is located. The vehicle state information and the environment information can be acquired in real time or after detecting that the user approaches the target vehicle, and the embodiments of the present application do not limit this. The action recognition strategy is a complete and configurable detection and verification rule adopted for opening the back door. In some embodiments, the vehicle state information is collected by a vehicle state sensor, and in a case where the vehicle state information includes a vehicle inclination angle, the vehicle state sensor is a gyroscope. The environment information includes weather information and road surface information, the weather information is used to reflect the weather condition, and the road surface information is used to reflect the road surface condition. In some embodiments, the road surface information is obtained based on the location of the target vehicle, or is obtained through road surface image or road surface point cloud recognition. The weather information is obtained based on the location of the target vehicle, or is obtained through a weather sensor of the target vehicle, for example, the weather sensor includes a rain sensor.

[0071] In a possible implementation, the vehicle controller determines a vehicle inclination angle of the target vehicle based on the vehicle state information. The vehicle controller determines a weather type and a road surface type of the location where the target vehicle is located based on the environment information. The vehicle controller determines the action recognition strategy of the target vehicle based on the vehicle inclination angle, the weather type, and the road surface type.

[0072] The vehicle inclination is an absolute value of the inclination, and the vehicle inclination can reflect a road slope of a road where the target vehicle is located. The size of the vehicle inclination directly reflects the size of the road slope. The embodiments of the present application provide a technical solution for controlling the back door by using the pattern projected by the projection assembly. The road slope affects the projection effect of the projection assembly, thereby affecting the motion recognition strategy. The weather types include sunny, light rain, heavy rain, rainstorm, light snow, heavy snow, and snowstorm, etc. In the embodiments of the present application, the weather types are divided into a first type of weather and a second type of weather. The second type of weather is a severe weather type, for example, rainstorm and snowstorm belong to the second type of weather. The severe weather type seriously affects the projection effect of the projection assembly. The first type of weather is a non-severe weather type, for example, sunny, light rain, heavy rain, light snow, and heavy snow belong to the first type of weather. The non-severe weather type does not affect or slightly affects the projection effect of the projection assembly. Of course, the above-mentioned enumeration and division of the weather types are only examples. In other possible implementations, the division can be made by the technician according to the actual situation and requirements. The embodiments of the present application do not limit this. The road surface types include cement road surface, asphalt road surface, sandstone road surface, muddy road surface, and soil road surface, etc. In the embodiments of the present application, the road surface types are divided into a preset road surface type and a non-preset road surface type. The preset road surface type is a road surface type with poor flatness. The road surface of the preset road surface type seriously affects the projection effect of the projection assembly. The road surface of the non-preset road surface type does not affect or slightly affects the projection effect of the projection assembly.

[0073] In this implementation, the motion recognition strategy of the target vehicle is determined in combination with the vehicle inclination, the weather type, and the road surface type. The motion recognition strategy matches the vehicle state of the target vehicle and the environmental state of the location. The motion recognition strategy can be used to accurately control the back door.

[0074] In order to more clearly illustrate the above-mentioned implementation, the following is divided into several parts to illustrate the above-mentioned implementation.

[0075] In a first part, the vehicle controller determines the vehicle inclination of the target vehicle based on the vehicle state information.

[0076] In a possible implementation, the vehicle state information includes three-axis acceleration measured by a gyroscope. The vehicle controller determines the vehicle inclination of the target vehicle based on the three-axis acceleration.

[0077] The way of determining the vehicle inclination by using the three-axis acceleration can adopt the way provided in the related art, and the embodiments of the present application do not limit this.

[0078] In this implementation, the three-axis acceleration measured by the gyroscope can be used to determine the vehicle inclination. The determination efficiency of the vehicle inclination is high.

[0079] The second part, the vehicle controller, determines the weather type and the road surface type of the location where the target vehicle is based on the environmental information.

[0080] To make the second part more clearly, the following first describes the way of determining the weather type based on the environmental information.

[0081] In a possible implementation, the environmental information includes rainfall information or snowfall information collected by a rainfall sensor. The vehicle controller determines the weather type of the location where the target vehicle is based on the rainfall information or the snowfall information.

[0082] The rainfall information can indicate the rainfall amount of the location where the target vehicle is, and the snowfall information can indicate the snowfall amount of the location where the target vehicle is. Since the rainfall information and the snowfall information are directly collected by the rainfall sensor, the accuracy is high.

[0083] In the above implementation, the rainfall information or the snowfall information collected by the sensor is used to determine the weather type, and the accuracy of the weather type is high.

[0084] For example, in a case where the rainfall amount indicated by the rainfall information is less than or equal to a rainfall threshold, or the snowfall amount indicated by the snowfall information is less than or equal to a snowfall threshold, the vehicle controller determines that the weather type of the location where the target vehicle is is a first weather type. In a case where the rainfall amount indicated by the rainfall information is greater than the rainfall threshold, or the snowfall amount indicated by the snowfall information is greater than the snowfall threshold, the vehicle controller determines that the weather type of the location where the target vehicle is is a second weather type.

[0085] The first weather type can be considered as non-adverse weather, and the second weather type can be considered as adverse weather.

[0086] In a possible implementation, in a case where the environmental information includes the weather type queried from a weather platform, the vehicle controller obtains the weather type from the environmental information.

[0087] In this implementation, the weather type can be directly obtained from the environmental information, the obtaining of the weather type does not depend on additional sensors, and the cost of obtaining the weather type is low.

[0088] The following first describes the way of determining the road surface type based on the environmental information.

[0089] In a possible implementation, the environmental information includes a road surface image or a road surface point cloud. The vehicle controller determines the road surface type based on the road surface image or the road surface point cloud.

