Method for adjusting internal temperature of vehicle and vehicle

Abstract

The invention relates to the field of vehicle temperature control, in particular to a vehicle internal temperature adjusting method and a vehicle. Comprises: detecting a target action of a target person in an internal space of a vehicle, the target action being an action formed by at least two gestures executed in sequence; and the temperature of the internal space of the vehicle is adjusted at least based on the target action of the target person and the body surface temperature of the target person, and the body surface temperature is determined by the part temperature of at least one key part of the target person. According to the method, the subjective feeling of the person on the temperature in the vehicle can be indirectly known by detecting the target action of the target person formed by at least two gestures executed in sequence; meanwhile, the body surface temperature is determined based on the part temperature of at least one key part, the physiological state difference of passengers is considered, and real-time sensing is achieved. The in-vehicle temperature is adjusted by integrating the target action and the body surface temperature, manual interaction is not needed, automatic operation can be achieved, the requirements of different passengers can be accurately met, and the problems in the prior art are effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of vehicle temperature control, and in particular to a vehicle interior temperature adjustment method and vehicle. BACKGROUND

[0002] With the rapid development of the automobile industry and the improvement of people's living standards, vehicles have become an essential means of transportation in daily life. As the core component of creating a comfortable driving environment, the performance of the vehicle air conditioning system directly affects the passenger's driving experience. Traditional vehicle air conditioning systems mostly use preset fixed temperatures or simple temperature adjustment logic, which has the advantages of easy implementation and low cost, but it is difficult to adjust the temperature individually according to the individual differences of different passengers in the vehicle and real-time environmental parameters.

[0003] In recent years, intelligent technology has driven the development of vehicle air conditioning systems towards intelligence and individualization, and technologies using machine learning or voice large models to adjust the temperature in the vehicle have emerged. Among them, the general machine learning method has poor adjustment effect, and the voice large model method relies on passenger voice interaction and cannot achieve automatic temperature adjustment. The common method based on machine learning also has the problems of ignoring the physiological state differences of passengers (such as different heat needs in sleep and active states) and lacking real-time sensing ability of passenger body temperature, which makes it difficult to meet the user's demand for comfortable and intelligent driving environment. SUMMARY

[0004] Therefore, the embodiments of the present application provide a vehicle interior temperature adjustment method and vehicle to solve the problems of poor temperature adjustment technology or adjustment effect of existing vehicle air conditioning, reliance on manual interaction for non-automatic temperature adjustment, ignoring physiological state differences of passengers and lacking real-time sensing ability of passenger body temperature, and difficulty in achieving individualized and intelligent temperature adjustment.

[0005] In a first aspect, the embodiments of the present application provide a vehicle interior temperature adjustment method, comprising:

[0006] detecting a target action of a target person in an internal space of a vehicle, the target action being an action composed of at least two postures executed in sequence;

[0007] adjusting the temperature of the internal space of the vehicle based at least on the target action of the target person and the body surface temperature of the target person, the body surface temperature being determined by the part temperature of at least one key part of the target person.

[0008] Further, before adjusting the temperature of the internal space of the vehicle based at least on the target action of the target person and the body surface temperature of the target person, the method further comprises: determining the environmental temperature characteristics of the vehicle based on the regional environmental parameters of the target location area where the vehicle is currently located and / or the in-vehicle environmental parameters of the internal space of the vehicle.

[0009] The temperature adjustment of the interior space of the vehicle based on the target action of the target person and the body surface temperature of the target person comprises:

[0010] The temperature adjustment of the interior space of the vehicle based on the target action of the target person, the body surface temperature of the target person and the environmental temperature feature; and / or,

[0011] The temperature adjustment of the interior space of the vehicle based on the environmental temperature feature and the personnel feature set of the vehicle; the personnel feature set is determined according to the target action of the target person and the body surface temperature of the target person.

[0012] Further, the personnel feature set of the vehicle is obtained by:

[0013] The personnel reference feature corresponding to each target person is determined based on the target action of each target person and the body surface temperature of each target person.

[0014] The personnel feature set of the vehicle is obtained according to the personnel reference feature corresponding to each target person.

[0015] Further, the personnel reference feature corresponding to each target person is determined based on the target action of each target person and the body surface temperature of each target person, comprising:

[0016] For each target person, the K1 part features in the target action and the body surface temperature of the target person are arranged according to a first preset feature arrangement to obtain a first data structure as the personnel reference feature corresponding to the target person; the first preset feature arrangement order is used to indicate a first arrangement order and / or a second arrangement order, the first arrangement order is used to indicate a sorting rule of the target action and the body surface temperature, and the second arrangement order is used to indicate a sorting rule of different part temperatures of K1 part temperatures in the body surface temperature, and the K1 is an integer greater than 0.

[0017] Further, the method further comprises:

[0018] The second data structure obtained by arranging the personnel reference feature corresponding to each target person according to a second preset feature arrangement is determined as the personnel feature set of the vehicle; and / or,

[0019] The K2 environmental parameters of the area environmental parameters of the target location area where the vehicle is currently located and / or the vehicle interior environmental parameters of the interior space of the vehicle are arranged according to a third preset feature arrangement to obtain a third data structure, and the third data structure is determined as the environmental temperature feature of the vehicle.

[0020] Further, the regional environment parameter comprises at least one of the following environment parameters: a region of the target position region, a regional temperature, a regional wind speed, and a regional humidity.

[0021] Further, the in-vehicle environment parameter comprises at least one of the following environment parameters: an in-vehicle temperature of an internal space of the vehicle, an in-vehicle humidity, a current time parameter, and an in-vehicle air conditioner operation parameter.

[0022] Further, the temperature adjustment of the internal space of the vehicle based on the target action of the target person and the body surface temperature of the target person comprises:

[0023] The target action of the target person, the body surface temperature of the target person, and the environment temperature feature are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the internal space of the vehicle.

