Vehicle environment adjusting method and device, vehicle and storage medium

Abstract

The invention relates to a vehicle environment adjusting method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining an identity label of a target pet in the vehicle in response to a preset pet monitoring instruction, and obtaining a preset state judgment feature baseline of the target pet based on the identity label; acquiring multispectral data of a target pet, and combining the multispectral data and a preset state judgment feature baseline to obtain a current state of the target pet; and in response to the situation that the current state is not the preset normal state, generating a corresponding environment adjustment strategy in combination with the environment data in the vehicle, the current state and the identity identifier, and controlling the vehicle to execute corresponding adjustment actions based on the environment adjustment strategy. Therefore, the technical problems that in the related technology, a single physical sensor is prone to being interfered by factors such as pet hair, clothes and light, monitoring stability is insufficient, differential diagnosis and adjustment for pet requirements are lacked, and the riding comfort of pets is not improved are solved.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a method, apparatus, vehicle, and storage medium for environmental regulation of a vehicle. Background Technology

[0002] In related technologies, when users travel with their pets in a car, they can use a single physical sensor inside the vehicle, such as a temperature sensor or a pressure sensor, to detect the pet's condition, obtain the pet's basic physiological data, and then determine the pet's condition so as to adjust the in-car environment accordingly.

[0003] However, in related technologies, single physical sensors are easily interfered with by factors such as pet hair, clothing, and light, resulting in insufficient monitoring stability. This makes it difficult to meet the needs of full-scene safety monitoring of pets during car rides. Furthermore, there is a lack of dedicated diagnosis and adjustment for pets' needs, making it impossible to make differentiated adjustments based on differences in pet breed, weight, etc. This leads to deviations in the status diagnosis results, and consequently, the adjusted in-car environment is still not conducive to improving the pet's condition, which urgently needs improvement. Summary of the Invention

[0004] This application provides a method, device, vehicle, and storage medium for adjusting the environment of a vehicle, in order to solve the technical problems in the related art where a single physical sensor is easily interfered with by factors such as pet hair, clothing, and light, resulting in insufficient monitoring stability and a lack of differentiated diagnosis and adjustment for pets' needs, which is not conducive to improving the comfort of pets while riding in a vehicle.

[0005] The first aspect of this application provides a method for adjusting the environment of a vehicle, comprising the following steps: in response to a preset pet monitoring command, obtaining an identification identifier of a target pet inside the vehicle, and obtaining a preset state judgment feature baseline of the target pet based on the identification identifier; obtaining multispectral data of the target pet, and combining the multispectral data and the preset state judgment feature baseline to obtain the current state of the target pet; in response to the current state not being a preset normal state, generating a corresponding environmental adjustment strategy by combining the environmental data inside the vehicle, the current state, and the identification identifier, and controlling the vehicle to perform corresponding adjustment actions based on the environmental adjustment strategy.

[0006] Optionally, in one embodiment of this application, the step of combining the multispectral data and the preset state judgment feature baseline to obtain the current state of the target pet includes: extracting the near-infrared light reflectance spectrum signal of the target pet from the multispectral data, and calculating the blood oxygen saturation and metabolic rate of the target pet based on the reflectance spectrum signal; extracting the thermal radiation signal of the target pet from the multispectral data, and generating a body surface temperature distribution map of the target pet based on the thermal radiation signal; extracting visible light image data of the target pet from the multispectral data, and identifying the behavioral data and pupil diameter of the target pet based on the visible light image data; and combining the blood oxygen saturation, the metabolic rate, the body surface temperature distribution map, the behavioral data, the pupil diameter, and the preset state judgment feature baseline to obtain the current state.

[0007] Optionally, in one embodiment of this application, the step of combining the blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and preset state judgment feature baseline to obtain the current state includes: capturing the influence relationship between the blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, and pupil diameter; generating global feature data based on the influence relationship; fusing the blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and the global feature data using a preset self-attention mechanism to obtain fused feature data; and matching the current state based on the fused feature data.

[0008] Optionally, in one embodiment of this application, matching the current state based on the fused feature data includes: calculating a matching score for each preset state based on the fused special data; normalizing the matching score to obtain a confidence level for each preset state; and determining the current state from multiple preset states based on the confidence level.

[0009] Optionally, in one embodiment of this application, the step of generating a corresponding environmental regulation strategy by combining the environmental data inside the vehicle, the current state, and the identity identifier includes: obtaining the historical state-environmental regulation data of the target pet based on the identity identifier; and generating the environmental regulation strategy by combining the historical state-environmental regulation data, the environmental data, and the current state.

[0010] Optionally, in one embodiment of this application, the method further includes: acquiring health diagnostic data of the target pet; and updating the preset state judgment feature baseline using the health diagnostic data.