[0090] The road surface image can reflect the road surface state of the road surface, and thus the corresponding road surface type can be determined by using the road surface state. The road surface point cloud can reflect the flatness of the road surface, and the corresponding road surface type can be determined in combination with the flatness.

[0091] In this implementation, the road surface type can be determined by using the road surface image or the road surface point cloud, and the accuracy of the road surface type is relatively high.

[0092] For example, in a case where the environment information includes the road surface image, the vehicle controller performs feature extraction on the environment image to obtain environment image features of the environment image. The vehicle controller performs classification based on the environment image features to obtain the road surface type. In a case where the environment information includes the road surface point cloud, the vehicle controller performs feature extraction on the environment point cloud to obtain environment point cloud features of the environment point cloud. The vehicle controller performs classification based on the environment point cloud features to obtain the road surface type.

[0093] The environment image features are high-dimensional abstract representations of the environment image, and the environment point cloud features are high-dimensional abstract representations of the environment point cloud. The classification based on the features is implemented by using a pre-trained classifier. The environment image features or the environment point cloud features are input into the classifier, and the corresponding road surface type can be obtained. The classifier is a multi-classifier, which can be any type of multi-classifier in the related art, and embodiments of the present application do not limit the classifier.

[0094] In a possible implementation, in a case where the environment information includes the road surface type queried from the geographic information platform, the vehicle controller obtains the road surface type from the environment information.

[0095] The third part, the vehicle controller determines the action recognition strategy of the target vehicle based on the vehicle inclination, the weather type, and the road surface type.

[0096] In a possible implementation, in a case where the weather type belongs to the first type of weather type, the vehicle controller determines the action recognition strategy of the target vehicle based on the vehicle inclination or the road surface type. In a case where the weather type belongs to the second type of weather type, the vehicle controller determines the action recognition strategy of the target vehicle as a first recognition strategy, and the first recognition strategy belongs to a second type of recognition strategy. The second type of recognition strategy is a strategy of not performing the rear door opening action recognition by projecting the first pattern by the projection component. The second type of weather type is more severe than the first type of weather type.

[0097] The first weather type is a non-adverse weather type, and the second weather type is an adverse weather type. The rear door opening operation recognition not through the first pattern refers to that the rear door opening operation recognition is not performed by using the projection component. For example, the first recognition strategy can be a voice recognition strategy or a key recognition strategy. In the voice recognition strategy, the user can control the rear door by using a voice instruction. In the key recognition strategy, the user can control the rear door by pressing a specific key. The action recognition strategy can be a single recognition strategy or a mixed recognition strategy, which is not limited in the embodiments of the present application.

[0098] In this embodiment, since the weather type has an important influence on the projection effect of the projection component, the determination of the action recognition strategy according to whether the weather type belongs to the first weather type or the second weather type can determine whether the projection component is used for subsequent recognition, thereby realizing the adaptive determination of the recognition strategy.

[0099] In order to more clearly illustrate the above embodiments, the determination of the action recognition strategy based on the vehicle inclination or the road surface type in the above embodiments is described below.

[0100] In some embodiments, when the vehicle inclination is greater than the first preset inclination, the action recognition strategy of the target vehicle is determined as a second recognition strategy by the vehicle controller, and the second recognition strategy is a repeated action recognition strategy. When the vehicle inclination is less than the second preset inclination, the action recognition strategy of the target vehicle is determined as a third recognition strategy, and the third recognition strategy is a preset recognition strategy. When the road surface type belongs to a preset road surface type, the action recognition strategy of the target vehicle is determined as a fourth recognition strategy, and the fourth recognition strategy is a time-delayed action recognition strategy.

[0101] The vehicle inclination is greater than the first preset inclination, which indicates that the vehicle inclination of the target vehicle is large. In the case that the front of the target vehicle is higher than the rear (the inclination corresponding to the vehicle inclination is positive), the items in the trunk may roll out of the vehicle after the tailgate is opened. In the case that the front of the target vehicle is lower than the rear (the inclination corresponding to the vehicle inclination is negative), the items put into the trunk by the user may move forward. The repeated action recognition strategy means that at least two correct tailgate opening actions are recognized before a response is performed, which can make the user pay more attention, pay attention to the items rolling out of the vehicle when the front is higher than the rear, and pay attention to the items moving forward when the front is lower than the rear. The preset recognition strategy is a recognition strategy configured in advance, which does not need additional strategy adjustment, that is, the default recognition strategy for controlling the tailgate by the projection assembly. The preset road surface type refers to a road surface type with a flatness less than a preset flatness. For example, the delayed action recognition strategy is a recognition strategy that prolongs the action recognition time. For example, in the preset recognition strategy, a correct tailgate opening action is recognized to perform a response. In the delayed action recognition strategy, after a correct tailgate opening action is recognized, a response is not performed immediately, but identification is continued. When the duration of the tailgate opening action is greater than a preset duration, a response is performed, which can prevent misidentification. The first preset inclination, the second preset inclination, and the preset duration are set or adjusted by a technician or a user according to actual conditions, and embodiments of the present application do not limit this. In addition, the tailgate recognition strategies corresponding to the second recognition strategy, the third recognition strategy, and the fourth recognition strategy all belong to the first type of recognition strategy, which is a strategy for tailgate opening action recognition by the first pattern projected by the projection assembly, that is, if the second recognition strategy, the third recognition strategy, or the fourth recognition strategy is used, the user needs to use the pattern projected on the ground by the projection assembly to control the tailgate.

