Vehicle door control method and device of target vehicle, medium and electronic equipment

Abstract

The invention discloses a vehicle door control method and device of a target vehicle, a medium and electronic device.The method comprises the steps that vehicle driving data of the target vehicle in the driving process is obtained; if it is determined that the target vehicle has a parking intention based on the vehicle driving data, vehicle door sensing data of at least one vehicle door on the target vehicle are obtained, environment information of each vehicle door is detected based on the vehicle door sensing data, and the environment information is used for indicating elements which exist in a preset space of the vehicle door and affect getting on and off of a user; the preset space is a space allowed by a user to get on and off from the door; determining an avoidance type of the vehicle door based on the environment information of the vehicle door; and generating a whole vehicle control strategy for the target vehicle based on the avoidance type of each vehicle door. Through the technical scheme provided by the invention, the vehicle door control accuracy of the target vehicle can be improved.

Description

Technical Field

[0001] The present application belongs to the field of vehicle control technology, and in particular relates to a door control method, device, medium and electronic equipment for a target vehicle. Background Art

[0002] With the rapid development of intelligent transportation systems and autonomous driving technologies, intelligent parking technology has gradually become a research hotspot. Currently, after a vehicle is parked, it passively responds to door opening commands to detect static obstacles at the door, outputting whether the door is obstructed or not. Dynamic obstacles are treated as static obstacles, resulting in a high misjudgment rate and a poor user experience.

[0003] Based on this, the low accuracy of target vehicle door control is a technical problem that needs to be solved urgently. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, medium, and electronic device for controlling the doors of a target vehicle, thereby improving the accuracy of controlling the doors of the target vehicle to at least a certain extent.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by practice of the present application.

[0006] According to a first aspect of an embodiment of the present application, a door control method for a target vehicle is provided, the method comprising:

[0007] Acquire vehicle driving data of the target vehicle during driving;

[0008] If it is determined based on the vehicle driving data that the target vehicle intends to park, door sensing data of at least one door of the target vehicle is obtained, and for each door, environmental information of the door is detected based on the door sensing data, wherein the environmental information is used to indicate factors in a preset space of the door that affect a user getting on or off the vehicle, where the preset space is a space allowed for a user to get on or off the vehicle;

[0009] determining an avoidance type of the vehicle door based on environmental information of the vehicle door, wherein the avoidance type is used to indicate a strategy adopted to ensure that a user safely gets on and off the vehicle from the vehicle door;

[0010] A vehicle control strategy is generated for the target vehicle based on the avoidance type of each door. The vehicle control strategy is a strategy adopted to ensure that the user can safely get on and off the target vehicle door to be opened.

[0011] In some embodiments of the present application, based on the aforementioned solution, determining whether the target vehicle has a parking intention based on vehicle driving data includes:

[0012] detecting whether a vehicle lateral displacement rate of the target vehicle falls within a target threshold interval, and comparing a longitudinal deceleration gradient of the target vehicle with a first preset threshold, wherein the vehicle driving data includes the vehicle lateral displacement rate and the longitudinal deceleration gradient;

[0013] If the vehicle's lateral displacement rate falls within a target threshold interval and the longitudinal deceleration gradient is greater than or equal to a first preset threshold, the vehicle driving data is input into an intention prediction model to obtain a parking intention probability output by the intention prediction model, wherein the intention prediction model is constructed based on a deep neural network, and the intention prediction model is used to analyze the time-series dependency characteristics of the target vehicle's steering angle parameters, wheel speed parameters, and vehicle stability parameters to obtain a target confidence level that the target vehicle has a parking intention, and output the target confidence level as the parking intention probability, wherein the vehicle driving data includes steering angle parameters, wheel speed parameters, and vehicle stability parameters;

[0014] If the parking intention probability is greater than or equal to the second preset threshold, it is determined that the target vehicle has a parking intention.

[0015] In some embodiments of the present application, based on the aforementioned solution, obtaining door sensing data of at least one door on a target vehicle includes: obtaining door sensing data of the door through a sensor array deployed on the door;

[0016] Detecting environmental information of a vehicle door based on vehicle door perception data includes: inputting the vehicle door perception data into an environmental detection model to obtain the environmental information of the vehicle door, wherein the environmental detection model includes an input layer, a pre-detector, a feature classifier, a fusion layer and an output layer, the input layer is used to receive the vehicle door perception data, the pre-detector is used to detect whether there is a target element in a preset space that affects the user getting on and off the vehicle based on the vehicle door perception data, the feature classifier is used to detect the element type of the target element based on the vehicle door perception data, the fusion layer is used to fuse the target element and the element type into environmental information, and the output layer is used to output the environmental information.

[0017] In some embodiments of the present application, based on the aforementioned solution, detecting whether there are target elements in a preset space that affect the user getting on and off the vehicle based on vehicle door sensing data includes:

[0018] If a reference element exists within the detection range of the sensor array based on the door sensing data, detecting whether the reference element falls within a preset space based on a reference distance parameter and a reference height parameter of the reference element, wherein the reference distance parameter is the shortest distance between the reference element and the target door, and the reference height parameter is the height of the reference element relative to the road surface, and the door sensing data includes the reference distance parameter and the reference height parameter;

[0019] If it is detected that the reference feature falls into the preset space, the reference feature is determined as the target feature and it is determined that the target feature exists in the preset space;

[0020] If it is detected that no reference element exists in the detection range of the sensor array, or if it is detected that the reference element does not fall into the preset space, it is determined that no target element exists in the preset space.

[0021] In some embodiments of the present application, based on the aforementioned solution, before detecting whether the reference element falls within the preset space based on the reference distance parameter and the reference height parameter of the reference element, the method further includes:

[0022] Construct the space that the door passes through from closing to opening to the maximum angle as the door movement space;

[0023] Construct the space where the door motion space is projected onto the road surface to obtain the door motion projection space;

[0024] The door motion space and the motion projection space are determined as preset spaces.