[0024] The personnel feature set and the environment temperature feature are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the internal space of the vehicle.

[0025] Further, the inputting of the personnel feature set and the environment temperature feature into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the internal space of the vehicle comprises:

[0026] A first feature vector of the personnel feature set is extracted, and a second feature vector of the environment temperature feature is extracted.

[0027] The first feature vector and the second feature vector are fused to obtain a third feature vector.

[0028] A similarity between the first feature vector and the third feature vector is calculated to obtain an attention weight, the first feature vector is weighted by using the attention weight, a result after the weighting is summed to obtain the adjustment temperature, and the attention weight is used to represent an importance degree of different personnel features to the environment temperature feature.

[0029] Further, the target action is obtained by the following way:

[0030] Performing posture recognition on each target video frame in a series of video frames collected for a target person; and determining a target action of the target person based on the target posture recognized from each target video frame and the collection order of each target video frame in the series of video frames, wherein the target video frames are at least two consecutive video frames in the series of video frames; and / or

[0031] The target person's body surface temperature is obtained in the following manner:

[0032] Acquire a thermal imaging image of the interior space of the vehicle, and analyze the thermal imaging image to obtain the body surface temperature of each target person; and / or

[0033] The target actions include at least one of the following: sleeping, talking, curling up, rubbing hands, putting on clothes, taking off coat, fanning with hands, wiping sweat, sitting normally; and / or, the target persons include one or at least two; and / or, the target persons include the driver and / or passengers of the vehicle.

[0034] In a second aspect, an embodiment of the present invention provides a device for regulating the temperature inside a vehicle, the device comprising:

[0035] A detection module is configured to detect a target action of a target person in the interior space of a vehicle, wherein the target action is an action consisting of at least two gestures performed in sequence.

[0036] The regulating module is configured to regulate the temperature of the interior space of the vehicle based at least on the target action of the target person and the body surface temperature of the target person, wherein the body surface temperature is determined by the temperature of at least one key part of the target person.

[0037] In a third aspect, an embodiment of the present invention provides a vehicle, comprising: a controller, wherein the controller comprises: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, wherein computer instructions are stored in the memory, and the processor executes the above method by executing the computer instructions.

[0038] In a fourth aspect, an embodiment of the present invention provides an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions being stored in the memory, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0039] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method of the first aspect or any corresponding embodiment thereof.

[0040] The application can indirectly understand the subjective feeling of the target person to the temperature in the vehicle by detecting the target action of the target person composed of at least two sequentially executed postures; meanwhile, the body surface temperature is determined based on the temperature of at least one key part, the physiological state difference of the passenger is considered, and real-time perception is realized. The temperature in the vehicle is adjusted by comprehensively considering the target action and the body surface temperature, manual interaction is not required, automatic operation can be realized, the needs of different passengers can be accurately met, and the problems of the prior art are effectively solved. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings required to be used in the specific embodiments or the prior art description will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0042] Figure 1 is a flowchart of a vehicle interior temperature adjustment method according to some embodiments of the present application;

[0043] Figure 2 is a flowchart of a vehicle interior temperature adjustment method according to some embodiments of the present application;

[0044] Figure 3 is a schematic diagram of real-time video data recognition according to some embodiments of the present application;

[0045] Figure 4 is a structural schematic diagram of a temperature prediction model according to some embodiments of the present application;

[0046] Figure 5 is a schematic diagram of vehicle interior temperature prediction according to some embodiments of the present application;

[0047] Figure 6 is a structural block diagram of a vehicle interior temperature adjustment device according to an embodiment of the present application;

[0048] Figure 7 is a hardware structure schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0050] According to the embodiment of the present application, a method for adjusting the temperature in a vehicle interior and a vehicle are provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0051] In the embodiment, a method for adjusting the temperature in a vehicle interior is provided, Figure 1 is a flowchart of the method for adjusting the temperature in a vehicle interior according to the embodiment of the present application, as shown in the figure, the flow includes the following steps: Figure 1

[0052] Step S101, detecting a target action of a target person in an interior space of a vehicle, the target action being an action composed of at least two postures executed in sequence.

[0053] In the embodiment of the present application, posture recognition is performed on each target video frame in the continuous video frames collected for the target person; and based on the target postures recognized from each target video frame and the collection order of each target video frame in the continuous video frames, the target action of the target person is determined, the target video frame being at least two video frames in the continuous video frames. Wherein, the target action includes at least one of the following: sleeping, talking, curling the body, rubbing the hands, wearing clothes, taking off the coat, fanning with hands, wiping sweat, and normal sitting; and / or, the target person includes one or at least two; and / or, the target person includes the driver and / or the passenger of the vehicle.

[0054] It should be noted that the target action refers to a series of sequential behavior actions made by the target person in the interior space of the vehicle, which can reflect the person's perception or comfort condition of the current temperature in the vehicle. These actions are not isolated single postures, but are composed of at least two postures executed in a specific order. For example, when a person feels cold, a continuous action sequence of "holding arms - curling the body - rubbing hands" may occur; when a person feels hot, actions such as "fanning with hands - taking off the coat - wiping sweat" may occur. By recognizing and analyzing these target actions, the subjective feelings and needs of the person for the temperature in the vehicle can be indirectly understood.

[0055] ​As an example, in a vehicle interior, a camera mounted on the roof continuously captures continuous video frames containing a target person inside the vehicle. For each target video frame (i.e. at least two continuous video frames) in the captured continuous video frames, by identifying the posture of the target person in the video frame, such as identifying that in a video frame the person is blowing air with hands on the mouth, and in the next video frame the body is slightly shaking. According to these target postures identified from each target video frame, and the sequence of capturing each target video frame in the continuous video frames, the target action of the target person can be determined, such as the continuous action sequence of "blowing air with hands - body shaking", which reflects that the person may feel cold.