[0102] In some embodiments, in the case that the action recognition strategy is the first recognition strategy, the working mode for tailgate control can be defined as a harsh weather mode; in the case that the action recognition strategy is the second recognition strategy, the working mode for tailgate control can be defined as a slope adaptive mode; in the case that the action recognition strategy is the third recognition strategy, the working mode for tailgate control can be defined as a standard mode; and in the case that the action recognition strategy is the fourth recognition strategy, the working mode for tailgate control can be defined as a complex terrain mode. In addition, in different working modes, the projection assembly has different projection strategies, and different projection strategies correspond to different projection parameters. The determination of the projection parameters will be described later. The determination of the above-mentioned action recognition strategy can be represented by Table 1.

[0103] Table 1

[0104] It should be noted that the above working modes and action recognition strategies are only examples, and other working modes and other action recognition strategies exist in other possible embodiments. For example, the slope adaptive mode and the complex terrain mode can be combined into a new working mode, and the repeated action recognition strategy and the time-delay action recognition strategy can be combined into a new action recognition strategy. The embodiments of the present application do not limit this.

[0105] For example, in the case that the vehicle inclination is greater than the first preset inclination and the road surface type is the preset road surface type, a composite mode of the slope adaptive mode and the complex terrain mode can be enabled, and the action recognition strategy is a composite recognition strategy of the repeated action recognition strategy and the time-delay action recognition strategy, that is, a recognition strategy that considers both the number of times and the time length.

[0106] Another embodiment of the third part will be described below.

[0107] In a possible embodiment, the vehicle controller inputs the vehicle inclination, the weather type, and the road surface type into an identification strategy determination model, extracts features of the vehicle inclination, the weather type, and the road surface type through the identification strategy determination model, and obtains strategy determination features of the target vehicle. The vehicle controller maps the strategy determination features through the identification strategy determination model, and obtains the action recognition strategy of the target vehicle.

[0108] The identification strategy determination model is a trained multi-classification model, and the corresponding action recognition strategy can be found by using the identification strategy determination model. The training method of the identification strategy determination model can adopt the training method of a multi-classification model in related technologies, such as using a cross-entropy loss function to train the multi-classification model. The embodiments of the present application do not limit this.

[0109] In this embodiment, the corresponding action recognition strategy can be determined by using the identification strategy determination model, the generalization ability of the action recognition model is used to determine the action recognition strategy in a complex scene, and the accuracy is high.

[0110] For example, the vehicle controller inputs the vehicle inclination, the weather type, and the road surface type after splicing into an identification strategy determination model, and performs multiple full connections or multiple convolutions on the spliced information through the identification strategy determination model to obtain strategy determination features of the target vehicle. The vehicle controller performs full connection and normalization on the strategy determination features through the identification strategy determination model to obtain a probability set, and the probability set includes multiple probabilities, and one probability corresponds to one candidate recognition strategy. The vehicle controller determines the candidate recognition strategy corresponding to the highest probability in the probability set as the action recognition strategy of the target vehicle.

[0111] The normalization can be implemented by a normalization function, which is a SoftMax function or a ReLu function, etc. The embodiments of the present application do not make any limitation in this aspect.

[0112] Optionally, after step 302, either step 303 or step 307 can be performed, and the embodiments of the present application do not make any limitation in this aspect.

[0113] 303. In a case where the action recognition strategy belongs to the first type of recognition strategy, the vehicle controller determines the projection parameter corresponding to the action recognition strategy.

[0114] The projection parameter is a parameter for controlling a projection component. The projection component is a device capable of projecting a pattern on the ground, such as a hardware module capable of realizing a ground projection function. The first type of recognition strategy refers to a recognition strategy that relies on a ground projection pattern for back door opening action recognition. The first type of recognition strategy can include multiple action recognition strategies.

[0115] In a possible implementation, in a case where the action recognition strategy of the target vehicle is the second recognition strategy, the vehicle controller determines the projection parameter corresponding to the second recognition strategy based on the vehicle inclination, the road surface type, and the distance between the projection component and the ground.

[0116] The action recognition strategy is the second recognition strategy, indicating that the vehicle inclination angle of the target vehicle is large. At this time, if the projection parameter is not adjusted, the pattern projected on the ground by the projection component will be distorted, and the position may also deviate greatly from the ideal position. At this time, the projection angle needs to be adjusted to eliminate the distortion of the pattern as much as possible and project the pattern at the ideal position. The distance between the projection component and the ground is the height of the projection component, which affects the projection position and projection distance of the pattern. In addition, different road types correspond to different flatness and reflectivity. The projection parameter determined in combination with the road type is more matched with the actual road condition at the position of the target vehicle, so better projection effect can be achieved subsequently. The projection parameter includes the projection pattern, the projection distance, the projection size, and the projection brightness. The projection pattern is the pattern that needs to be projected on the ground. The projection distance is the distance between the pattern projected on the ground and the target vehicle. The projection distance is associated with the projection center offset. The determination method of the projection center offset is shown in the following formula (1). The projection size is the size of the pattern projected on the ground. The projection brightness is the brightness of the pattern projected on the ground. The projection parameter corresponding to the second recognition strategy is the projection parameter for compensating the projection position. On the one hand, the first pattern is close to the vehicle due to the slope, so when the user uses the first pattern to open the back door, the back door may collide with the user. On the other hand, the first pattern is far from the vehicle due to the slope, which is not convenient for the user to perform the action.

[0117] ΔS = K * L * sinθ (1) Wherein, ΔS is the projection center offset, K is the terrain coefficient, which is associated with the ground type, L is the height of the projection component, and θ is the vehicle inclination angle.

[0118] In this embodiment, when the action recognition strategy is the second recognition strategy, the vehicle inclination angle, the road type, and the distance between the projection component and the ground are used to determine the projection parameter. The projection parameter is adapted to the vehicle state and the environmental state of the target vehicle. Using the projection parameter to control the projection of the projection component can achieve better projection effect.