[0025] In some embodiments of the present application, based on the aforementioned solution, detecting whether a reference element falls within a preset space based on a reference distance parameter and a reference height parameter of the reference element includes:

[0026] Construct a three-dimensional coordinate system with the door center as the origin;

[0027] Convert the reference distance parameter and the reference height parameter into coordinates in a three-dimensional coordinate system to obtain reference distance coordinates and reference height coordinates;

[0028] Add preset space, reference distance coordinates and reference height coordinates to the 3D coordinate system;

[0029] If the reference distance coordinate or the reference height coordinate falls within the preset space, it is determined that the reference element falls within the preset space;

[0030] If both the reference distance coordinate and the reference height coordinate do not fall within the preset space, it is determined that the reference element does not fall within the preset space.

[0031] In some embodiments of the present application, based on the aforementioned solution, determining the avoidance type of the vehicle door based on the environmental information of the vehicle door includes:

[0032] The avoidance type corresponding to the environmental information is searched from the first corresponding relationship as the avoidance type of the vehicle door, wherein the first corresponding relationship is used to record environmental information and avoidance types having a corresponding relationship.

[0033] In some embodiments of the present application, based on the aforementioned solution, a vehicle control strategy is generated for the target vehicle based on the avoidance type of each door, including:

[0034] Determine a door to be opened among the vehicle doors as a target door to obtain at least one set of target doors and target avoidance types having a corresponding relationship;

[0035] If a first avoidance type exists in at least one set of corresponding target doors and target avoidance types, determining a first control strategy corresponding to the first avoidance type as the vehicle control strategy, wherein the first control strategy includes prohibiting the target door from being opened;

[0036] If the first avoidance type does not exist in at least one set of corresponding target doors and target avoidance types, for each target door, the target control strategy corresponding to the target avoidance type is obtained to obtain the target control strategy for each target door; the target control strategy for each target door is integrated to obtain the control strategy for the entire vehicle.

[0037] According to a second aspect of an embodiment of the present application, a door control device for a target vehicle is provided, the device comprising:

[0038] An acquisition module is used to acquire vehicle driving data of a target vehicle during driving;

[0039] a processing module configured to, if it is determined based on the vehicle driving data that the target vehicle has an intention to park, obtain door sensing data for at least one door of the target vehicle, and, for each door, detect door environmental information based on the door sensing data, wherein the environmental information indicates factors within a preset space of the door that affect a user getting on or off the vehicle, the preset space being a space permitted for a user to get on or off the vehicle;

[0040] A first determining module is configured to determine an avoidance type of the vehicle door based on environmental information of the vehicle door, wherein the avoidance type is used to indicate a strategy for ensuring that a user safely gets on and off the vehicle from the vehicle door;

[0041] The generation module is used to generate a vehicle control strategy for the target vehicle based on the avoidance type of each door, wherein the vehicle control strategy is a strategy adopted to ensure that the user safely gets on and off the target vehicle door to be opened.

[0042] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by any method of the first aspect above.

[0043] According to the fourth aspect of the embodiments of the present application, an electronic device is provided, which includes one or more processors and one or more memories, wherein at least one computer program instruction is stored in the one or more memories, and the at least one computer program instruction is loaded and executed by the one or more processors to implement the method of any embodiment of the first aspect above.

[0044] In this application, when a target vehicle is detected to be attempting to stop while in motion, environmental information indicating factors affecting user entry and exit within the permitted space for entry and exit is detected based on the vehicle's door perception data. Based on this environmental information, an avoidance strategy is determined to indicate a strategy for ensuring safe entry and exit from the vehicle, ultimately resulting in a vehicle control strategy for the target vehicle. In other words, while the target vehicle is in motion, by real-time detection of factors affecting user entry and exit within the permitted space for entry and exit, a strategy is generated to ensure the door opens normally, allowing the user to enter and exit safely. This not only improves the accuracy of the target vehicle's door control, but also enhances the safety of user entry and exit.

[0045] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are incorporated into and constitute a part of the specification, illustrating embodiments consistent with the present application and, together with the specification, explaining the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can derive other drawings based on these drawings without inventive effort. In the drawings:

[0047] Figure 1 A flow chart showing a door control method for a target vehicle in an embodiment of the present application is shown;

[0048] Figure 2 A schematic diagram showing a preset space in an embodiment of the present application is shown;

[0049] Figure 3 A demonstration diagram of detecting target elements by an environment detection model in an embodiment of the present application is shown;

[0050] Figure 4 A block diagram of a door control device of a target vehicle in an embodiment of the present application is shown;

[0051] Figure 5 A schematic structural diagram of an electronic device in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0053] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0054] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0055] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0056] It should be noted that the term "plurality" as used herein refers to two or more. The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described.

[0057] In order to make those skilled in the art better understand this application, first combine Figure 1 A brief description of the application scenarios involved in this application is given.

[0058] Reference Figure 1 , shows a flow chart of a door control method for a target vehicle in an embodiment of the present application, the door control method for the target vehicle can be executed by a device having a computing and processing function, with reference to Figure 1 As shown, the door control method of the target vehicle includes:

[0059] Step 101: Acquire vehicle driving data of a target vehicle during driving;

[0060] Step 102: If it is determined based on the vehicle driving data that the target vehicle intends to park, door sensing data of at least one door of the target vehicle is obtained. For each door, environmental information of the door is detected based on the door sensing data. The environmental information indicates factors within a preset space of the door that affect a user getting on or off the vehicle. The preset space is the space allowed for a user to get on or off the vehicle.