[0011] A second aspect of this application provides an environmental regulation device for a vehicle, comprising: a first acquisition module, configured to acquire an identification identifier of a target pet inside the vehicle in response to a preset pet monitoring command, and to acquire a preset state judgment feature baseline of the target pet based on the identification identifier; a second acquisition module, configured to acquire multispectral data of the target pet, and to obtain the current state of the target pet by combining the multispectral data and the preset state judgment feature baseline; and an adjustment module, configured to generate a corresponding environmental regulation strategy in response to the current state not being a preset normal state, by combining the environmental data inside the vehicle, the current state, and the identification identifier, and to control the vehicle to perform corresponding adjustment actions based on the environmental regulation strategy.

[0012] Optionally, in one embodiment of this application, the second acquisition module includes: a first extraction unit, configured to extract the near-infrared reflectance spectrum signal of the target pet from the multispectral data, and calculate the blood oxygen saturation and metabolic rate of the target pet based on the reflectance spectrum signal; a second extraction unit, configured to extract the thermal radiation signal of the target pet from the multispectral data, and generate a body surface temperature distribution map of the target pet based on the thermal radiation signal; a third extraction unit, configured to extract visible light image data of the target pet from the multispectral data, and identify the behavioral data and pupil diameter of the target pet based on the visible light image data; and a first acquisition unit, configured to combine the blood oxygen saturation, the metabolic rate, the body surface temperature distribution map, the behavioral data, the pupil diameter, and the preset state judgment feature baseline to obtain the current state.

[0013] Optionally, in one embodiment of this application, the first acquisition unit includes: a capture subunit, configured to capture the influence relationship between the blood oxygen saturation, the metabolic rate, the body surface temperature distribution map, the behavioral data, and the pupil diameter, and generate global feature data based on the influence relationship; a fusion subunit, configured to fuse the blood oxygen saturation, the metabolic rate, the body surface temperature distribution map, the behavioral data, the pupil diameter, and the global feature data using a preset self-attention mechanism to obtain fused feature data; and a matching subunit, configured to match the current state based on the fused feature data.

[0014] Optionally, in one embodiment of this application, the matching subunit includes: a calculation subunit, used to calculate a matching score for each preset state based on the fused special data; a processing subunit, used to normalize the matching score to obtain a confidence level for each preset state; and a determination subunit, used to determine the current state from multiple preset states based on the confidence level.

[0015] Optionally, in one embodiment of this application, the adjustment module includes: a second acquisition unit, configured to acquire historical state-environment adjustment data of the target pet based on the identity identifier; and a generation unit, configured to combine the historical state-environment adjustment data, the environmental data, and the current state to generate the environmental adjustment strategy.

[0016] Optionally, in one embodiment of this application, it further includes: a third acquisition module, used to acquire the health diagnostic data of the target pet; and an update module, used to update the preset state judgment feature baseline using the health diagnostic data.

[0017] A third aspect of this application provides a vehicle, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the environmental regulation method for the vehicle as described in the above embodiments.

[0018] A fourth aspect of this application provides a computer-readable storage medium storing computer instructions for causing the computer to perform the vehicle environmental conditioning method as described in the above embodiments.

[0019] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the above-described vehicle environmental regulation method.

[0020] This application embodiment can respond to preset pet monitoring commands, identify pets inside the vehicle, and obtain corresponding files through the target pet's identification. It then uses these files to determine a baseline of characteristics for diagnosing the target pet's condition, enabling differentiated adjustments to diagnostic benchmarks for pets of different breeds, weights, and ages. Furthermore, it collects multispectral data of the target pet to avoid environmental factors affecting the diagnostic results. In cases of abnormal pet condition, it adaptively generates corresponding environmental adjustment strategies based on the pet's characteristics and current state, thereby adjusting the in-vehicle environment to improve the pet's comfort and prevent the pet's abnormal condition from affecting the user's normal driving. This solves the technical problems in related technologies where single physical sensors are easily interfered with by factors such as pet hair, clothing, and lighting, resulting in insufficient monitoring stability and a lack of differentiated diagnosis and adjustment based on pet needs, which is detrimental to improving pet riding comfort.

[0021] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0022] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0023] Figure 1 This is a flowchart of a vehicle environmental conditioning method according to an embodiment of this application; Figure 2 This is a schematic diagram of a multimodal data acquisition process according to an embodiment of this application; Figure 3 This is a schematic diagram illustrating the principle of a CNN-Transformer hybrid deep learning model according to an embodiment of this application; Figure 4 This is a schematic diagram illustrating the principle of pet detection according to an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a vehicle environmental regulation device according to an embodiment of this application; Figure 6 This is a structural schematic diagram of a vehicle provided according to an embodiment of this application. Detailed Implementation