[0056] In step S102, at least based on the target action of the target person and the body surface temperature of the target person determined by the part temperature of at least one key part of the target person, the interior space of the vehicle is temperature-regulated.

[0057] In the embodiments of the present application, a thermal imaging image of the interior space of the vehicle is obtained, and the thermal imaging image is analyzed to obtain the body surface temperature of each target person.

[0058] Specifically, a thermal imager is installed at a suitable position in the vehicle interior to continuously collect thermal imaging information of the person inside the vehicle. Secondly, the thermal imaging information collected by the thermal imager is spatio-temporally aligned and fused with the video information collected by the camera, so that the temperature data in the thermal imaging can be accurately corresponded to the body parts of the person in the video picture.

[0059] Then, the key parts of the target person, such as forehead, hands, neck, etc. are determined, and the part temperatures of these key parts are extracted from the fused information. The plurality of extracted key part temperatures are comprehensively processed, for example, a weighted average method is used, and according to the influence degree of different key parts on the overall body surface temperature, corresponding weights are given to calculate the body surface temperature of the target person.

[0060] Then, in combination with the identified target action and the calculated body surface temperature of the target person, the feeling and demand of the person to the current temperature in the vehicle are analyzed. Finally, according to the analysis result, according to the preset temperature regulation rule, such as when the action of feeling cold is identified and the body surface temperature is low, the temperature in the vehicle is increased; when the action of feeling hot is identified and the body surface temperature is high, the temperature in the vehicle is decreased, the interior space of the vehicle is correspondingly temperature-regulated.

[0061] It should be noted that the body surface temperature of the person is a physical quantity that can reflect the body surface temperature condition of the target person in the vehicle. By selecting at least one representative key part (such as the forehead, hands, neck, etc., the temperature change of which can better reflect the overall thermal state of the human body) of the target person, the part temperature of the key part is obtained by using a thermal imager. The body surface temperature of the person can be used as an important index to determine whether the person is in a comfortable temperature environment and whether the temperature in the vehicle is appropriate.

[0062] As an example, in the vehicle, in order to determine the body surface temperature of the target person (driver and / or passenger), the forehead is selected as the key part. A non-contact infrared thermometer is installed at a suitable position in the vehicle, and when the target person is in the measurement range of the thermometer, the thermometer continuously emits infrared rays and receives the infrared signals reflected by the forehead. Through analysis and processing of the received signals, real-time temperature data of the forehead part can be quickly obtained. Since the forehead temperature is relatively stable and can better reflect the overall body surface temperature condition of the human body, the measured forehead temperature is directly determined as the body surface temperature of the target person.

[0063] As another example, in the vehicle, in order to more accurately determine the body surface temperature of the target person (driver and / or passenger), the forehead and wrist are selected as two key parts. Two non-contact infrared thermometers are installed at corresponding positions in the vehicle, one of which is aimed at the forehead of the target person and the other of which is aimed at the wrist. When the target person enters the measurement area, the two thermometers work simultaneously to obtain the temperature data of the forehead and wrist parts, respectively. Then, the temperatures of the two parts are comprehensively analyzed. Considering that the forehead temperature is relatively small affected by the external environment but may be different due to individual physiological differences, and the wrist temperature can reflect the temperature change caused by the peripheral blood circulation of the human body, the average value of the temperatures of the two parts is determined as the body surface temperature of the target person.

[0064] The present application can indirectly understand the subjective feeling of the person to the temperature in the vehicle by detecting the target action of the target person composed of at least two sequentially executed postures, and determine the body surface temperature based on the part temperature of at least one key part, considering the physiological state difference of the passenger and realizing real-time perception. The temperature in the vehicle is adjusted by comprehensively considering the target action and the body surface temperature, without manual interaction, and can be automatically operated, which can accurately meet the needs of different passengers and effectively solve the problems of the prior art.

[0065] In another embodiment of the present application, a method for adjusting the temperature in a vehicle is provided, Figure 2 is a flowchart of the method for adjusting the temperature in the vehicle according to an embodiment of the present application, as Figure 2 shown, the flow includes the following steps:

[0066] Step S201, detecting a target action of a target person in the internal space of the vehicle, the target action being an action composed of at least two poses executed in sequence.

[0067] In the embodiment of the present application, the camera inside the vehicle continuously captures video information of the internal space of the vehicle. Secondly, each captured frame of video is preprocessed to remove noise and interference in the frame and enhance the clarity and contrast of the frame, so as to more accurately identify the person and the pose. Then, the target person is located in the preprocessed frame, and the main parts of the body such as the head, arms, legs, etc. are marked.

[0068] Then, according to the pre-set pose sequence template of the target action, the body part actions of the target person are tracked and analyzed frame by frame. The position changes and action transitions of the body parts between different frames are observed, and when a series of continuous pose changes are found to conform to a pose sequence template of a target action, it is determined that the target person is performing the target action, which is composed of at least two poses executed in sequence.

[0069] As an example, as shown in Figure 3 The in-vehicle camera is in an automatic continuous capture mode, continuously capturing video frames containing the driver and the passenger in the back row. In specific operation, a sliding window mechanism is used to maintain 10 frames of data for recognition. When the number of captured video frames does not exceed 10 frames, data is normally accumulated; once it exceeds 10 frames, a new frame is added to the sliding window every time a new inference is performed, and the farthest frame is deleted, so that the 10 frames of data in the window are always the latest and continuous.