[0061] Step 103: determining an avoidance type of the vehicle door based on the environmental information of the vehicle door, wherein the avoidance type is used to indicate a strategy for ensuring that the user safely gets on and off the vehicle from the vehicle door;

[0062] Step 104: Generate a vehicle control strategy for the target vehicle based on the avoidance type of each door, wherein the vehicle control strategy is a strategy adopted to ensure that the user safely gets on and off the target vehicle door to be opened.

[0063] Through the above steps, when the target vehicle detects an intention to stop while in motion, environmental information indicating factors affecting user entry and exit within the permitted space for entry and exit is detected based on the vehicle's door perception data. Based on this environmental information, an avoidance strategy is determined to ensure safe entry and exit from the vehicle, ultimately resulting in a vehicle control strategy for the target vehicle. In other words, while the target vehicle is in motion, by real-time detecting factors affecting user entry and exit within the permitted space for entry and exit, a strategy is generated to ensure the door opens normally, allowing for safe entry and exit from the vehicle. This not only improves the accuracy of the target vehicle's door control, but also enhances the safety of user entry and exit.

[0064] In the embodiment provided in step 101, the above-mentioned vehicle driving data are dynamic parameters of the target vehicle during driving, which may include but are not limited to: vehicle motion parameters such as vehicle speed, vehicle acceleration (for example: longitudinal deceleration gradient), lateral displacement rate (for example: vehicle lateral moving speed); vehicle control parameters such as steering wheel angle, wheel speed; vehicle stability parameters such as yaw angular velocity, body roll angle, etc.

[0065] In the embodiment provided in step 102, the original environmental data can be collected as the above-mentioned door perception data by, but is not limited to, door sensors, binocular vision systems, V2X roadside units and other devices deployed on the vehicle.

[0066] Optionally, in an embodiment of the present application, data on factors affecting users getting on and off the vehicle can be extracted from the vehicle door sensing data, but is not limited to. If the vehicle door sensing data includes data on factors affecting users getting on and off the vehicle, environmental information of the vehicle door is generated based on the data on factors affecting users getting on and off the vehicle; or, if the vehicle door sensing data does not include data on factors affecting users getting on and off the vehicle, environmental information is generated to indicate factors affecting users getting on and off the vehicle that do not exist in the preset space of the vehicle door.

[0067] Optionally, in an embodiment of the present application, the above-mentioned factors affecting users getting on and off the vehicle may include but are not limited to obstacles, traffic flow, pedestrian flow, road slope, etc.

[0068] Optionally, in an embodiment of the present application, vehicle driving data of the target vehicle during driving is obtained in real time. Since whether the target vehicle has an intention to park is determined based on the vehicle driving data of the target vehicle during driving, a forward-looking prediction of parking behavior is achieved, that is, the evaluation process is started in advance when the target vehicle has an intention to park.

[0069] In one embodiment of the present application, it is possible but not limited to determine whether the target vehicle has an intention to park based on vehicle driving data in the following manner: detecting whether the vehicle lateral displacement rate of the target vehicle falls within a target threshold interval, and comparing the longitudinal deceleration gradient of the target vehicle with a first preset threshold, wherein the vehicle driving data includes the vehicle lateral displacement rate and the longitudinal deceleration gradient; if the vehicle lateral displacement rate falls within the target threshold interval, and the longitudinal deceleration gradient is greater than or equal to the first preset threshold, inputting the vehicle driving data into an intention prediction model to obtain a parking intention probability output by the intention prediction model, wherein the intention prediction model is constructed based on a deep neural network, and the intention prediction model is used to analyze the time-series dependence characteristics of the steering angle parameters, wheel speed parameters and vehicle stability parameters of the target vehicle to obtain a target confidence that the target vehicle has an intention to park, and output the target confidence as a parking intention probability, and the vehicle driving data includes steering angle parameters, wheel speed parameters and vehicle stability parameters; if the parking intention probability is greater than or equal to the second preset threshold, it is determined that the target vehicle has an intention to park.

[0070] Optionally, in this embodiment, a dual verification mechanism of the lateral displacement rate threshold interval and the longitudinal deceleration gradient is used to determine the timing of waking up the intention prediction model used to detect parking intentions, thereby reducing the possibility of misjudgment. Combined with the intention prediction model based on a deep neural network, accurate identification of parking intentions is achieved. In other words, this application quickly screens potential parking scenarios through dynamic parameter thresholds, then analyzes the timing characteristics of parameters such as steering angle and wheel speed through an LSTM (Long Short-Term Memory) timing model, and ultimately outputs a probabilistic judgment result. From passively responding to door opening to predicting parking intentions, the door opening strategy is constructed in advance, improving the efficiency of door control.

[0071] Optionally, in this embodiment, the target threshold range may be set to, but not limited to, 0.5 m / s to 1.2 m / s; or, the target threshold range may be determined by performing a cluster analysis based on actual application scenarios of the target vehicle.

[0072] Optionally, in this embodiment, the first preset threshold may be set to, but is not limited to, 0.3 g; or, the first preset threshold may be determined by performing a cluster analysis based on actual application scenarios of the target vehicle.

[0073] Optionally, in this embodiment, the above-mentioned intention prediction model can be trained but not limited to through the following process: Construct an initial prediction model: obtain an LSTM for processing time-series dependent features as the initial prediction model, and set the initial prediction model to use the Attention mechanism to enhance the weight of key time steps (such as: the moment of sudden braking). Further, the initial prediction model includes an input layer, a time-series feature extraction layer, an attention layer, a fully connected classification layer and an output layer, wherein the input layer is used to standardize the input data and process missing values; the time-series feature extraction layer is used to capture the time-series dependencies (such as the causality of turning after deceleration) and output the hidden state of each time step; the attention layer is used to focus on key time steps (such as the moment of sudden braking or large turning) to calculate the attention weight and output a condensed representation of the time-series information; the fully connected classification layer is used to use binary cross entropy as the loss function and the Sigmoid activation function to calculate the confidence; the output layer is used to determine the confidence as the probability of parking intention and output it.