[0024] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0025] The following description, with reference to the accompanying drawings, outlines a vehicle environment adjustment method, apparatus, vehicle, and storage medium according to embodiments of this application. Addressing the technical problems mentioned in the background section, where single physical sensors are easily interfered with by factors such as pet hair, clothing, and light, resulting in insufficient monitoring stability and a lack of differentiated diagnosis and adjustment tailored to pet needs, thus hindering improvements in pet comfort during vehicle use, this application provides a vehicle environment adjustment method. In this method, in response to a preset pet monitoring command, the method identifies the pet inside the vehicle and obtains its corresponding profile through the pet's identification identifier. The profile then determines a baseline of status judgment features for diagnosing the pet's condition, enabling differentiated adjustments to diagnostic benchmarks for pets of different breeds, weights, and ages. Furthermore, the method collects multispectral data of the pet to avoid environmental factors affecting the status diagnosis results. In cases of abnormal pet condition, the method adaptively generates corresponding environmental adjustment strategies based on the pet's characteristics and current state, thereby adjusting the in-vehicle environment, improving pet comfort, and preventing pet abnormalities from affecting the user's normal driving. This solves the technical problems in related technologies, such as the susceptibility of single physical sensors to interference from pet hair, clothing, light, etc., insufficient monitoring stability, lack of differentiated diagnosis and adjustment for pets' needs, and the inability to improve pets' riding comfort.

[0026] Specifically, Figure 1 This is a schematic flowchart of a vehicle environmental conditioning method provided in an embodiment of this application.

[0027] like Figure 1 As shown, the environmental conditioning method for this vehicle includes the following steps: In step S101, in response to a preset pet monitoring instruction, the identification of the target pet inside the vehicle is obtained, and a preset status judgment feature baseline of the target pet is obtained based on the identification.

[0028] In practice, the pet monitoring command can be triggered by the user activating the pet safety mode, or by recognizing any pet feature through image recognition. The pet features can be extracted and categorized from a large number of different pet species, allowing for further identification of the pet's type and other information after feature recognition.

[0029] The vehicle can respond to pet monitoring commands, identify the pet inside the vehicle through image data, and determine the pet's identity.

[0030] To identify the pet, users can input basic pet information via vehicle or mobile terminals, including breed (e.g., Golden Retriever, British Shorthair), size (small / medium / large), and age. Based on the entered pet breed, the system loads the breed's status assessment baseline, such as spectral baseline data, from the data storage unit. This status assessment baseline can be obtained based on statistical analysis of a large amount of experimental data from pets of the same breed, or it can be continuously updated based on the pet's historical data. For example, the normal blood oxygen saturation threshold range for large dogs (Golden Retrievers) is 95%-98%, the pupil diameter threshold range is 3-5mm, and the body surface temperature distribution threshold range is 38.0℃-39.2℃; for small cats (British Shorthairs), the normal blood oxygen saturation threshold range is 94%-97%, the pupil diameter threshold range is 2-4mm, and the body surface temperature distribution threshold range is 38.5℃-39.5℃.

[0031] There are several ways to verify the identity of a pet. For example, users can pre-set the identification feature to be the collar worn by the pet, or they can take multiple photos of the pet in advance and enter them into the vehicle so that the vehicle can extract the pet's features for matching.

[0032] In addition, if the pet in the car is not identified, the user can be prompted to register the pet's identity, or a common identification standard and common baseline data can be used.

[0033] In step S102, multispectral data of the target pet is acquired, and the current state of the target pet is obtained by combining the multispectral data with the preset state judgment feature baseline.

[0034] Furthermore, embodiments of this application can utilize multiple sensors installed in the vehicle, such as near-infrared spectral sensors, thermal infrared spectral sensors, and visible light spectral sensors, to acquire the pet's vital signs data, i.e., multispectral data, and then combine this data with the pet's corresponding state judgment feature baseline to make a comprehensive judgment on the current state.

[0035] Optionally, in one embodiment of this application, the current state of the target pet is obtained by combining multispectral data and a preset state judgment feature baseline, including: extracting the near-infrared light reflectance spectrum signal of the target pet from the multispectral data, and calculating the blood oxygen saturation and metabolic rate of the target pet based on the reflectance spectrum signal; extracting the thermal radiation signal of the target pet from the multispectral data, and generating a body surface temperature distribution map of the target pet based on the thermal radiation signal; extracting visible light image data of the target pet from the multispectral data, and identifying the behavioral data and pupil diameter of the target pet based on the visible light image data; and obtaining the current state by combining blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and a preset state judgment feature baseline.