[0070] After these video frames containing person action information are transmitted to the controller, the controller first uses the ConvNext network to extract features from each frame of video in the sliding window, converting each frame of video into representative feature information. Then, these extracted frame features are input into the ConvGRU network in the order of the video frames in the sliding window. Since the actions of the people inside the vehicle are continuous, the ConvGRU network can well capture the time dependence between the continuous frames and deeply mine the correlation information between the frames due to its advantage in processing sequence data.

[0071] After being processed by the ConvGRU network, the features are input into the FC (full connection layer) for further processing to integrate and refine the feature information. Finally, the softmax layer is used for classification, so as to accurately identify the pose of the driver and the passenger in the back row corresponding to each frame in the continuous video frames.

[0072] After analysis, it is identified that the driver first holds hands in front of the chest, then the body slowly curls up to the seat back, and then rubs hands on arms quickly; the rear passenger first frequently waves hands near the face, then unbuttons the coat, and finally wipes sweat with hands. Then, based on the collection sequence of each frame in the sliding window and the identified posture, the target action sequence of the driver is determined as "hands on arms-body curling-rubbing hands", which reflects that the driver feels cold; the target action sequence of the rear passenger is "hand fan-outer coat-wiping sweat", which indicates that the rear passenger feels hot.

[0073] In step S202, an environmental temperature feature of the vehicle is determined based on a regional environmental parameter of a target location region where the vehicle currently locates, and / or an in-vehicle environmental parameter of an internal space of the vehicle.

[0074] In the embodiments of the present application, first, the target location region information where the vehicle currently locates is obtained by a positioning device equipped on the vehicle. Second, the regional environmental parameter of the target location region, such as outdoor temperature, humidity, and light intensity, is obtained by using an external environmental data platform in communication connection with the vehicle or an environmental sensor carried by the vehicle itself; meanwhile, the in-vehicle environmental parameter of the internal space of the vehicle, such as air temperature, humidity, and ventilation condition in the vehicle, is collected by using various environmental sensors installed in the vehicle.

[0075] Then, the obtained regional environmental parameter and in-vehicle environmental parameter are sorted and analyzed, and the key data related to temperature, such as the high and low of outdoor temperature and the distribution of in-vehicle temperature, are extracted. Finally, the key data related to temperature are comprehensively processed, and the environmental temperature feature accurately reflecting the environmental temperature condition of the vehicle is determined according to certain rules and logic.

[0076] In step S203, the internal space of the vehicle is temperature-regulated based on at least the target action of the target person and the body surface temperature of the target person, the body surface temperature being determined by the part temperature of at least one key part of the target person.

[0077] In the embodiments of the present application, the internal space of the vehicle can be temperature-regulated based on the target action of the target person, the body surface temperature of the target person, and the environmental temperature feature; and / or the internal space of the vehicle can be temperature-regulated based on the personnel feature set of the vehicle and the environmental temperature feature, the personnel feature set being determined according to the target action of the target person and the body surface temperature of the target person.

[0078] The embodiment of the application detects a target action composed of sequential postures, can understand the subjective feeling of the passenger on the temperature, determines the environmental temperature characteristics in combination with the regional and in-vehicle environmental parameters, comprehensively considers the external and in-vehicle environmental factors. Meanwhile, the personnel feature set is determined according to the target action and the body surface temperature. The target action, the body surface temperature, the environmental temperature characteristics or the personnel feature set and the environmental temperature characteristics are comprehensively combined to adjust the temperature, which not only realizes automation, but also fully considers the physiological state and the body surface temperature of the passenger, and can also combine the environmental factors, so as to accurately meet the needs of different passengers and achieve personalized and intelligent temperature adjustment.

[0079] In the embodiment of the application, the personnel feature set of the vehicle is obtained through the following steps A1-A2:

[0080] Step A1, determining the personnel reference features corresponding to each target person based on the target action of each target person and the body surface temperature of each target person.

[0081] Specifically, determining the personnel reference features corresponding to each target person based on the target action of each target person and the body surface temperature of each target person includes: for each target person, taking the K1 part features in the target action and the body surface temperature of the target person as the personnel reference features corresponding to the target person according to the first data structure obtained by arranging the first preset features; the first preset feature arrangement order is used to indicate the first arrangement order and / or the second arrangement order, the first arrangement order is used to indicate the sorting rule of the target action and the body surface temperature, and the second arrangement order indicates the sorting rule of the different part temperatures of the K1 part temperatures in the body surface temperature, and K1 is an integer greater than 0.

[0082] It can be understood that the first arrangement order is the sorting rule of the target action (corresponding to the action composed of multiple postures such as taking off the coat and curling the body mentioned in the foregoing description) and the body surface temperature. In the scene described in the foregoing description, the in-vehicle personnel has a specific target action and a body surface temperature composed of multiple key part temperatures. The first arrangement order is to determine the arrangement mode of the target action related data and the body surface temperature related data in the data structure involved in forming the personnel reference features.

[0083] For example, the data of the target action is arranged first, and then the data of the body surface temperature is arranged; or conversely, the data of the body surface temperature is arranged first, and then the data of the target action is arranged. Such sorting rule helps to integrate different types of information related to the personnel state in a unified and orderly manner into the subsequent data structure, and provides a standardized data organization form for subsequent temperature adjustment and other operations based on these information.

[0084] The second arrangement order is an arrangement rule for K1 different part temperatures in the body surface temperature. It is mentioned above that the body surface temperature is determined by the part temperature of at least one key part of the target person, such as the head temperature, the neck temperature, etc. When considering the K1 part temperatures in the body surface temperature, the second arrangement order is to determine the arrangement order of the K1 different part temperatures in the data structure.

[0085] For example, the part temperatures are arranged in the order from the head to the foot, or arranged according to some rule such as the size of the temperature value. Through such an arrangement rule, the temperature information of different parts in the body surface temperature can be presented in order in the data structure, which facilitates subsequent comprehensive analysis and processing of the temperature information, so as to more accurately reflect the body surface temperature condition of the person.