[0074] Construct a training set for training the initial prediction model: Onboard sensors (such as the IMU, wheel speed sensors, and steering angle sensors) are used to collect data from both parking and non-parking scenarios. Extract temporal features from this data: A sliding window is used to extract the mean, variance, and trend of parameters such as lateral displacement rate, longitudinal deceleration gradient, and steering angle. Contextual features of the data are also extracted: gear status (P / N / D), turn signal signal, and location data (e.g., whether the GPS location is near a parking lot). Label the data with "parking intention" (a binary classification of 1 / 0) to create a training set.

[0075] Train the initial prediction model using the training set: Set the learning rate of the initial prediction model to 0.001 and train the initial prediction model using the training set. Terminate training when the accuracy of the initial prediction model exceeds 95% or the initial prediction model completes the training set. The trained initial prediction model is designated as the intent prediction model.

[0076] It should be noted that the training process of the above-mentioned intention prediction model can be supplemented with corresponding model verification and scenario-based testing based on actual application scenarios. This intention prediction model uses a deep learning algorithm to comprehensively analyze vehicle driving data (such as lateral displacement, deceleration gradient, and steering angle) to proactively predict the driver's parking intention, reserving sufficient response time for door safety control. This effectively avoids the misjudgment caused by traditional methods based on single signals or simple rules, ensuring the real-time and safety of the system.

[0077] Optionally, in this embodiment, the parking intention probability is a value between 0 and 1.

[0078] Optionally, in this embodiment, the second preset threshold may be set to, but is not limited to, 0.5; or, the second preset threshold may be determined by performing a cluster analysis based on actual application scenarios of the target vehicle.

[0079] Optionally, in this embodiment, if the parking intention probability is less than a second preset threshold, it is determined that the target vehicle has no parking intention.

[0080] In one embodiment of the present application, the door sensing data of at least one door on the target vehicle may be obtained in the following manner, but is not limited to: obtaining the door sensing data of the door through a sensor array deployed on the door.

[0081] Optionally, in this embodiment, a sensor array is deployed on each door of the target vehicle. The door perception data of the door obtained through the sensor array deployed on the door may include, but is not limited to: obtaining echo signals through millimeter-wave radar, obtaining digital images through a binocular vision system, obtaining roadside traffic events and traffic sign messages through a V2X roadside unit, etc.

[0082] In one embodiment of the present application, the environmental information of the vehicle door can be detected based on the vehicle door perception data in the following manner, but is not limited to: the vehicle door perception data is input into the environmental detection model to obtain the environmental information of the vehicle door, wherein the environmental detection model includes an input layer, a pre-detector, a feature classifier, a fusion layer and an output layer, the input layer is used to receive the vehicle door perception data, the pre-detector is used to detect whether there is a target element in the preset space that affects the user getting on and off the vehicle based on the vehicle door perception data, the feature classifier is used to detect the element type of the target element based on the vehicle door perception data, the fusion layer is used to fuse the target element and the element type into environmental information, and the output layer is used to output the environmental information.

[0083] Optionally, in this embodiment, the door sensing data of each door can be input into the environmental detection model in turn, but is not limited to, to obtain the environmental information of each door output by the environmental detection model; or, multiple groups of corresponding doors and door sensing data can be input into the environmental detection model to obtain multiple groups of corresponding doors and environmental information output by the environmental detection model.

[0084] Optionally, in this embodiment, the environment detection model may be trained in, but is not limited to, the following manner: constructing an initial detection model including an input layer, a pre-detector, a feature classifier, a fusion layer, and an output layer;

[0085] Constructing a training set: Initial perception data from the sensor arrays of different vehicle doors is collected in various scenarios through simulated or actual driving. The initial perception data is annotated to determine whether there are target elements that affect user entry and exit, as well as the element types of these target elements. These scenarios include, but are not limited to, the presence of one or more elements within the pre-set space of the vehicle door that affect user entry and exit. These elements include static obstacles such as curbs, fire hydrants, and walls of varying heights and angles; dynamic obstacles such as pedestrians, bicycles, and pets; environmental conditions such as rainy days, nighttime, strong light interference, accumulated water, road slope, and potholes; and regulatory restrictions such as time-limited parking spaces. These element types include, but are not limited to, static obstacle types, super-high static obstacle types, non-motorized vehicles occupying the road, dynamic obstacle types, regulatory constraints, environmental conditions (accumulated water), environmental conditions (icing), environmental conditions (slope), and environmental conditions (lighting).

[0086] Use the training set to train the initial detection model until a trained environment detection model is obtained.

[0087] Optionally, in this embodiment, the above-mentioned pre-detector is used to determine whether there is a target element that affects getting on and off the vehicle within the preset space of the vehicle door. As shown in Table 1, the pre-detector can be provided with, but is not limited to, the function of identifying elements based on the acquired vehicle door sensing data (for example, identifying whether there are fire hydrants, streetlight poles, etc. in the image data based on the image data) and detecting whether the identified elements fall within the preset space (for example, identifying the positions of fire hydrants and streetlight poles in the image data based on the radar raw data). It is conceivable that for static elements, the pre-detector is used to identify the elements, and then monitor the relative position of the elements and the vehicle from the subsequently input vehicle door sensing data until the elements fall within the preset space, and determine that the target elements that affect getting on and off the vehicle are detected in the preset space; for dynamic elements, the pre-detector can predict the flow direction of the dynamic elements based on the vehicle door sensing data at adjacent moments, thereby determining whether to continue monitoring the elements.

[0088] Table 1

[0089]

[0090] Optionally, in this embodiment, the feature classifier is used to identify the element type of the target element. As shown in Table 1, the feature classifier can be trained, but is not limited to, to classify the identified target element according to judgment criteria and determine its element type. For example, if the pre-detector identifies the target element as a fire hydrant, the feature classifier determines whether its height exceeds 15 cm. If the fire hydrant's height exceeds 15 cm, its element type is determined to be an ultra-high static obstacle type; alternatively, if the fire hydrant's height does not exceed 15 cm, its element type is determined to be a static obstacle type.