[0036] For example, in this embodiment of the application, three sets of multispectral sensing modules can be deployed in the pet cabin of a vehicle, respectively installed above the back of the front seats, above the rear seats, and on top of the pet area in the trunk, ensuring coverage of all areas where the pet is active while riding in the vehicle. The overlap rate of the detection range of adjacent modules is ≥30%, avoiding blind spots in monitoring. Figure 2 As shown, each group of multispectral sensing modules integrates three types of sensors: The near-infrared spectral sensor emits near-infrared light to penetrate pet fur (maximum penetration thickness 8cm), receives the reflected spectral signal, and calculates the pet's blood oxygen saturation (accuracy ±2%) and metabolic rate (error ≤5%) through spectral analysis algorithms to determine the pet's fatigue and stress level. The thermal infrared spectroscopy sensor is used to capture the thermal radiation signal of the pet's body surface, generate a body surface temperature distribution map (resolution 0.1℃), accurately locate local overheated areas (such as the abdomen and paws), and the temperature detection range is 35℃-42℃, covering the normal body temperature and abnormal high temperature range of pets. The visible light spectral sensor is used to capture pet images at a frame rate of ≥15fps. It extracts behavioral data such as facial expressions (e.g., dilated pupils, drooping ears, and taut whiskers) and limb movements (e.g., wriggling, lying down, and standing) through image recognition. The pupil diameter recognition accuracy is ±0.1mm.

[0037] By acquiring blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, and pupil diameter, this application embodiment can combine the pet's targeted state judgment feature baseline to make a state judgment.

[0038] The infrared thermal imaging sensor can be replaced with a millimeter-wave radar sensor to acquire physiological data by detecting the pet's micro-movements (breathing, limb movements), maintaining stable monitoring even under poor lighting conditions. The ordinary visual sensor can be replaced with an infrared night vision camera to improve the accuracy of pet position and posture recognition at night or in low-light environments; alternatively, a TOF camera can be used to acquire the pet's three-dimensional spatial information, enhancing recognition capabilities in occluded scenarios.

[0039] Optionally, in one embodiment of this application, the current state is obtained by combining blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and a preset state judgment feature baseline. This includes: capturing the influence relationship between blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, and pupil diameter; generating global feature data based on the influence relationship; fusing blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and global feature data using a preset self-attention mechanism to obtain fused feature data; and matching the current state based on the fused feature data.

[0040] To ensure that the acquired data can be used for feature extraction and fusion, the embodiments of this application can first preprocess the data. First, spatiotemporal synchronization alignment is performed, and the sampling time difference of the three types of spectral sensors and environmental sensors is corrected based on a unified timestamp, with the error controlled within 1ms. Second, the median filtering method is used to remove abnormal noise data (such as sudden interference in spectral signals and light spot occlusion in images).

[0041] Furthermore, embodiments of this application may employ a CNN-Transformer hybrid deep learning model, wherein the training method of the CNN-Transformer hybrid deep learning model can be as follows: Figure 3 As shown, iteration continues until convergence conditions are met, such as stopping when the model accuracy is greater than or equal to 98.5%. This model includes a feature extraction layer, a feature fusion layer, and a classification output layer. The feature extraction layer can extract local features of multispectral data (such as spectral peaks and temperature distribution hotspots) and local features of behavioral data (such as pupil region features and limb posture features) through a CNN network (using a ResNet50 structure); and capture global features and the relationships between features (such as the linkage between oxygen saturation and limb movements, and the relationship between body surface temperature and ambient temperature) through the self-attention mechanism of the Transformer network.

[0042] The feature fusion layer can use a weighted summation method to fuse local features extracted by CNN and global features extracted by Transformer. The weights are dynamically assigned according to the importance of the features. The higher the weight, the greater the influence of the feature on the final output (e.g., under stress, the oxygen saturation feature has a weight of 0.4 and the pupil diameter feature has a weight of 0.3, which means that the oxygen saturation has a greater influence).

[0043] Based on the fused feature data output by the feature fusion layer, the current state of the target pet can be obtained in this embodiment of the application.

[0044] Optionally, in one embodiment of this application, matching the current state based on fused feature data includes: calculating a matching score for each preset state based on fused special data; normalizing the matching scores to obtain the confidence level of each preset state; and determining the current state from multiple preset states based on the confidence level.

[0045] Furthermore, in the classification output layer of the CNN-Transformer hybrid deep learning model, the current state of the pet and its state confidence can be output through a fully connected layer. The states include normal, mild stress, moderate stress, severe stress, mild fatigue, moderate fatigue, heatstroke precursors, and health abnormalities.

[0046] Specifically, the embodiments of this application can perform a comprehensive analysis of the fused integrated feature data, and calculate an original matching score for each preset pet state (normal, mild stress, etc.) to determine the degree of matching between the target heavy feature data and each state.

[0047] Subsequently, in this embodiment of the application, the original matching scores can be uniformly processed, and the original scores of all states can be converted into values ​​between 0 and 1, thereby obtaining the 0-1 value corresponding to each state, which is the confidence level of the model in determining that the pet belongs to that state, and the highest confidence level is selected as the final judgment result of the target pet's current state.