[0119] For example, when the action recognition strategy of the target vehicle is the second recognition strategy, the vehicle control unit obtains the projection pattern, the projection size, and the projection brightness corresponding to the second recognition strategy. The vehicle control unit determines the projection center offset based on the vehicle inclination angle, the road type, and the distance between the projection component and the ground. The vehicle control unit determines the projection distance based on the projection center offset and the preset projection distance. The above-mentioned projection pattern, projection size, projection distance, and projection brightness constitute the projection parameter corresponding to the second recognition strategy.

[0120] The action recognition strategy corresponds to the projection pattern, the projection size, and the projection brightness, and the action recognition strategy can be used to query the corresponding projection pattern, the projection size, and the projection brightness.

[0121] Another embodiment of the step 303 is described below.

[0122] In a possible implementation, when the action recognition strategy of the target vehicle is the third recognition strategy, the vehicle controller determines the preset projection parameter as the projection parameter corresponding to the third recognition strategy.

[0123] The third recognition strategy is a preset recognition strategy, and the corresponding projection parameter is a preset projection parameter. The preset projection parameter includes a preset projection pattern, a preset projection distance, a preset projection size, and a preset projection brightness. The preset projection parameter is set by a technician or a user according to actual conditions or requirements, and embodiments of the application do not limit this. In some embodiments, the preset projection pattern is a circular target.

[0124] In this implementation, when the action recognition strategy is a preset recognition strategy, the corresponding preset projection parameter is determined as the projection parameter, and a better projection effect can be obtained.

[0125] Another embodiment of the step 303 is described below.

[0126] In a possible implementation, when the action recognition strategy of the target vehicle is the fourth recognition strategy, the vehicle controller determines a target projection size and a target projection brightness of the projection pattern based on the road surface type. The vehicle controller determines the projection parameter corresponding to the fourth recognition strategy based on the target projection size and the target projection brightness.

[0127] The target projection size and the target projection brightness are associated with the road surface type. In the fourth recognition strategy, it is indicated that the road surface type belongs to a preset road surface type. The sandy and muddy road surface belongs to the preset road surface type. At this time, the flatness of the road surface is poor, and therefore the target projection size and the target projection brightness are determined again, so as to expect a better projection effect of the pattern on the road surface of the preset road surface type. The road surface type corresponds to the projection size and the projection brightness, and the road surface type can be used to directly query the corresponding target projection size and target projection brightness. The target projection size is greater than the preset projection size, and the target projection brightness is higher than the preset projection brightness. The projection parameter corresponding to the fourth recognition strategy is a projection parameter for magnifying and enhancing the brightness of the projection pattern.

[0128] In this implementation, when the action recognition strategy is the fourth recognition strategy, the road surface type is used to determine the target projection size and the target projection brightness, and then the projection parameters are determined in combination with the target projection size and the target projection brightness. The matching degree between the projection parameters and the fourth recognition strategy is higher.

[0129] For example, when the action recognition strategy of the target vehicle is the fourth recognition strategy, the vehicle controller queries the road surface type to obtain the target projection size and the target projection brightness of the projection pattern. The integration controller determines the set of the target projection size, the target projection brightness, the preset projection pattern, and the preset projection distance as the projection parameters corresponding to the fourth recognition strategy.

[0130] For example, the preset projection pattern is a circular target, the preset projection size is 30 cm in diameter, the preset projection brightness is 400 nit, and the preset projection position is 2 m. When the action recognition strategy is the second recognition strategy, the projection position 2 m is enlarged or reduced in combination with the vehicle inclination to adapt to different road slopes, and other preset parameters remain unchanged. When the action recognition strategy is the third recognition strategy, the above preset projection parameters can be directly used. When the action recognition strategy is the fourth recognition strategy, the preset projection size and the preset projection brightness are enlarged, for example, the preset projection size is enlarged from 30 cm to 50 cm, and the preset projection brightness is enlarged from 400 nit to 600 nit (increased by 50%).

[0131] The following Table 2 shows examples of the projection parameters corresponding to the different action recognition strategies.

[0132] Table 2

[0133] 304、The vehicle controller controls the projection assembly of the target vehicle to project a first pattern on the ground based on the projection parameters corresponding to the action recognition strategy.

[0134] The first pattern is a projection pattern projected by the projection assembly on the ground to guide the user to perform the back door opening action.

[0135] In one possible implementation, the vehicle controller determines component control parameters of the projection assembly based on the projection parameters. The vehicle controller controls the projection assembly to project the first pattern on the ground by using the component control parameters.

[0136] The projection parameters include a projection pattern, a projection distance, a projection size, and a projection brightness, and actually describe the projection effect of the first pattern. The component control parameters are used to control the projection component to project, so that the projection effect of the first pattern projected on the ground is close to the projection effect indicated by the projection parameters. There is a corresponding relationship between the projection parameters and the component control parameters, which is calibrated by a technician according to the actual situation, and embodiments of the present application do not limit this. In some embodiments, the projection component is an AR projection control module, and the AR projection control module includes an adjustable focus DLP projector and a dynamic pattern generator. In some embodiments, the adjustable focus DLP projector supports electric adjustment in the range of -15°~+30°, so that the first pattern can be projected on multiple positions on the ground. The dynamic pattern generator is used to generate the first pattern that needs to be projected, and the dynamic pattern generator is built-in with multiple patterns that can be projected.

[0137] For example, referring to Figure 4 The vehicle controller can control the projection component of the target vehicle 400 to project the first pattern 401 on the ground.

[0138] 305, the vehicle controller determines whether the back door opening action exists in the range indicated by the first pattern.