[0086] In general, when determining the person reference feature based on the target action and the body surface temperature of each target person, for each target person, K1 part features in the target action and the body surface temperature are arranged according to the first preset feature containing the above two arrangement orders, to form a first data structure in the form of a sequence or a one-dimensional vector as the person reference feature.

[0087] Step A2, obtaining the person feature set of the vehicle according to the person reference feature corresponding to each target person.

[0088] According to the person reference feature corresponding to each target person, which is formed by the target action and the body surface temperature related feature according to the specific arrangement rule, the person reference features of all target persons are integrated and summarized to obtain the person feature set of the vehicle.

[0089] In the embodiments of the present application, the method further comprises: determining the second data structure obtained by arranging the person reference feature corresponding to each target person according to the second preset feature as the person feature set of the vehicle; and / or, obtaining K2 environment parameters in the area environment parameters of the target location area where the vehicle is currently located and / or the in-vehicle environment parameters of the internal space of the vehicle, and determining the third data structure obtained by arranging the K2 environment parameters according to the third preset feature as the environmental temperature feature of the vehicle. The area environment parameters include at least one of the following environment parameters: the area, the area temperature, the area wind speed, and the area humidity of the target location area; and / or, the in-vehicle environment parameters include at least one of the following environment parameters: the in-vehicle temperature, the in-vehicle humidity, the current time parameter, and the in-vehicle air conditioner running parameter of the internal space of the vehicle.

[0090] It should be noted that the personnel reference features corresponding to each target person formed by the target action and the body surface temperature related features in a specific arrangement rule are arranged according to the second preset feature to form a second data structure which can be a set or a one-dimensional vector. For example, taking a 5-seater vehicle as an example, the maximum number of people in the vehicle is 5, and the total data of 5 people is 5x5, which is pulled into a one-dimensional vector of 1x25. The formula is as follows, where p i represents the ith person. p i = (a, t1, t2, t3, t4), where (i = 1, 2, 3, 4, 5). The second data structure is: X1 = {p i |i = 1, 2, 3, 4, 5}.

[0091] The second data structure is determined as the personnel feature set of the vehicle; at the same time or separately, the K2 environmental parameters of the area environmental parameters of the target location area where the vehicle is currently located and / or the in-vehicle environmental parameters of the internal space of the vehicle are obtained, and arranged according to the third preset feature to form a third data structure, which is determined as the environmental temperature feature of the vehicle.

[0092] For example, the third data structure is X2 = {t in , t out , v out , h out , h in , v ac , t d , t m d ac}. Wherein, t in is the original internal temperature, t out is the external temperature, v out is the external wind speed, h out is the external humidity, h in is the internal humidity, v ac is the internal air conditioning wind speed, t d is the current time, t m is the current month, and d ac is the air conditioning on duration.

[0093] In the embodiments of the present application, the internal space of the vehicle is temperature adjusted based on at least the target action of the target person and the body surface temperature of the target person, including the following two ways:

[0094] ① When the internal space of the vehicle is temperature adjusted based on the target action of the target person, the body surface temperature of the target person and the environmental temperature feature, the target action of the target person, the body surface temperature of the target person and the environmental temperature feature are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the internal space of the vehicle.

[0095] For example, during the driving of the vehicle, the in-vehicle camera continuously captures continuous video frames containing the driver and the front passenger. The controller performs posture recognition on each target video frame (at least two continuous) in the continuous video frames, and finds that the driver first holds his hands in front of his chest, then slowly curls his body, and then continuously rubs his hands, determines that the target action is "holding hands- body curling-rubbing hands", and reflects that he feels cold. The front passenger first uses his hand to quickly fan his face, then unzips his jacket, and then uses his hand to wipe his forehead sweat, determines that the target action is "hand fanning- jacket off- sweat wiping", and indicates that he feels hot. At the same time, two non-contact infrared thermometers in the vehicle obtain the forehead and wrist temperatures of the two people, and take the average value to obtain the body surface temperature. In addition, the vehicle obtains the regional temperature, wind speed of the target location area currently located, and the in-vehicle temperature, humidity and current time parameters of the in-vehicle space, arranges these regional environmental parameters and in-vehicle environmental parameters according to a preset to form an environmental temperature feature. Finally, the target action, body surface temperature and environmental temperature feature of the target person are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature suitable for the internal space of the vehicle.

[0096] ②When adjusting the temperature of the internal space of the vehicle based on the personnel feature set and the environmental temperature feature of the vehicle, the personnel feature set and the environmental temperature feature are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the internal space of the vehicle.

[0097] For example, during the driving of the vehicle, the in-vehicle camera continuously captures continuous video frames containing the driver and the front passenger. The controller performs posture recognition on each target video frame (at least two continuous) in the continuous video frames, and finds that the driver first holds his hands in front of his chest, then slowly curls his body, and then continuously rubs his hands, determines that the target action is "holding hands- body curling-rubbing hands", and reflects that he feels cold. The front passenger first uses his hand to quickly fan his face, then unzips his jacket, and then uses his hand to wipe his forehead sweat, determines that the target action is "hand fanning- jacket off- sweat wiping", and indicates that he feels hot. At the same time, two non-contact infrared thermometers in the vehicle obtain the forehead and wrist temperatures of the two people, and take the average value to obtain the body surface temperature. In addition, the vehicle obtains the regional temperature, wind speed of the target location area currently located, and the in-vehicle temperature, humidity and current time parameters of the in-vehicle space, arranges these regional environmental parameters and in-vehicle environmental parameters according to a preset to form an environmental temperature feature. Finally, the target action, body surface temperature and environmental temperature feature of the target person are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature suitable for the internal space of the vehicle.