[0048] In addition to the CNN+LSTM hybrid model, the embodiments of this application can also use support vector machines (SVM) to classify multimodal feature vectors, which is suitable for vehicle hardware platforms with limited computing power; for high-end models, the Transformer model can be used to improve the feature capture capability of long time series data; the weighted fusion algorithm can be replaced with the evidence theory fusion algorithm to enhance the decision reliability when there is conflict between multiple sources of data; attention mechanism fusion algorithms can also be used to automatically focus on key features (such as abnormal temperature areas on a pet's body surface, high concentration of harmful gases); in ambiguous risk scenarios, a Bayesian inference model can be used to dynamically update the safety risk level through probability distribution, thereby improving the flexibility of judgment.

[0049] In step S103, in response to the current state not being a preset normal state, a corresponding environmental adjustment strategy is generated by combining the vehicle's internal environmental data, current state, and identity identifier, so as to control the vehicle to perform corresponding adjustment actions based on the environmental adjustment strategy.

[0050] Furthermore, in this embodiment of the application, after calculating the current state, it can determine whether the target pet is normal. If the target pet is not in a normal state, the environment inside the vehicle needs to be adjusted to ensure the pet's health.

[0051] This application embodiment can determine the vehicle's environmental adjustment strategy based on environmental data such as temperature, humidity, and noise inside the vehicle, combined with the target pet's current state and attributes and baselines obtained from its identification. For example, if the target pet is in a state of moderate fatigue, the seat space can be adjusted to provide ample resting space, low-frequency soothing music can be played, and the temperature can be adjusted to the target pet's baseline comfort range.

[0052] Among them, the temperature and humidity sensor used to collect environmental data can be replaced with a high-precision digital temperature and humidity module, combined with a dew point sensor, to more accurately determine whether there is a risk of condensation in the cabin and prevent pets from getting damp.

[0053] Optionally, in one embodiment of this application, a corresponding environmental regulation strategy is generated by combining the vehicle's internal environmental data, current state, and identity identifier, including: obtaining the target pet's historical state-environmental regulation data based on the identity identifier; and generating an environmental regulation strategy by combining the historical state-environmental regulation data, environmental data, and current state.

[0054] Before generating an environmental regulation strategy, this embodiment of the application can also obtain the stored historical state-environment regulation data based on the target pet's identity identifier, that is, obtain the environmental regulation strategy corresponding to the target pet in the corresponding historical state, and then generate the corresponding environmental regulation strategy by combining the historical state-environment regulation data.

[0055] Optionally, in one embodiment of this application, the method further includes: acquiring health diagnostic data of the target pet; and updating a preset state judgment feature baseline using the health diagnostic data.

[0056] It is understandable that although different types and breeds of pets have different environmental needs, different health conditions of pets will also affect their environmental needs.

[0057] Therefore, embodiments of this application can obtain pet health diagnostic data, such as physical examination and health diagnostic data of template pets scanned regularly by users, in order to optimize the baseline of status judgment features.

[0058] This application embodiment can periodically remind users to update their pet's health diagnostic data. For example, it can infer that the pet may be sick based on the user's driving destination (such as a pet hospital), and then remind the user to upload the health diagnostic data.

[0059] In summary, the embodiments of this application can use multimodal spectral sensors (infrared thermal imaging sensors, near-infrared spectral sensors), environmental sensors (temperature and humidity sensors, air quality sensors), and visual sensors to collaboratively collect pet physiological data (body surface temperature distribution, respiratory rate, metabolic state) and cabin environmental data (temperature, humidity, CO2 concentration, harmful gas content).

[0060] This application embodiment can extract features from the data collected by each sensor, obtain the pet's physiological feature vector and the cabin environment feature vector respectively, and use a multimodal fusion algorithm to generate comprehensive feature data; perform a comprehensive analysis on the fused comprehensive feature data, and calculate an original matching score for each preset pet state (normal, mild stress, etc.) to determine the degree of matching between the target heavy feature data and each state.

[0061] This application's embodiments can directly classify a pet's condition into corresponding pet safety levels (normal, warning, dangerous) and cabin environment levels. For example, a normal condition corresponds to the normal level, mild to moderate fatigue to the warning level, etc. The cabin environment level is then determined based on the degree of matching between the cabin environment and the pet's condition. Furthermore, by combining basic information such as pet breed, weight, and health status with historical monitoring data, the initial identification results are dynamically corrected to optimize the accuracy of safety level determination.

[0062] Based on the final safety level, the corresponding preset control strategy is triggered, and cabin control commands are output (adjusting the air conditioning temperature, turning on the fresh air system, slightly lowering the windows, activating the emergency alarm, etc.) to achieve active safety protection; after the control commands are executed, the pet's status and environmental change data are collected in real time, and the model parameters are optimized to form a closed-loop learning.