[0139] The range indicated by the first pattern on the ground is the detection range of the back door opening action, that is, the first pattern is actually used to indicate the range of the back door opening action. While the projection component projects the first pattern on the ground, the vehicle controller controls the action recognition component to perform action detection in the range indicated by the first pattern to determine whether the back door opening action exists, that is, the detection range of the action recognition component is the same as the range indicated by the first pattern, and in the case that the range indicated by the first pattern changes, the detection range of the action recognition component will also change. The back door opening action refers to a specific body movement performed by the user to open the back door. In the first type of recognition strategy, the back door opening action refers to the action performed by the user's foot on the first pattern on the ground.

[0140] In a possible implementation, the action recognition component is a TOF component, and in the case that the action recognition component is started, the action recognition component measures the distance in the range indicated by the first pattern. In response to detecting that the distance changes through the action recognition component, the vehicle controller determines that the back door opening action exists in the range indicated by the first pattern. In the case that the distance does not change through the action recognition component, the vehicle controller determines that the back door opening action does not exist in the range indicated by the first pattern.

[0141] The TOF component utilizes infrared light to measure distance. The TOF component periodically emits near-infrared light modulation waves outward, and the modulation waves are reflected after encountering an object. The TOF component measures the round-trip phase difference of the modulation waves to obtain the time of flight, and then calculates the relative distance between the TOF component and the measured target. When the user taps the first pattern, the distance is detected to decrease. In some embodiments, when the action recognition strategy of the target vehicle is the fourth recognition strategy, it indicates that the flatness of the road at this time is poor, and the accuracy of action recognition using the action recognition component alone is low. At this time, the millimeter wave radar of the target vehicle is controlled by the vehicle controller to measure the distance of the range indicated by the first pattern, thereby assisting the TOF component to implement action recognition and improving the accuracy of action recognition.

[0142] Another embodiment of the above step 305 is described below.

[0143] In a possible implementation, the action recognition component is a camera. When the action recognition component is started, the action recognition component will collect an image in the range indicated by the first pattern. In response to identifying that there is a preset action in the image, the vehicle controller determines that there is a back door opening action in the range indicated by the first pattern. In the case where the preset action is not identified in the image by the action recognition component, the vehicle controller determines that there is no back door opening action in the range indicated by the first pattern.

[0144] The preset action is set by a technician or a user according to actual conditions, and embodiments of the present application do not limit this. The way of identifying whether there is a preset action in the image can be realized by using the image recognition method in related technologies.

[0145] Correspondingly, on the basis of the above embodiments, when the action recognition strategy of the target vehicle is the fourth recognition strategy, the millimeter wave radar of the target vehicle can also be controlled by the vehicle controller to measure the distance of the range indicated by the first pattern, thereby assisting the camera to implement action recognition and improving the accuracy of action recognition.

[0146] 306、In response to the back door opening action in the range indicated by the first pattern, the vehicle controller uses the action recognition strategy to recognize the back door opening action, and controls the back door of the target vehicle according to the recognition result.

[0147] The recognition result is the final judgment obtained after detecting and algorithmically analyzing the back door opening action. The result is usually a binary judgment: it is an effective opening signal or it is not an effective opening signal. Only when the result is the former, the vehicle controller will issue an unlock instruction to the back door controller, thereby controlling the back door to open.

[0148] In a possible implementation, when the action recognition strategy is the second recognition strategy, the vehicle controller identifies the number of times that the back door opening action is performed within the first pattern indication range. When the number of times is greater than or equal to a preset number of times, the vehicle controller controls the back door to open. When the number of times is less than the preset number of times, the vehicle controller does not respond to the back door opening action.

[0149] The preset number of times is greater than or equal to 2, and the preset number of times is set by a technician according to an actual situation, which is not limited in the embodiments of the present application. The second recognition strategy is a repeated action recognition strategy, and the preset number of times is used to implement the repeated action recognition. In some embodiments, when the action recognition component is a TOF component, the number of times that the back door opening action is performed can be represented by the number of cycles of distance changes that are identified; when the action recognition component is a camera, the number of times that the back door opening action is performed can be represented by the number of preset actions that are identified.

[0150] In this implementation, when the action recognition strategy is the second recognition strategy, whether to respond to the back door opening action is determined by counting the number of times that the back door opening action is performed, and the accuracy of action response is relatively high. Because the vehicle inclination corresponding to the second recognition strategy is relatively large, through the repeated action recognition described above, the accuracy of action recognition can be improved, and the user can be reminded to pay attention to the situation that the objects in the target vehicle fall or move forward.

[0151] In a possible implementation, when the action recognition strategy is the third recognition strategy, the vehicle controller identifies whether the back door opening action is located in a preset position range of the first pattern. When the back door opening action is located in the preset position range, the vehicle controller controls the back door to open. When the back door opening action is not located in the preset position range, the vehicle controller does not respond to the back door opening action.

[0152] The preset position range is smaller than the range indicated by the first pattern. In the third recognition strategy, the weather condition of the position where the target vehicle is located is relatively good, and the road surface is relatively flat. At this time, whether to respond to the back door opening action is determined by refining the judgment of the back door opening action, which can reduce the probability of false response. The preset position range is set by a technician according to an actual situation, which is not limited in the embodiments of the present application. In some embodiments, when the action recognition component is a TOF component, the position of distance change can be identified, so as to determine whether the back door opening action is located in the preset position range of the first pattern; when the action recognition component is a camera, whether the back door opening action is located in the preset position range of the first pattern can be determined by identifying the relative positional relationship between the back door opening action and the first pattern.

[0153] In this implementation, when the action recognition strategy is the third recognition strategy, the action response accuracy is higher by judging whether to respond to the back door opening action by judging whether the back door opening action is located in the preset position range of the first pattern.