[0098] Specifically, the personnel feature set and the ambient temperature feature are input into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the interior space of the vehicle, including: extracting a first feature vector of the personnel feature set, and extracting a second feature vector of the ambient temperature feature; fusing the first feature vector and the second feature vector to obtain a third feature vector; calculating the similarity between the first feature vector and the third feature vector to obtain an attention weight, and weighting the first feature vector by using the attention weight, and summing the weighted result to obtain the adjustment temperature, wherein the attention weight is used to represent the importance of different personnel features to the ambient temperature feature.

[0099] As shown in Figure 4 The personnel feature set X1 includes the target actions, body surface temperatures and other related information of the target persons in the vehicle. The X1 is mapped to a feature space by a fully connected layer (FC layer) to obtain a first feature vector. This step aims to extract more representative and distinguishable feature representations from the original personnel feature data.

[0100] The ambient temperature feature X2 covers the regional environment parameters (such as regional temperature, regional wind speed, regional humidity, etc.) of the target location area where the vehicle is currently located, and the in-vehicle environment parameters (such as in-vehicle temperature, in-vehicle humidity, current time parameter, in-vehicle air conditioner running parameter, etc.) of the interior space of the vehicle. Similarly, the X2 is processed by a fully connected layer to extract a second feature vector to represent the ambient temperature related feature information.

[0101] The first feature vector and the second feature vector extracted are fused. This fusion process can be realized by simple splicing or other more complex fusion methods, and the purpose is to integrate the information of the personnel features and the ambient temperature features to form a comprehensive feature representation, i.e., a third feature vector. The fused feature vector can contain both the subjective feeling of the personnel to the temperature and the objective information of the ambient temperature.

[0102] The importance of different personnel features to the ambient temperature feature is determined by calculating the similarity between each element in the first feature vector and the third feature vector. The similarity calculation can use methods such as dot product, cosine similarity, etc. The obtained similarity value is processed (such as normalized) to form an attention weight. The attention weight reflects the relative importance of different personnel features when considering the personnel features and the ambient temperature features comprehensively.

[0103] According to the calculated attention weight, each element in the first feature vector is weighted. The weighted result is summed to finally obtain the adjusted temperature of the interior space of the vehicle. This process through the attention mechanism enables the model to pay more attention to the personnel features that have greater influence on temperature adjustment, so as to more accurately determine the temperature adjustment scheme suitable for the personnel in the vehicle.

[0104] For example, in the process of determining the adjusted temperature of the interior space of the vehicle, it is assumed that the first feature vector contains personnel feature information such as the driver feeling cold (such as the feature corresponding to the "double-arm hugging-body curling-frictioning hands" action) and the passenger feeling hot (such as the feature corresponding to the "hand fanning-removing coat-wiping sweat" action), and the third feature vector fuses the personnel features and the environmental temperature features. By calculating the cosine similarity of each element of the first feature vector and the third feature vector, the importance of different personnel features to the environmental temperature features is determined, such as the similarity of the driver feeling cold feature to the comprehensive feature is high, and the obtained attention weight is large. Then, according to these attention weights, the elements representing different personnel features in the first feature vector are weighted, such as the driver feeling cold feature is given a higher weight, and after weighting, the sum is obtained, and finally an adjusted temperature that accurately adapts to the needs of the personnel in the vehicle is obtained.

[0105] As an example, as shown in Figure 5 , the process of determining the adjusted temperature of the interior space of the vehicle is as follows:

[0106] Step 1: Install a high-definition camera at a suitable position in the vehicle to continuously and uninterruptedly collect video pictures inside the vehicle, covering the main area of personnel activities in the vehicle to ensure that the behavior and body part information of the personnel can be clearly captured. Arrange a thermal imager to have a certain overlap with the camera view angle to perform thermal imaging shooting on the body parts of the personnel in the vehicle to obtain accurate temperature distribution data.

[0107] Collect environmental parameters inside and outside the vehicle by using various sensors (such as temperature sensors, humidity sensors, wind speed sensors, etc.), including but not limited to the temperature, humidity, and wind speed inside the vehicle, the temperature, humidity, and light intensity outside the vehicle, and other information to comprehensively reflect the environmental state of the vehicle.

[0108] Step 2: Input the collected video data into a target detection algorithm (such as the YOLO series algorithm based on deep learning, etc.), and the algorithm analyzes each frame of the video to identify the specific personnel and the corresponding body parts (such as the head, neck, arms, etc.) in the picture.

[0109] For the identified personnel, further input the personnel into a behavior recognition model (such as a model based on a convolutional neural network combined with a recurrent neural network), and analyze the action, posture, and other information of the personnel to determine the specific behavior (such as resting, exercising, and talking, etc.).

[0110] For the thermal imaging data collected by the thermal imager, combined with the body part information identified by the target detection algorithm, the corresponding area of each body part in the thermal imaging image is determined, and then the temperature value of the area is obtained. Combine these temperature values with personnel behavior to generate a feature vector X1.

[0111] Step 3, integrate the parameters collected by other sensors to construct a feature vector X2.

[0112] Step 4, input the constructed feature vectors X1 and X2 into the trained algorithm model, the model analyzes and calculates the input feature vectors, and outputs a predicted optimal adjustment temperature value through a series of operations such as internal weight matrix operation and activation function processing.

[0113] Step 5, send the optimal adjustment temperature value output by the algorithm model to the air conditioning control system or other temperature adjusting equipment of the vehicle. These devices automatically adjust the cooling or heating power, air speed and other parameters according to the received temperature instructions, realize accurate adjustment of the temperature in the vehicle, and make the vehicle environment reach a relatively comfortable temperature state.