[0091] That is, the judgment criteria shown in Table 1 are used to determine whether the identified element is a target element that affects the user's getting on and off the vehicle after the element has been identified. The specific method for identifying the element is to train the pre-detector to have the function of identifying the element based on the acquired door sensing data. For example, the pre-detector determines that there is ice on the road surface (the element is ice) through the temperature sensor combined with visual reflection analysis. If the area of ice exceeds 0.5m 2 Icing is identified as the target factor that affects users getting on and off the vehicle, and the factor type is environmental condition type - icing.

[0092] In one embodiment of the present application, it is possible but not limited to detect whether there is a target element in a preset space that affects users getting on and off the vehicle based on the vehicle door sensing data in the following manner: if it is detected based on the vehicle door sensing data that there is a reference element in the detection range of the sensor array, detect whether the reference element falls into the preset space based on the reference distance parameter and reference height parameter of the reference element, wherein the reference distance parameter is the shortest distance between the reference element and the target vehicle door, the reference height parameter is the height of the reference element relative to the road surface, and the vehicle door sensing data includes the reference distance parameter and the reference height parameter; if it is detected that the reference element falls into the preset space, determine the reference element as the target element and determine that the target element exists in the preset space; if it is detected that the reference element does not exist in the detection range of the sensor array, or it is detected that the reference element does not fall into the preset space, determine that the target element does not exist in the preset space.

[0093] In one embodiment of the present application, before detecting whether a reference element falls into a preset space based on a reference distance parameter and a reference height parameter of the reference element, the preset space can be generated in the following manner, but is not limited to: constructing the space that the door passes through from closing to opening to the maximum angle as the door movement space; constructing the space where the door movement space is projected onto the road surface to obtain the door movement projection space; and determining the door movement space and the movement projection space as the preset space.

[0094] Optionally, in this embodiment, Figure 2 A schematic diagram showing the preset space in the embodiment of the present application is shown, referring to Figure 2 As shown, the three-dimensional space of the door opening path simulated from closing to opening to the maximum angle is used as the door movement space, and the three-dimensional space of the door movement space projected onto the road surface is used as the motion projection space, thereby obtaining a preset space including the door movement space and the motion projection space.

[0095] In one embodiment of the present application, it is possible to detect whether the reference element falls into the preset space based on the reference distance parameter and reference height parameter of the reference element in the following manner, but not limited to: construct a three-dimensional coordinate system with the door center of the vehicle door as the origin; convert the reference distance parameter and the reference height parameter into coordinates in the three-dimensional coordinate system to obtain reference distance coordinates and reference height coordinates; add the preset space, reference distance coordinates and reference height coordinates to the three-dimensional coordinate system; if the reference distance coordinate or the reference height coordinate falls into the preset space, it is determined that the reference element falls into the preset space; if neither the reference distance coordinate nor the reference height coordinate falls into the preset space, it is determined that the reference element does not fall into the preset space.

[0096] In the embodiment provided in step 103, the corresponding environmental information and avoidance types may be pre-built, but is not limited to; or the avoidance type may be detected in real time according to the environmental information.

[0097] In one embodiment of the present application, the avoidance type of the vehicle door can be determined based on the environmental information of the vehicle door in the following manner, but is not limited to: searching the avoidance type corresponding to the environmental information from the first correspondence as the avoidance type of the vehicle door, wherein the first correspondence is used to record the environmental information and avoidance type with a corresponding relationship.

[0098] Optionally, in this embodiment, as shown in Table 2, the avoidance types may be divided into, but not limited to, primary, secondary, and tertiary levels, and the first correspondence includes environmental information and avoidance types (primary, secondary, and tertiary) having a corresponding relationship.

[0099] Table 2

[0100]

[0101] In the embodiment provided in step 104, the target door can be determined in a variety of ways, but is not limited to, such as: determining the door closest to each target user as the target door to obtain one or more target doors, specifically, collecting the gravity sensing parameters of each vehicle seat of the target vehicle through a gravity sensor, wherein a gravity sensor is deployed on each vehicle seat of the target vehicle; if the gravity sensing parameter is greater than or equal to a preset gravity threshold (set according to the actual scenario), the vehicle seat is determined to be the target vehicle seat where the target user is sitting; and searching the target correspondence for a door that has a corresponding relationship with the target vehicle seat as the target door, wherein the target correspondence is used to record the correspondence between each vehicle seat and the door closest to the vehicle seat.

[0102] It should be noted that for a target vehicle, a user with vehicle control rights (e.g., the driver) is allowed to set one or more doors to always be closed, i.e., the target correspondence does not include one or more doors that are always closed. Alternatively, a user may be allowed to open only one or more specific doors, i.e., the target correspondence only includes the permission to open one or more specific doors.

[0103] In one embodiment of the present application, a whole vehicle control strategy can be generated for a target vehicle based on the avoidance type of each door in the following manner, but is not limited to: a door to be opened among the doors is determined as a target door to obtain at least one set of target doors and target avoidance types with corresponding relationships; if a first avoidance type exists in at least one set of target doors and target avoidance types with corresponding relationships, a first control strategy corresponding to the first avoidance type is determined as the whole vehicle control strategy, wherein the first control strategy includes prohibiting the opening of the target door; if the first avoidance type does not exist in at least one set of target doors and target avoidance types with corresponding relationships, for each target door, a target control strategy corresponding to the target avoidance type is obtained to obtain a target control strategy for each target door; and the target control strategy for each target door is integrated to obtain a whole vehicle control strategy.