[0154] In a possible implementation, when the action recognition strategy is the fourth recognition strategy, the vehicle controller identifies the duration of the back door opening action within the first pattern indication range. When the duration is greater than or equal to a preset duration, the vehicle controller controls the back door to open. When the duration is less than the preset duration, the vehicle controller does not respond to the back door opening action.

[0155] The preset duration is set by a technician according to actual conditions, such as 3s or 5s, and the embodiments of the present application do not limit this. The fourth recognition strategy is a delay action recognition strategy, which is mainly to prolong the action recognition duration to improve the recognition accuracy. This is because the flatness of the road where the target vehicle is located is lower under the fourth recognition strategy, and the action recognition accuracy will decrease. The action recognition duration is prolonged to compensate for the decrease in accuracy.

[0156] In this implementation, when the action recognition strategy is the fourth recognition strategy, whether to respond to the back door opening action can be determined by judging the duration of the back door opening action, so as to eliminate the influence of the road with low flatness as much as possible and improve the accuracy of the action response.

[0157] In some embodiments, when the weather type indicates that it is raining or snowing, the vehicle controller controls the opening angle of the back door based on the vehicle inclination angle to prevent rainwater or accumulated snow from flowing into the target vehicle.

[0158] The vehicle inclination angle and the opening angle of the back door have a corresponding relationship, which is set by a technician according to actual conditions, and the embodiments of the present application do not limit this.

[0159] 307、When the action recognition strategy belongs to the second type of recognition strategy, the vehicle controller controls the projection component of the target vehicle to project a second pattern on the ground, and the second pattern is used to prompt that the back door of the target vehicle cannot be opened by the pattern projected by the projection component.

[0160] In the above description, the first identification strategy belongs to the second type of identification strategy, and the weather type under the first identification strategy belongs to the second type of weather type, that is, the severe weather type. In this scenario, the projection effect of the projection component is seriously affected, and the identification accuracy of the motion recognition component is also reduced, so the method of controlling the back door by using the first pattern is disabled. The second pattern is a reminder pattern, which is used to remind the user to control the back door through a voice instruction or by pressing a button, including any one of pressing a button on a vehicle key, pressing a button on the target vehicle, and pressing a button on the target mobile terminal.

[0161] According to the technical scheme provided by the embodiment of the present application, in the case that the user is identified to approach the target vehicle, the vehicle state of the target vehicle and the environmental condition of the location are combined to dynamically determine the motion recognition strategy corresponding to the target vehicle, and the motion recognition strategy is matched with the vehicle state and the environmental condition. In the case that the motion recognition strategy belongs to the first type of identification strategy, the corresponding projection parameter is used to control the projection component to project the first pattern on the ground, and the user can use the first pattern to control the back door. In the case that the back door opening action within the indication range of the first pattern is identified, the back door opening action is identified by using the motion recognition strategy, and the back door is controlled according to the identification result, thereby meeting the user's demand for controlling the back door in a specific scenario.

[0162] Figure 5 is a structural schematic diagram of a vehicle back door control device provided by an embodiment of the present application, referring to Figure 5 The device comprises a strategy determination module 501, a projection control module 502, and a back door control module 503.

[0163] The strategy determination module 501 is configured to determine the motion recognition strategy corresponding to the target vehicle based on the vehicle state information of the target vehicle and the environmental information of the location in response to the user approaching the target vehicle.

[0164] The projection control module 502 is configured to control the projection component of the target vehicle to project the first pattern on the ground based on the projection parameter corresponding to the motion recognition strategy in the case that the motion recognition strategy belongs to the first type of identification strategy. The first type of identification strategy is a strategy for identifying the back door opening action by using the first pattern.

[0165] The back door control module 503 is configured to identify the back door opening action by using the motion recognition strategy in response to the back door opening action within the indication range of the first pattern, and control the back door of the target vehicle according to the identification result.

[0166] In a possible implementation, the strategy determination module 501 is configured to determine a vehicle inclination of the target vehicle based on the vehicle state information. The weather type and the road surface type of the location where the target vehicle is located are determined based on the environment information. The action recognition strategy of the target vehicle is determined based on the vehicle inclination, the weather type, and the road surface type.

[0167] In a possible implementation, the strategy determination module 501 is configured to determine the action recognition strategy of the target vehicle based on the vehicle inclination or the road surface type, in a case where the weather type belongs to a first weather type. In a case where the weather type belongs to a second weather type, the action recognition strategy of the target vehicle is determined as a first recognition strategy, the first recognition strategy belongs to a second recognition strategy, the second recognition strategy is a strategy of not performing the rear door opening action recognition by using the first pattern, and the second weather type is more severe than the first weather type.

[0168] In a possible implementation, the strategy determination module 501 is configured to determine the action recognition strategy of the target vehicle as a second recognition strategy in a case where the vehicle inclination is greater than a first preset inclination, the second recognition strategy being a repeated action recognition strategy. The action recognition strategy of the target vehicle is determined as a third recognition strategy in a case where the vehicle inclination is less than a second preset inclination, the third recognition strategy being a preset recognition strategy. The action recognition strategy of the target vehicle is determined as a fourth recognition strategy in a case where the road surface type belongs to a preset road surface type, the fourth recognition strategy being a time-delayed action recognition strategy. The rear door recognition strategies corresponding to the second recognition strategy, the third recognition strategy, and the fourth recognition strategy all belong to the first recognition strategy.

[0169] In a possible implementation, the apparatus further includes: In a possible implementation, the projection parameter determination module is configured to determine the projection parameter based on the vehicle inclination, the road surface type, and a distance between the projection component and the ground, in a case where the action recognition strategy of the target vehicle is the second recognition strategy. In a case where the action recognition strategy of the target vehicle is the third recognition strategy, a preset projection parameter is determined as the projection parameter. In a case where the action recognition strategy of the target vehicle is the fourth recognition strategy, a target projection size and a target projection brightness of the projection pattern are determined based on the road surface type. The projection parameter is determined based on the target projection size and the target projection brightness.