[0114] In this embodiment, a vehicle interior temperature adjusting device is also provided, which is used to implement the above embodiments and preferred embodiments, and has been described above. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware is also possible and is contemplated.

[0115] The embodiment provides a vehicle interior temperature adjusting device, as shown in Figure 6 , comprising:

[0116] The detection module 601 is configured to detect a target action of a target person in an internal space of a vehicle, the target action being an action composed of at least two postures executed in sequence;

[0117] The adjusting module 602 is configured to adjust the temperature of the internal space of the vehicle based on at least the target action of the target person and the body surface temperature of the target person, the body surface temperature being determined by the part temperature of at least one key part of the target person.

[0118] In the embodiments of the present application, the device further comprises a determination module configured to determine an environmental temperature feature of the vehicle based on an area environmental parameter of a target location area where the vehicle is currently located, and / or an in-vehicle environmental parameter of the internal space of the vehicle.

[0119] The adjusting module 602 is configured to adjust the temperature of the internal space of the vehicle based on the target action of the target person, the body surface temperature of the target person, and the environmental temperature feature; and / or adjust the temperature of the internal space of the vehicle based on the set of person features of the vehicle and the environmental temperature feature; and the set of person features is determined according to the target action of the target person and the body surface temperature of the target person.

[0120] In the embodiment of the present application, the device further comprises a collecting module configured to determine, based on the target action of each target person and the body surface temperature of each target person, the person reference feature corresponding to each target person; and obtain the set of person features of the vehicle according to the person reference feature corresponding to each target person.

[0121] In the embodiment of the present application, the collecting module is configured to, for each target person, take the first data structure obtained by arranging K1 part features in the target action and the body surface temperature of the target person according to a first preset feature arrangement as the person reference feature corresponding to the target person; the first preset feature arrangement order is used to indicate a first arrangement order and / or a second arrangement order, the first arrangement order is used to indicate a sorting rule of the target action and the body surface temperature, and the second arrangement order is used to indicate a sorting rule of different part temperatures of K1 part temperatures in the body surface temperature, and K1 is an integer greater than 0.

[0122] In the embodiment of the present application, the device further comprises an arrangement module configured to determine, as the set of person features of the vehicle, a second data structure obtained by arranging the person reference feature corresponding to each target person according to a second preset feature arrangement; and / or obtain K2 environmental parameters from the area environmental parameters of the target location area where the vehicle is currently located and / or the in-vehicle environmental parameters of the internal space of the vehicle, and determine, as the environmental temperature feature of the vehicle, a third data structure obtained by arranging the K2 environmental parameters according to a third preset feature arrangement.

[0123] In the embodiment of the present application, the area environmental parameters include at least one of the following environmental parameters: a region of the target location area, a region temperature, a region wind speed, and a region humidity; and / or the in-vehicle environmental parameters include at least one of the following environmental parameters: an in-vehicle temperature of the internal space of the vehicle, an in-vehicle humidity, a current time parameter, and an in-vehicle air conditioner operating parameter.

[0124] In the embodiment of the present application, when the adjusting module 602 is configured to adjust the temperature of the internal space of the vehicle based on the target action of the target person, the body surface temperature of the target person, and the environmental temperature feature, the target action of the target person, the body surface temperature of the target person, and the environmental temperature feature are input into the trained in-vehicle temperature adjustment model to obtain an adjusted temperature of the internal space of the vehicle; or when the adjusting module 602 is configured to adjust the temperature of the internal space of the vehicle based on the set of person features of the vehicle and the environmental temperature feature, the set of person features and the environmental temperature feature are input into the trained in-vehicle temperature adjustment model to obtain an adjusted temperature of the internal space of the vehicle.

[0125] In the embodiment of the present application, the adjusting module 602 is configured to extract a first feature vector of the set of personal features and a second feature vector of the environmental temperature feature, fuse the first feature vector and the second feature vector to obtain a third feature vector, calculate a similarity between the first feature vector and the third feature vector to obtain an attention weight, and perform weighted processing on the first feature vector by using the attention weight, and sum the weighted result to obtain the adjusting temperature, wherein the attention weight is used to represent the importance of different personal features to the environmental temperature feature.

[0126] In the embodiment of the present application, the target action is obtained by: performing posture recognition on each target video frame in the continuous video frames collected for the target person; and determining the target action of the target person based on the target posture recognized from each target video frame and the collection sequence of each target video frame in the continuous video frames, the target video frame being at least two continuous video frames in the continuous video frames; and / or, the body surface temperature of the target person is obtained by: obtaining a thermal imaging image of the internal space of the vehicle, and analyzing the thermal imaging image to obtain the body surface temperature of each target person; and / or, the target action includes at least one of the following: sleeping, talking, curling the body, rubbing the hands, wearing clothes, taking off the coat, fanning with hands, wiping sweat, and normally sitting; and / or, the target person includes one or at least two; and / or, the target person includes the driver and / or the passenger of the vehicle.

[0127] Please refer to Figure 7 , Figure 7 is a structural schematic diagram of an electronic device provided by an optional embodiment of the present application, as shown in Figure 7 , the electronic device includes one or more processors 10, a memory 20, and an interface for connecting various components, including a high-speed interface and a low-speed interface. Various components are communicatively connected to each other by using different buses, and can be installed on a common mainboard or in other ways as needed. The processor can process instructions executed within the electronic device, including instructions stored in the memory or graphical information stored on the memory to display a GUI on an external input / output device, such as a display device coupled to the interface. In some optional embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memories, if necessary. Similarly, multiple electronic devices can be connected, each device providing part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0128] The processor 10 can be a central processing unit, a network processing unit, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a generic array logic, or any combination thereof.

[0129] The memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform the methods illustrated in the above embodiments.