[0104] Optionally, in this embodiment, a corresponding control strategy is set for each avoidance type, including: the first control strategy for the first avoidance type is to prohibit opening the door; the second control strategy for the second avoidance type is to control the target door opening angle; the third control strategy for the third avoidance type is to allow opening the door, but priority is given to broadcasting a "recommendation to avoid" reminder in the car. Furthermore, it can also be broadcast in combination with environmental information, such as "Be careful of scratches when opening the door on a curb, it is recommended to avoid."

[0105] Optionally, in this embodiment, the target angle for allowing the target door to open can be detected in the following ways, but is not limited to: introducing a dynamic simulation algorithm for the door motion envelope, constructing a four-dimensional space-time model (three-dimensional space + time dimension) of the door opening process through inverse kinematics calculations to predict the minimum safe distance between the target door and the obstacle, and determining the target angle based on the minimum safe distance.

[0106] Optionally, in this embodiment, if the vehicle control strategy includes a third control strategy, it can generate, but is not limited to, multimodal navigation suggestions based on the location information of the elements, such as: dynamically recommending the best alternative parking location, and providing a progressive guidance path (the typical recommended distance is 15-30m forward of the current residence), etc.

[0107] In order to make those skilled in the art better understand the door control method of the target vehicle, the following will be combined with Figure 3 For illustration, take the target threshold range as 0.5m / s-1.2m / s, the first preset threshold as 0.3g, and the second preset threshold as 0.5 as an example.

[0108] Obtain vehicle driving data of the target vehicle during driving.

[0109] Extract the vehicle's lateral displacement rate and longitudinal deceleration gradient from the vehicle's driving data. If the vehicle's lateral displacement rate falls between 0.5m / s and 1.2m / s, and the longitudinal deceleration gradient is greater than or equal to 0.3g, input the vehicle's driving data into the intention prediction model to obtain the stopping intention probability output by the intention prediction model.

[0110] If the parking intention probability is greater than or equal to 0.5, the door perception data of each door on the target vehicle is collected through the sensor array deployed on each door.

[0111] Reference Figure 3 , showing a demonstration diagram of the environmental detection model detecting target elements in an embodiment of the present application. The door sensing data for each vehicle door is input into the environmental detection model, which detects whether an element exists within the detection range of the sensor array of each vehicle door. Based on the element's location information and the location information of the preset space, the model determines whether the element falls within the preset space. If the element falls within the preset space, the element is identified as a target element, its element type is detected, and the model ultimately outputs environmental information (including the target element and element type).

[0112] The avoidance type of each door is determined based on the environmental information; the avoidance type of the target door is then integrated to obtain the vehicle control strategy.

[0113] In the door control method of the target vehicle proposed in this application, the passive mode of "after-the-fact remediation" of traditional door collision prevention is changed, and an active control strategy of "perception-prediction-protection" is established. From flexible reminders to limit and then to forced locking, a hierarchical control strategy is implemented for different sources of danger, thereby improving the accuracy and control efficiency of the door control of the target vehicle.

[0114] The following describes an embodiment of the device of the present application, which can be used to execute the door control method of the target vehicle in the above-mentioned embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the door control method of the target vehicle in the above-mentioned embodiment of the present application.

[0115] See also Figure 4 , shows a block diagram of the door control device of the target vehicle in an embodiment of the present application.

[0116] like Figure 4 As shown, the door control device (400) of the target vehicle according to an embodiment of the present application includes: an acquisition module 401, a processing module 402, a first determination module 403, and a generation module 404.

[0117] The acquisition module 401 is used to acquire the vehicle driving data of the target vehicle during the driving process;

[0118] Processing module 402 is configured to, if it is determined based on the vehicle driving data that the target vehicle intends to park, obtain door sensing data for at least one door of the target vehicle, and detect, for each door, door environment information based on the door sensing data, where the environment information indicates factors within a preset space of the door that affect a user getting on or off the vehicle, where the preset space is a space permitted for a user to get on or off the vehicle;

[0119] A first determining module 403 is configured to determine an avoidance type of the vehicle door based on environmental information of the vehicle door, wherein the avoidance type indicates a strategy for ensuring that a user safely gets on and off the vehicle from the vehicle door;

[0120] The generation module 404 is used to generate a vehicle control strategy for the target vehicle based on the avoidance type of each door, wherein the vehicle control strategy is a strategy adopted to ensure that the user safely gets on and off the target vehicle door to be opened.

[0121] In some embodiments of the present application, based on the aforementioned scheme, the processing module 402 is configured to: detect whether the vehicle lateral displacement rate of the target vehicle falls within the target threshold interval, and compare the longitudinal deceleration gradient of the target vehicle with a first preset threshold, wherein the vehicle driving data includes the vehicle lateral displacement rate and the longitudinal deceleration gradient; if the vehicle lateral displacement rate falls within the target threshold interval, and the longitudinal deceleration gradient is greater than or equal to the first preset threshold, the vehicle driving data is input into the intention prediction model to obtain the parking intention probability output by the intention prediction model, wherein the intention prediction model is constructed based on a deep neural network, and the intention prediction model is used to analyze the time-series dependence characteristics of the steering angle parameters, wheel speed parameters and vehicle stability parameters of the target vehicle to obtain the target confidence that the target vehicle has the intention to park, and output the target confidence as the parking intention probability, and the vehicle driving data includes the steering angle parameters, wheel speed parameters and vehicle stability parameters; if the parking intention probability is greater than or equal to the second preset threshold, it is determined that the target vehicle has the intention to park.

[0122] In some embodiments of the present application, based on the aforementioned scheme, the processing module 402 is further configured to: obtain door perception data of the vehicle door through a sensor array deployed on the vehicle door; detect environmental information of the vehicle door based on the door perception data, including: inputting the door perception data into an environmental detection model to obtain environmental information of the vehicle door, wherein the environmental detection model includes an input layer, a pre-detector, a feature classifier, a fusion layer and an output layer, the input layer is used to receive the door perception data, the pre-detector is used to detect whether there is a target element in a preset space that affects the user getting on and off the vehicle based on the door perception data, the feature classifier is used to detect the element type of the target element based on the door perception data, the fusion layer is used to fuse the target element and the element type into environmental information, and the output layer is used to output the environmental information.