[0170] In a possible implementation, the projection control module 502 is configured to determine a component control parameter of the projection component based on the projection parameter. The projection component is controlled to project the first pattern on the ground by using the component control parameter.

[0171] In a possible implementation, the back door control module 503 is configured to, in a case where the action recognition strategy is a second recognition strategy, recognize a number of times of execution of the back door opening action within the first pattern indication range. In a case where the number of times of execution is greater than or equal to a preset number of times, the back door is controlled to be opened. In a case where the number of times of execution is less than the preset number of times, the back door opening action is not responded to. In a case where the action recognition strategy is a third recognition strategy, it is recognized whether the back door opening action is located in a preset position range of the first pattern. In a case where the back door opening action is located in the preset position range, the back door is controlled to be opened. In a case where the back door opening action is not located in the preset position range, the back door opening action is not responded to. In a case where the action recognition strategy is a fourth recognition strategy, it is recognized a duration for which the back door opening action is within the first pattern indication range. In a case where the duration is greater than or equal to a preset duration, the back door is controlled to be opened. In a case where the duration is less than the preset duration, the back door opening action is not responded to.

[0172] In a possible implementation, the device further includes: The user recognition module is configured to, in response to detecting a vehicle key of the target vehicle, determine that a user is approaching the target vehicle. Alternatively, in response to detecting a target mobile terminal, determine that a user is approaching the target vehicle, the target mobile terminal being a mobile terminal having a binding relationship with the target vehicle. Alternatively, in response to identifying a target user from the environment information, determine that a user is approaching the target vehicle, the target user being a user having a binding relationship with the target vehicle.

[0173] In a possible implementation, the projection control module 502 is further configured to, in a case where the action recognition strategy belongs to a second type of recognition strategy, control a projection component of the target vehicle to project a second pattern on the ground, the second pattern being used to prompt the back door of the target vehicle to be opened by a pattern that cannot be projected by the projection component. In this implementation, in a case where the back door cannot be controlled by the first type of recognition strategy, the projection component is controlled to project the second pattern to prompt the user to use other ways to control the back door, which can improve the efficiency of human-computer interaction.

[0174] It should be noted that the vehicle back door control device provided in the above embodiments is only used as an example for the division of the above functional modules when controlling the back door of the vehicle. In actual applications, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the vehicle back door control device and the vehicle back door control method provided in the above embodiments belong to the same concept, and the specific implementation process is described in the method embodiments, which will not be repeated here.

[0175] Through the technical solutions provided by the embodiments of the present application, in the case of identifying that the user is close to the target vehicle, the action recognition strategy corresponding to the target vehicle is dynamically determined in combination with the vehicle state of the target vehicle and the environmental condition of the position where the target vehicle is located, and the action recognition strategy is matched with the vehicle state and the environmental condition. In the case of the action recognition strategy belonging to the first type of recognition strategy, the corresponding projection parameter is used to control the projection component to project the first pattern on the ground, and the user can use the first pattern to control the back door. In the case of identifying the back door opening action within the indication range of the first pattern, the back door opening action is recognized by using the action recognition strategy, and the back door is controlled according to the recognition result, thereby meeting the user's demand for controlling the back door in a specific scenario.

[0176] The embodiments of the present application also provide a vehicle, Figure 6 FIG. 1 is a structural schematic diagram of a vehicle provided by an embodiment of the present application.

[0177] Generally, the vehicle 600 comprises one or more processors 601 and one or more memories 602.

[0178] The processor 601 can comprise one or more processing cores, such as a 4-core processor, a 6-core processor, etc. The processor 601 can be implemented in at least one of a hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), a PLA (Programmable Logic Array). The processor 601 can also comprise a main processor and a coprocessor, the main processor being a processor for processing data in an awake state, also known as a CPU (Central Processing Unit), and the coprocessor being a low-power processor for processing data in a standby state. In some embodiments, the processor 601 can be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed by the display screen. In some embodiments, the processor 601 can further comprise an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0179] The memory 602 can include one or more computer-readable storage media. The memory 602 can also include high-speed random access memory and non-volatile, computer-readable storage media such as one or more magnetic disk storage devices, flash memory devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 602 is used to store at least one computer program for being executed by the processor 601 to implement the control method of the vehicle back door provided by the method embodiments in the present application.

[0180] Those skilled in the art can understand that, Figure 6 The structure shown in the figure does not constitute a limitation on the vehicle 600, and can include more or fewer components than shown, or combine certain components, or adopt a different arrangement of components.

[0181] In addition, the device provided by the embodiments of the present application can be a chip, a component or a module, which can include a processor and a memory connected to each other. The memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the control method of the vehicle back door provided by the above-mentioned embodiments.

[0182] The embodiment also provides a computer-readable storage medium, which stores computer program codes, and when the computer program codes run on a computer, the computer executes the above-mentioned related method steps to implement the control method of the vehicle back door provided by the above-mentioned embodiments.

[0183] The embodiment also provides a computer program product, which makes the computer execute the above-mentioned related steps to implement the control method of the vehicle back door provided by the above-mentioned embodiments when the computer program product runs on the computer.

[0184] The device, computer-readable storage medium, computer program product or chip provided by the embodiment are used to execute the corresponding method provided above, so the beneficial effects they can achieve can refer to the beneficial effects of the corresponding method provided above, which will not be repeated here.

[0185] Through the description of the above embodiments, those skilled in the art can understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.