[0063] In addition to fixed-level corresponding strategies, the system can automatically learn user preference strategies based on historical adjustment data. When a risk level is triggered, it will prioritize common user operations (e.g., if the owner habitually sets the air conditioning temperature to 26℃ during a warning, the system will automatically adapt to this parameter). Besides slightly lowering windows and adjusting the air conditioning, a cabin fragrance system can be introduced to release calming scents (such as lavender) to alleviate pet stress. For pet-specific transport vehicles, ventilation control for pet cages and heating / cooling pad adjustment functions can be added. When an alarm is needed, in addition to sending a message to the user, the system can trigger a short horn blast (1 second each time, with a 5-second interval) and flashing lights to alert nearby people to the safety of the pet in the cabin. It can also link with third-party rescue platforms to automatically send a rescue request if the danger level persists for 10 minutes.

[0064] like Figure 5 As shown, the case of monitoring and intervention for driving fatigue in large dogs (Golden Retrievers) is an example.

[0065] This application embodiment may include the following steps: Step S1: The user enters pet information (breed: Golden Retriever, size: large, age: 3 years old). The system loads the Golden Retriever's spectral characteristic baseline: blood oxygen saturation 95%-98%, pupil diameter 3-5mm, body surface temperature 38.0℃-39.2℃. Step S2: During vehicle operation, the multispectral sensing module collects data every 10 seconds to obtain: near-infrared spectral data (oxygen saturation 92%), thermal infrared spectral data (body surface temperature 38.8℃), visible light data (pupil diameter 6mm, drooping ears), and environmental data (cabin temperature 28℃, noise 55 decibels). Step S3, Data Preprocessing: After spatiotemporal synchronization and alignment, noisy data is removed and converted into standardized feature vectors and quantization parameters; Step S4: Input the CNN-Transformer hybrid model and output a moderate fatigue state with a confidence level of 0.93. Step S5: Based on the moderate fatigue state and combined with cabin environment data, determine the safety level as the warning level, initiate Level 1 intervention, adjust the seat to a 15° reclining angle, play 40Hz low-frequency soothing music, and cool the air conditioner to 25°C with the fan speed on the low setting. Step S6: Data collected after 5 minutes: oxygen saturation 96%, pupil diameter 4mm, body surface temperature 38.3℃, condition returned to normal, intervention parameters were recorded, and the model's adaptation weights for the Golden Retriever were updated.

[0066] For example, the monitoring and intervention of early signs of heatstroke in small cats (British Shorthairs) while parking may include the following steps in the embodiments of this application: Step S1: The user enters the pet's information (breed: British Shorthair, size: small, age: 2 years). The system loads the spectral characteristic baseline of the British Shorthair: blood oxygen saturation 94%-97%, pupil diameter 2-4mm, body surface temperature 38.5℃-39.5℃. Step S2: After triggering the pet monitoring command, the multispectral sensing module collects data every 10 seconds to obtain: near-infrared spectral data (oxygen saturation 91%), thermal infrared spectral data (body surface temperature 39.8℃, abdominal local temperature 40.2℃), visible light data (pupil diameter 4.5mm, whiskers taut), and environmental data (cabin temperature 32℃, noise 40 decibels). Step S3: After spatiotemporal synchronization and alignment, noisy data is removed and converted into standardized feature vectors and quantization parameters; Step S4: Input the CNN-Transformer hybrid model and output the heatstroke precursor state with a confidence level of 0.95.

[0067] Step S5: In this embodiment of the application, the cabin environment level can be determined based on the degree of matching between the cabin environment and the pet's condition, such as whether the ambient temperature is too high. Combined with the signs of heatstroke, the cabin environment level can be determined to be low. Then, by combining basic information such as the pet's breed, weight, and health status with historical monitoring data, the preliminary identification result is dynamically corrected to obtain the warning danger level, and corresponding interventions are implemented.

[0068] In summary, the embodiments of this application can acquire core physiological indicators such as blood oxygen saturation and metabolic rate of pets through multispectral sensing fusion technology. Compared with traditional technologies that only monitor body temperature and respiration, it can detect abnormal states such as pet fatigue, stress, and heatstroke precursors earlier. By adopting a CNN-Transformer hybrid deep learning model, the local features and global correlations of multimodal data are fused to effectively distinguish various pet states and avoid misaligned interventions. Multispectral sensing technology is not affected by pet fur, clothing, or lighting conditions, and still maintains high stability in complex scenarios such as nighttime, low light, and when pet fur is dense.