[0123] In some embodiments of the present application, based on the aforementioned scheme, the processing module 402 is further configured to: if the presence of a reference element in the detection range of the sensor array is detected based on the vehicle door sensing data, detect whether the reference element falls into a preset space based on the reference distance parameter and reference height parameter of the reference element, wherein the reference distance parameter is the shortest distance between the reference element and the target vehicle door, the reference height parameter is the height of the reference element relative to the road surface, and the vehicle door sensing data includes the reference distance parameter and the reference height parameter; if it is detected that the reference element falls into the preset space, determine the reference element as the target element and determine that the target element exists in the preset space; if it is detected that the reference element does not exist in the detection range of the sensor array, or it is detected that the reference element does not fall into the preset space, determine that the target element does not exist in the preset space.

[0124] In some embodiments of the present application, based on the aforementioned solution, before detecting whether the reference element falls within the preset space based on the reference distance parameter and the reference height parameter of the reference element, the device further includes:

[0125] The first construction module is used to construct the space that the door passes through from closing to opening to the maximum angle as the door movement space;

[0126] The second construction module is used to construct the space where the door motion space is projected onto the road surface to obtain the door motion projection space;

[0127] The second determining module is configured to determine the door motion space and the motion projection space as a preset space.

[0128] In some embodiments of the present application, based on the aforementioned scheme, the processing module 402 is further configured to: construct a three-dimensional coordinate system with the door center of the vehicle door as the origin; convert the reference distance parameter and the reference height parameter into coordinates in the three-dimensional coordinate system to obtain the reference distance coordinate and the reference height coordinate; add the preset space, the reference distance coordinate and the reference height coordinate to the three-dimensional coordinate system; if the reference distance coordinate or the reference height coordinate falls into the preset space, it is determined that the reference element falls into the preset space; if neither the reference distance coordinate nor the reference height coordinate falls into the preset space, it is determined that the reference element does not fall into the preset space.

[0129] In some embodiments of the present application, based on the aforementioned scheme, the first determination module 403 is further configured to: search for the avoidance type corresponding to the environmental information from the first correspondence as the avoidance type of the vehicle door, wherein the first correspondence is used to record the environmental information and avoidance type with a corresponding relationship.

[0130] In some embodiments of the present application, based on the aforementioned scheme, the generation module 404 is configured to: determine the door to be opened among the vehicle doors as the target door to obtain at least one group of target doors and target avoidance types with corresponding relationships; if there is a first avoidance type in at least one group of target doors and target avoidance types with corresponding relationships, determine the first control strategy corresponding to the first avoidance type as the whole vehicle control strategy, wherein the first control strategy includes prohibiting the opening of the target door; if there is no first avoidance type in at least one group of target doors and target avoidance types with corresponding relationships, for each target door, obtain the target control strategy corresponding to the target avoidance type to obtain the target control strategy for each target door; and integrate the target control strategy of each target door to obtain the whole vehicle control strategy.

[0131] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by the previous method.

[0132] Based on the same inventive concept, the present application also provides an electronic device, referring to Figure 5 , shows a structural diagram of an electronic device in an embodiment of the present application, the electronic device includes one or more memories 504, one or more processors 502 and at least one computer program (computer program instruction) stored in the memory 504 and executable on the processor 502, and the processor 1202 implements the above method when executing the computer program.

[0133] Among them, Figure 5 In the embodiment of the present invention, a bus architecture (represented by bus 500) is shown. Bus 500 may include any number of interconnected buses and bridges, and bus 500 links various circuits including one or more processors represented by processor 502 and memory represented by memory 504. Bus 500 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 505 provides an interface between bus 500 and receiver 501 and transmitter 503. Receiver 501 and transmitter 503 may be the same component, namely a transceiver, which provides a unit for communicating with various other devices over a transmission medium. Processor 502 is responsible for managing bus 500 and general processing, while memory 504 may be used to store data used by processor 502 when performing operations.

[0134] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of this application and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hardwiring, or a combination of any of these. Furthermore, the functional units may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit.

[0135] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0136] The units described as separate components may or may not be physically separate, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0137] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, and other media that can store computer program instructions.

[0138] The above are merely examples of the present application and are not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of the claims of the present application.

Claims

1. A door control method for a target vehicle, characterized in that: The method comprises: Acquire vehicle driving data of the target vehicle during driving; If it is determined based on the vehicle driving data that the target vehicle has an intention to park, obtaining door sensing data of at least one door of the target vehicle, and detecting, for each door, environmental information of the door based on the door sensing data, wherein the environmental information is used to indicate factors in a preset space of the door that affect a user getting on or off the vehicle, the preset space being a space allowed for a user to get on or off the vehicle from the door; determining an avoidance type of the vehicle door based on the environmental information of the vehicle door, wherein the avoidance type is used to indicate a strategy adopted to ensure that a user safely gets on and off the vehicle from the vehicle door; A whole vehicle control strategy is generated for the target vehicle based on the avoidance type of each door, wherein the whole vehicle control strategy is a strategy adopted to ensure that a user safely gets on and off the vehicle from the target door to be opened.