[0186] In the embodiments of the present disclosure, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic; for example, the division of the modules or units is only a logical function division; there can be another division manner in actual implementation; for example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, or the among different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0187] The foregoing is merely specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto; any modification or replacement within the technical range disclosed by the present disclosure can be easily conceived by those skilled in the art, and should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A method for controlling a vehicle's tailgate, characterized in that, The method includes: In response to a user approaching a target vehicle, a motion recognition strategy corresponding to the target vehicle is determined based on the vehicle status information of the target vehicle and the environmental information of its location. When the action recognition strategy belongs to the first type of recognition strategy, the projection component of the target vehicle is controlled to project a first pattern on the ground based on the projection parameters corresponding to the action recognition strategy. The first type of recognition strategy is a strategy that uses the first pattern to recognize the action of opening the tailgate. In response to a tailgate opening action within the first pattern indication range, the action recognition strategy is used to recognize the tailgate opening action, and the tailgate of the target vehicle is controlled according to the recognition result.

2. The method according to claim 1, characterized in that, The action recognition strategy for the target vehicle is determined based on the vehicle status information of the target vehicle and the environmental information of its location. Based on the vehicle status information, the vehicle tilt angle of the target vehicle is determined; Based on the environmental information, the weather type and road surface type at the location of the target vehicle are determined; Based on the vehicle tilt angle, the weather type, and the road surface type, a motion recognition strategy for the target vehicle is determined.

3. The method according to claim 2, characterized in that, The motion recognition strategy for determining the target vehicle based on the vehicle tilt angle, the weather type, and the road surface type includes: If the weather type is classified as the first type, the action recognition strategy for the target vehicle is determined based on the vehicle tilt angle or the road surface type. When the weather type is classified as the second type of weather, the action recognition strategy of the target vehicle is determined as the first recognition strategy. The first recognition strategy belongs to the second type of recognition strategy. The second type of recognition strategy is a strategy that does not recognize the tailgate opening action through the first pattern. The severity of the second weather type is greater than that of the first type of weather.

4. The method according to claim 3, characterized in that, The motion recognition strategy for determining the target vehicle based on the vehicle tilt angle or the road surface type includes: When the vehicle tilt angle is greater than the first preset tilt angle, the action recognition strategy of the target vehicle is determined as the second recognition strategy, which is a repetitive action recognition strategy. When the vehicle tilt angle is less than the second preset tilt angle, the action recognition strategy of the target vehicle is determined as the third recognition strategy, which is the preset recognition strategy. When the road surface type is a preset road surface type, the action recognition strategy of the target vehicle is determined as the fourth recognition strategy, which is a delayed action recognition strategy. Among them, the rear door recognition strategies corresponding to the second recognition strategy, the third recognition strategy, and the fourth recognition strategy all belong to the first type of recognition strategy.

5. The method according to claim 4, characterized in that, Before controlling the projection component of the target vehicle to project a first pattern on the ground based on the projection parameters corresponding to the action recognition strategy, the method further includes: When the target vehicle's motion recognition strategy is the second recognition strategy, the projection parameters are determined based on the vehicle tilt angle, the road surface type, and the distance between the projection component and the ground. When the target vehicle's action recognition strategy is the third recognition strategy, the preset projection parameters are determined as the projection parameters; When the target vehicle's motion recognition strategy is the fourth recognition strategy, the target projection size and target projection brightness of the projected pattern are determined based on the road surface type; the projection parameters are determined based on the target projection size and the target projection brightness.

6. The method according to any one of claims 1-5, characterized in that, The step of controlling the projection component of the target vehicle to project a first pattern on the ground based on the projection parameters corresponding to the action recognition strategy includes: Based on the projection parameters, determine the component control parameters of the projection component; The component control parameters are used to control the projection component to project the first pattern onto the ground.

7. The method according to any one of claims 1-5, characterized in that, The sound uses the action recognition strategy to recognize the opening action of the tailgate, and controls the tailgate of the target vehicle according to the recognition result, including: When the action recognition strategy is the second recognition strategy, the number of times the tailgate opening action is executed within the range indicated by the first pattern is identified; when the number of executions is greater than or equal to a preset number, the tailgate is controlled to open; when the number of executions is less than the preset number, the tailgate opening action is not responded to. When the action recognition strategy is the third recognition strategy, it is determined whether the tailgate opening action is within a preset position range of the first pattern; if the tailgate opening action is within the preset position range, the tailgate is controlled to open; if the tailgate opening action is not within the preset position range, the tailgate opening action is not responded to. When the action recognition strategy is the fourth recognition strategy, the duration of the tailgate opening action within the first pattern indication range is recognized; if the duration is greater than or equal to a preset duration, the tailgate is controlled to open; if the duration is less than the preset duration, the tailgate opening action is not responded to.

8. The method according to claim 1, characterized in that, The method further includes: In response to detecting the vehicle key of the target vehicle, it is determined that a user is approaching the target vehicle; Alternatively, in response to detecting a target mobile terminal, it is determined that a user is approaching the target vehicle, wherein the target mobile terminal is a mobile terminal that is bound to the target vehicle; Alternatively, in response to identifying a target user from the environmental information, it is determined that a user is approaching the target vehicle, and the target user is a user with a binding relationship to the target vehicle.

9. The method according to claim 1, characterized in that, The method further includes: When the action recognition strategy belongs to the second type of recognition strategy, the projection component of the target vehicle is controlled to project a second pattern on the ground. The second pattern is used to indicate that the rear door of the target vehicle cannot be opened by the pattern projected by the projection component.

10. A vehicle, characterized in that, The vehicles include: Memory, used to store executable program code; A processor is configured to call and run the executable program code from the memory, causing the vehicle to perform the vehicle tailgate control method as described in any one of claims 1 to 9.

Images