[0069] The vehicle environment adjustment method proposed in this application can respond to a preset pet monitoring command, identify the pet in the vehicle, obtain the corresponding file through the target pet's identification, and then determine the status judgment feature baseline for diagnosing the target pet's status through the file. This enables differentiated adjustment of diagnostic benchmarks for pets of different breeds, weights, and ages. Multispectral data of the target pet is collected to avoid the influence of environmental factors on the status diagnosis results. In the event of an abnormal state of the target pet, a corresponding environmental adjustment strategy is adaptively generated based on the characteristics and current state of the target pet, thereby adjusting the in-vehicle environment, improving the pet's riding comfort, and preventing the pet's abnormal state from affecting the user's normal driving. This solves the technical problems in related technologies where single physical sensors are easily interfered with by factors such as pet hair, clothing, and light, resulting in insufficient monitoring stability and a lack of differentiated diagnosis and adjustment based on pet needs, which is detrimental to improving pet riding comfort.

[0070] Next, the environmental conditioning device for a vehicle according to an embodiment of this application is described with reference to the accompanying drawings.

[0071] Figure 5 This is a block diagram of the environmental control device for a vehicle according to an embodiment of this application.

[0072] like Figure 5 As shown, the environmental adjustment device 10 of the vehicle includes: a first acquisition module 100, a second acquisition module 200, and an adjustment module 300.

[0073] Specifically, the first acquisition module 100 is used to acquire the identity identifier of the target pet inside the vehicle in response to a preset pet monitoring instruction, so as to acquire the preset status judgment feature baseline of the target pet based on the identity identifier.

[0074] The second acquisition module 200 is used to acquire the multispectral data of the target pet, and combine the multispectral data with the preset state judgment feature baseline to obtain the current state of the target pet.

[0075] The adjustment module 300 is used to respond to situations where the current state is not a preset normal state, and to generate a corresponding environmental adjustment strategy by combining the vehicle's internal environmental data, current state, and identity identifier, so as to control the vehicle to perform corresponding adjustment actions based on the environmental adjustment strategy.

[0076] Optionally, in one embodiment of this application, the second acquisition module 200 includes: a first extraction unit, a second extraction unit, a third extraction unit, and a first acquisition unit.

[0077] The first extraction unit is used to extract the near-infrared light reflection spectrum signal of the target pet from the multispectral data, and to calculate the blood oxygen saturation and metabolic rate of the target pet based on the reflection spectrum signal.

[0078] The second extraction unit is used to extract the thermal radiation signal of the target pet from the multispectral data and generate a body surface temperature distribution map of the target pet based on the thermal radiation signal.

[0079] The third extraction unit is used to extract visible light image data of the target pet from multispectral data, and to identify the target pet's behavioral data and pupil diameter based on the visible light image data.

[0080] The first acquisition unit is used to combine blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter and preset state judgment feature baseline to obtain the current state.

[0081] Optionally, in one embodiment of this application, the first acquisition unit includes: a capture subunit, a fusion subunit, and a matching subunit.

[0082] The capture subunit is used to capture the influence relationship between blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data and pupil diameter, and generate global feature data based on the influence relationship.

[0083] The fusion subunit is used to fuse blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter and global feature data using a preset self-attention mechanism to obtain fused feature data.

[0084] The matching subunit is used to match the current state based on the fused feature data.

[0085] Optionally, in one embodiment of this application, the matching subunit includes: a calculation subunit, a processing subunit, and a determination subunit.

[0086] The calculation sub-component is used to calculate the matching score for each preset state based on the fused special data.

[0087] The processing component is used to normalize the matching score to obtain the confidence level of each preset state.

[0088] The determination sub-component is used to determine the current state from multiple preset states based on confidence level.

[0089] Optionally, in one embodiment of this application, the adjustment module 300 includes a second acquisition unit and a generation unit.

[0090] The second acquisition unit is used to acquire the target pet's historical status-environmental regulation data based on its identity identifier.

[0091] The generation unit is used to combine historical state-environmental regulation data, environmental data, and the current state to generate an environmental regulation strategy.

[0092] Optionally, in one embodiment of this application, the vehicle's environmental conditioning device 10 further includes a third acquisition module and an update module.

[0093] The third acquisition module is used to acquire the target pet's health diagnostic data.

[0094] The update module is used to update the preset status judgment feature baseline using health diagnostic data.

[0095] It should be noted that the foregoing explanation of the vehicle environmental conditioning method embodiment also applies to the vehicle environmental conditioning device of this embodiment, and will not be repeated here.

[0096] The vehicle environment adjustment device proposed in this application can respond to preset pet monitoring commands, identify pets inside the vehicle, and obtain corresponding files through the target pet's identification identifier. It then uses these files to determine a baseline of characteristics for diagnosing the target pet's condition, enabling differentiated adjustments to diagnostic benchmarks for pets of different breeds, weights, and ages. Furthermore, it collects multispectral data of the target pet to avoid environmental factors affecting the diagnostic results. In cases of abnormal pet condition, it adaptively generates corresponding environmental adjustment strategies based on the pet's characteristics and current state, thereby adjusting the in-vehicle environment to improve the pet's comfort and prevent the pet's abnormal condition from affecting the user's normal driving. This solves the technical problems in related technologies where single physical sensors are easily interfered with by factors such as pet hair, clothing, and light, resulting in insufficient monitoring stability and a lack of differentiated diagnosis and adjustment based on pet needs, which is detrimental to improving pet riding comfort.