2. The method according to claim 1, characterized in that The determining, based on the vehicle driving data, that the target vehicle has a parking intention includes: detecting whether a vehicle lateral displacement rate of the target vehicle falls within a target threshold interval, and comparing a longitudinal deceleration gradient of the target vehicle with a first preset threshold, wherein the vehicle driving data includes the vehicle lateral displacement rate and the longitudinal deceleration gradient; If the lateral displacement rate of the vehicle falls within the target threshold range, and the longitudinal deceleration gradient is greater than or equal to the first preset threshold, the vehicle driving data is input into an intention prediction model to obtain a parking intention probability output by the intention prediction model, wherein the intention prediction model is constructed based on a deep neural network, and the intention prediction model is used to analyze the time-series dependency characteristics of the steering angle parameter, the wheel speed parameter, and the vehicle stability parameter of the target vehicle to obtain a target confidence level that the target vehicle has a parking intention, and output the target confidence level as the parking intention probability, and the vehicle driving data includes the steering angle parameter, the wheel speed parameter, and the vehicle stability parameter; If the parking intention probability is greater than or equal to a second preset threshold, it is determined that the target vehicle has the parking intention.

3. The method according to claim 1, characterized in that The acquiring of door sensing data of at least one door on the target vehicle comprises: acquiring the door sensing data of the door through a sensor array deployed on the door; The detecting of the environmental information of the vehicle door based on the vehicle door perception data includes: inputting the vehicle door perception data into an environmental detection model to obtain the environmental information of the vehicle door, wherein the environmental detection model includes an input layer, a pre-detector, a feature classifier, a fusion layer and an output layer, the input layer is used to receive the vehicle door perception data, the pre-detector is used to detect whether there is a target element in the preset space that affects the user getting on and off the vehicle based on the vehicle door perception data, the feature classifier is used to detect the element type of the target element based on the vehicle door perception data, the fusion layer is used to fuse the target element and the element type into the environmental information, and the output layer is used to output the environmental information.

4. The method according to claim 3, characterized in that The detecting, based on the vehicle door sensing data, whether there is a target element in the preset space that affects the user getting on and off the vehicle includes: If a reference element is detected within the detection range of the sensor array based on the vehicle door sensing data, detecting whether the reference element falls within the preset space based on a reference distance parameter and a reference height parameter of the reference element, wherein the reference distance parameter is the shortest distance between the reference element and the target vehicle door, and the reference height parameter is the height of the reference element relative to the road surface, and the vehicle door sensing data includes the reference distance parameter and the reference height parameter; If it is detected that the reference element falls into the preset space, the reference element is determined as the target element and the target element is determined to exist in the preset space; If it is detected that the reference element does not exist in the detection range of the sensor array, or if it is detected that the reference element does not fall into the preset space, it is determined that the target element does not exist in the preset space.

5. The method according to claim 4, characterized in that Before detecting whether the reference element falls within the preset space based on the reference distance parameter and the reference height parameter of the reference element, the method further includes: Constructing the space that the door passes through from closing to opening to the maximum angle as the door movement space; Constructing a space where the door motion space is projected onto a road surface to obtain the door motion projection space; The door motion space and the motion projection space are determined as the preset space.

6. The method according to claim 4, characterized in that The detecting, based on the reference distance parameter and the reference height parameter of the reference element, whether the reference element falls within the preset space includes: Constructing a three-dimensional coordinate system with the door center of the vehicle door as the origin; Converting the reference distance parameter and the reference height parameter into coordinates in the three-dimensional coordinate system to obtain reference distance coordinates and reference height coordinates; Adding the preset space, the reference distance coordinates, and the reference height coordinates to the three-dimensional coordinate system; If the reference distance coordinate or the reference height coordinate falls within the preset space, determining that the reference element falls within the preset space; If both the reference distance coordinate and the reference height coordinate do not fall within the preset space, it is determined that the reference element does not fall within the preset space.

7. The method according to claim 1, characterized in that The determining the avoidance type of the vehicle door based on the environmental information of the vehicle door includes: The avoidance type corresponding to the environmental information is searched from a first corresponding relationship as the avoidance type of the vehicle door, wherein the first corresponding relationship is used to record environmental information and avoidance types having a corresponding relationship.

8. The method according to claim 1, characterized in that Generating a vehicle control strategy for the target vehicle based on the avoidance type of each door includes: Determining a door to be opened among the vehicle doors as the target door to obtain at least one set of the target doors and target avoidance types having a corresponding relationship; If a first avoidance type exists in at least one set of corresponding target doors and target avoidance types, determining a first control strategy corresponding to the first avoidance type as the vehicle control strategy, wherein the first control strategy includes prohibiting the target door from being opened; If the first avoidance type does not exist in at least one set of corresponding target doors and target avoidance types, for each target door, the target control strategy corresponding to the target avoidance type is obtained to obtain the target control strategy of each target door; the target control strategy of each target door is integrated to obtain the whole vehicle control strategy.

9. A door control device for a target vehicle, characterized in that: The device comprises: An acquisition module is used to acquire vehicle driving data of a target vehicle during driving; a processing module configured to, if it is determined based on the vehicle driving data that the target vehicle has an intention to park, obtain door sensing data of at least one door of the target vehicle, and, for each door, detect environmental information of the door based on the door sensing data, wherein the environmental information indicates factors in a preset space of the door that affect a user getting on or off the vehicle, the preset space being a space permitted for a user to get on or off the vehicle from the door; a first determining module, configured to determine an avoidance type of the vehicle door based on the environmental information of the vehicle door, wherein the avoidance type is used to indicate a strategy adopted to ensure that a user safely gets on and off the vehicle from the vehicle door; A generation module is used to generate a whole vehicle control strategy for the target vehicle based on the avoidance type of each door, wherein the whole vehicle control strategy is a strategy adopted to ensure that the user safely gets on and off the target vehicle door to be opened.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer program instructions, which are loaded and executed by a processor to implement the operations performed by the method according to any one of claims 1 to 8.

11. An electronic device comprising a processor and a memory, characterized in that: The memory stores computer program instructions that can be executed by the processor, and when the processor executes the computer program instructions, the processor implements the method according to any one of claims 1 to 8.

Images