[0130] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system and applications required by at least one function. The data storage area can store data created by the use of the electronic device according to the presentation of a small program landing page, and the like. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely from the processor 10, and these remote memories can be connected to the electronic device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0131] The memory 20 can include a volatile memory, such as a random access memory, and can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid state disk. The memory 20 can further include a combination of the above-mentioned types of memories.

[0132] The electronic device further includes a communication interface 30 for communication of the electronic device with other devices or communication networks.

[0133] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or be implemented as computer codes stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded through a network and stored in a local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer codes, when the software or computer codes are accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0134] Although the embodiments of the present application are described in conjunction with the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A method of regulating the temperature of a vehicle interior, characterized by, The method comprises: detecting a target action of a target person in an internal space of a vehicle, the target action being an action composed of at least two postures executed in sequence; based at least on the target action of the target person and a body surface temperature of the target person, the body surface temperature being determined by a part temperature of at least one key part of the target person, temperature regulation is performed on the internal space of the vehicle.

2. The method of claim 1, wherein, Before the at least based on the target action of the target person and the body surface temperature of the target person, the temperature regulation is performed on the internal space of the vehicle, further comprising: based on the area environment parameter of the target location area where the vehicle is currently located and / or the in-vehicle environment parameter of the internal space of the vehicle, the environmental temperature characteristics of the vehicle are determined; The at least based on the target action of the target person and the body surface temperature of the target person, the temperature regulation is performed on the internal space of the vehicle, comprising: based on the target action of the target person, the body surface temperature of the target person, and the environmental temperature characteristics, temperature regulation is performed on the internal space of the vehicle; and / or, based on the environmental temperature characteristics and the personnel feature set of the vehicle, temperature regulation is performed on the internal space of the vehicle; the personnel feature set is determined according to the target action of the target person and the body surface temperature of the target person.

3. The method of claim 2, wherein, The personnel feature set of the vehicle is obtained by: based on the target action of each target person and the body surface temperature of each target person, the personnel reference characteristics corresponding to each target person are determined; based on the personnel reference characteristics corresponding to each target person, the personnel feature set of the vehicle is obtained.

4. The method of claim 3, wherein, The based on the target action of each target person and the body surface temperature of each target person, the personnel reference characteristics corresponding to each target person are determined, comprising: for each target person, the K1 part characteristics in the target action and the body surface temperature of the target person are arranged according to a first preset characteristic arrangement to obtain a first data structure as the personnel reference characteristics corresponding to the target person; the first preset characteristic arrangement order is used to indicate a first arrangement order and / or a second arrangement order, the first arrangement order is used to indicate a sorting rule of the target action and the body surface temperature, and the second arrangement order indicates a sorting rule of different part temperatures of K1 part temperatures in the body surface temperature, and K1 is an integer greater than 0.

5. The method of claim 3, wherein, The method further comprises: arranging the personnel reference characteristics corresponding to each target person according to a second preset characteristic arrangement to obtain a second data structure, and determining the second data structure as the personnel feature set of the vehicle; and / or, obtaining K2 environment parameters in the area environment parameter of the target location area where the vehicle is currently located and / or the in-vehicle environment parameter of the internal space of the vehicle, and determining a third data structure obtained by arranging K2 environment parameters according to a third preset characteristic arrangement as the environmental temperature characteristics of the vehicle.

6. The method of claim 5, wherein, The area environment parameter comprises at least one of the following environment parameters: the area, the area temperature, the area wind speed, and the area humidity of the target location area; and / or, the in-vehicle environment parameter comprises at least one of the following environment parameters: the in-vehicle temperature, the in-vehicle humidity, the current time parameter, and the in-vehicle air conditioner running parameter of the internal space of the vehicle.

7. The method of claim 2, wherein, The temperature adjustment of the interior space of the vehicle based on at least the target action of the target person and the body surface temperature of the target person comprises: The temperature adjustment of the interior space of the vehicle based on the target action of the target person, the body surface temperature of the target person, and the environmental temperature feature comprises inputting the target action of the target person, the body surface temperature of the target person, and the environmental temperature feature into a trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the interior space of the vehicle. The temperature adjustment of the interior space of the vehicle based on the set of person features of the vehicle and the environmental temperature feature comprises inputting the set of person features and the environmental temperature feature into a trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the interior space of the vehicle.

8. The method of claim 6, wherein, The inputting of the set of person features and the environmental temperature feature into the trained in-vehicle temperature adjustment model to obtain the adjustment temperature of the interior space of the vehicle comprises: extracting a first feature vector of the set of person features and a second feature vector of the environmental temperature feature; fusing the first feature vector and the second feature vector to obtain a third feature vector; calculating the similarity between the first feature vector and the third feature vector to obtain an attention weight, and weighting the first feature vector using the attention weight, and summing the weighted result to obtain the adjustment temperature, wherein the attention weight is used to represent the importance of different person features to the environmental temperature feature.

9. The method of claim 1, wherein, The target action is obtained by: performing posture recognition on each target video frame in the continuous video frames collected for the target person, and determining the target action of the target person based on the target posture recognized from each target video frame and the collection sequence of each target video frame in the continuous video frames, wherein the target video frame is at least two continuous video frames in the continuous video frames; and / or The body surface temperature of the target person is obtained by: obtaining a thermal imaging image of the interior space of the vehicle, and analyzing the thermal imaging image to obtain the body surface temperature of each target person; and / or The target action comprises at least one of the following: sleeping, talking, curling the body, rubbing the hands, wearing clothes, taking off a coat, fanning with hands, wiping sweat, and normally sitting; and / or, the target person comprises one or at least two; and / or, the target person comprises the driver and / or the passenger of the vehicle.

10. A vehicle characterized by comprising: The vehicle comprises a controller, wherein the controller comprises a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the method of any one of claims 1 to 9.

Images