[0097] Figure 6 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include: The memory 601, the processor 602, and the computer program stored on the memory 601 and capable of running on the processor 602.

[0098] When the processor 602 executes the program, it implements the vehicle environment adjustment method provided in the above embodiments.

[0099] Furthermore, the vehicle also includes: Communication interface 603 is used for communication between memory 601 and processor 602.

[0100] The memory 601 is used to store computer programs that can run on the processor 602.

[0101] The memory 601 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0102] If the memory 601, processor 602, and communication interface 603 are implemented independently, then the communication interface 603, memory 601, and processor 602 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0103] Optionally, in a specific implementation, if the memory 601, processor 602, and communication interface 603 are integrated on a single chip, then the memory 601, processor 602, and communication interface 603 can communicate with each other through an internal interface.

[0104] The processor 602 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0105] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described vehicle environmental regulation method.

[0106] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the vehicle environmental adjustment method provided in this embodiment of the invention.

[0107] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0108] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0109] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0110] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0111] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0112] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0113] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0114] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for adjusting the environment of a vehicle, characterized in that, Includes the following steps: In response to a preset pet monitoring command, the system obtains the identification of the target pet inside the vehicle, and uses the identification to obtain a preset status judgment feature baseline of the target pet. The multispectral data of the target pet is acquired, and the current state of the target pet is obtained by combining the multispectral data with the preset state judgment feature baseline. In response to the current state not being a preset normal state, a corresponding environmental adjustment strategy is generated by combining the environmental data inside the vehicle, the current state, and the identity identifier, so as to control the vehicle to perform corresponding adjustment actions based on the environmental adjustment strategy.

2. The method according to claim 1, characterized in that, The step of combining the multispectral data and the preset state judgment feature baseline to obtain the current state of the target pet includes: The near-infrared reflectance spectrum signal of the target pet is extracted from the multispectral data, and the blood oxygen saturation and metabolic rate of the target pet are calculated based on the reflectance spectrum signal. The thermal radiation signal of the target pet is extracted from the multispectral data, and a body surface temperature distribution map of the target pet is generated based on the thermal radiation signal. Visible light image data of the target pet is extracted from the multispectral data, and behavioral data and pupil diameter of the target pet are identified based on the visible light image data; The current state is obtained by combining the blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and preset state judgment feature baseline.

3. The method according to claim 2, characterized in that, The current state is obtained by combining the blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and preset state judgment feature baseline, including: The influence relationship between blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data and pupil diameter is captured, and global feature data is generated based on the influence relationship. By using a preset self-attention mechanism to fuse the blood oxygen saturation, metabolic rate, body surface temperature distribution map, behavioral data, pupil diameter, and global feature data, fused feature data is obtained. The current state is matched based on the fused feature data.

4. The method according to claim 3, characterized in that, The matching of the current state based on the fused feature data includes: Based on the fused special data, a matching score is calculated for each preset state; The matching scores are normalized to obtain the confidence level of each preset state; Based on the confidence level, the current state is determined from a plurality of preset states.

5. The method according to claim 1, characterized in that, The step of combining the vehicle's internal environmental data, the current state, and the identity identifier to generate a corresponding environmental adjustment strategy includes: Based on the identification, the historical status-environmental regulation data of the target pet are obtained; The environmental regulation strategy is generated by combining the historical state-environmental regulation data, the environmental data, and the current state.

6. The method according to claim 1, characterized in that, Also includes: Obtain the health diagnostic data of the target pet; The preset state judgment feature baseline is updated using the health diagnostic data.

7. An environmental control device for a vehicle, characterized in that, include: The first acquisition module is used to respond to a preset pet monitoring command, acquire the identity identifier of the target pet inside the vehicle, and acquire a preset state judgment feature baseline of the target pet based on the identity identifier; The second acquisition module is used to acquire the multispectral data of the target pet, and combine the multispectral data with the preset state judgment feature baseline to obtain the current state of the target pet; The adjustment module is used to respond to the current state not being a preset normal state, and to generate a corresponding environmental adjustment strategy by combining the environmental data inside the vehicle, the current state, and the identity identifier, so as to control the vehicle to perform corresponding adjustment actions based on the environmental adjustment strategy.

8. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, the processor executing the program to implement the environmental conditioning method for a vehicle as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the environmental conditioning method for a vehicle as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed, it is used to implement the environmental conditioning method for a vehicle as described in any one of claims 1-6.

Images