A method for recommending vehicle sentry modes, electronic devices, and storage media.
By analyzing the environmental risks and remaining battery power of the vehicle when it is parked, and combining multi-dimensional data for fusion reasoning, the activation and deactivation decisions of the Sentinel Mode are optimized. This solves the problem that existing technologies cannot combine the actual parking status of the vehicle with environmental risks, and improves the decision-making accuracy and energy utilization efficiency of the Sentinel Mode.
Patent Information
- Application Number
- CN202611271420.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-20
- Publication Date
- 2026-11-17
AI Technical Summary
The existing sentry mode relies solely on location information for judgment, and cannot combine the actual parking status of the vehicle with the risks of the surrounding environment to make analysis, which makes it difficult to meet users' needs for precise security and energy consumption management.
By acquiring images of the vehicle's surrounding environment and remaining vehicle battery power, the system analyzes environmental risk types and parking posture regularity, and performs fusion reasoning to recommend sentry mode. It considers the vehicle's remaining battery power, environmental risk assessment results, and parking risk assessment results, and combines historical accident statistics and vehicle operation modes to optimize the activation and deactivation decisions of sentry mode.
It achieves fusion reasoning based on the vehicle's remaining battery power and the dual risk assessment results of the environment and parking in complex parking scenarios, which improves the accuracy of sentry mode decision-making and balances the protection needs of vehicle parking with range management.
Smart Images

Figure CN123104571A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of vehicle safety technology, and more particularly to a vehicle sentry mode recommendation method, electronic device, and storage medium. Background Technology
[0002] Sentry Mode is a vehicle parking safety monitoring function equipped in smart cars. Once the vehicle is parked, Sentry Mode uses onboard cameras and various sensors to continuously monitor the surrounding environment. If the system detects an abnormal event, it will automatically start recording the scene, allowing the owner to understand the vehicle's status in a timely manner.
[0003] However, the application of sentry mode typically relies on geofencing technology. Users pre-define locations such as their home or workplace, and sentry mode is activated when a vehicle enters the area and deactivated when it leaves. This singular decision-making mechanism relies solely on location information and cannot analyze the vehicle's actual parking status or the risks of its surrounding environment, making it difficult to meet users' needs for precise security and energy management. Summary of the Invention
[0004] In view of this, this disclosure provides a vehicle sentry mode recommendation method, the method comprising: If the vehicle is detected as parked, obtain images of the vehicle's surrounding environment and the vehicle's remaining battery power; The environmental risk type, the parking pose normality of the vehicle and the parking pose normality of other vehicles in the vehicle surrounding environment image are analyzed to obtain environmental risk assessment results and parking risk assessment results. The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result.
[0005] Optionally, the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result for the sentry mode includes: If any of the preset conditions are met, it is recommended to enable the sentry mode. Otherwise, it is recommended to turn off the Sentinel mode; The preset conditions include: the remaining battery power of the vehicle is not lower than a preset battery power threshold, and the comprehensive risk assessment result jointly indicated by the environmental risk assessment result and the parking risk assessment result is higher than a preset risk threshold.
[0006] Optionally, the method further includes: Obtain the current coordinates of the vehicle and upload them to a big data service. The big data service then queries the historical accident statistics data of the vehicle's current coordinates. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the historical accident statistics.
[0007] Optionally, the method further includes: The ambient temperature around the vehicle, the vehicle model, and the operating mode of the Sentinel mode are obtained to calculate the power consumption of the Sentinel mode per unit time based on the ambient temperature around the vehicle, the vehicle model, and the operating mode. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the power consumption per unit time in the Sentinel Mode.
[0008] Optionally, the method further includes: obtaining the vehicle parking duration; The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, the parking risk assessment results, and the power consumption per unit time in Sentinel Mode to obtain the recommended result for Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, the power consumption per unit time in Sentinel Mode, and the vehicle parking duration.
[0009] Optionally, the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result for the sentry mode includes: The method further includes: merging and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results and reasons for the Sentinel Mode; The vehicle user is shown the recommended results and reasons for the Sentry Mode, so that the vehicle user can select from the recommended results and determine whether the Sentry Mode is enabled based on the selection.
[0010] Optionally, the method further includes: Based on the vehicle user's selection of the recommendation results, the vehicle user's preferences are generated and recorded; The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the vehicle user's preferences.
[0011] This disclosure also provides an electronic device, including a communication interface, a processor, a memory, and a bus, wherein the communication interface, the processor, and the memory are interconnected via the bus; The memory stores machine-readable instructions, and the processor executes the above method by invoking the machine-readable instructions.
[0012] This disclosure also provides a machine-readable storage medium storing machine-readable instructions that, when called and executed by a processor, implement the above-described method.
[0013] Therefore, in the technical solution disclosed herein, by acquiring images of the vehicle's surrounding environment and the vehicle's remaining battery power when the vehicle is parked, the environmental risk type, the vehicle's parking pose normality, and the parking pose normality of other vehicles in the images of the surrounding environment are analyzed to obtain quantitative environmental risk assessment results and parking risk assessment results. Then, based on the vehicle's remaining battery power, environmental risk assessment results, and parking risk assessment results, a fusion reasoning is performed to obtain the sentinel mode recommendation result.
[0014] Through the above methods, the technical solution disclosed herein enables the system to perform fusion reasoning based on the vehicle's remaining battery power and the dual risk assessment results of the environment and parking during the sentry mode decision-making process, thereby balancing the protection needs of vehicle parking and range management, and improving the decision accuracy of activating sentry mode in complex parking scenarios. Attached Figure Description
[0015] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 This is a flowchart illustrating a vehicle sentinel mode recommendation method as an exemplary embodiment; Figure 2 This is a block diagram illustrating a vehicle sentinel mode recommendation architecture as an exemplary embodiment; Figure 3This is a schematic diagram of a sentinel mode recommendation interface shown in an exemplary embodiment; Figure 4 This is an exemplary embodiment illustrating the hardware structure of an electronic device; Figure 5 This is a block diagram illustrating a vehicle sentry mode recommendation device, as shown in an exemplary embodiment. Detailed Implementation
[0017] To enable those skilled in the art to better understand the technical solutions in this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.
[0018] It should be noted that the steps of the corresponding methods are not necessarily performed in the order shown and described in this disclosure in other embodiments. In some other embodiments, the methods may include more or fewer steps than those described in this disclosure. Furthermore, a single step described in this disclosure may be broken down into multiple steps in other embodiments; and multiple steps described in this disclosure may be combined into a single step in other embodiments.
[0019] Sentry Mode is a vehicle parking safety monitoring function equipped in smart cars. Once the vehicle is parked, Sentry Mode uses onboard cameras and various sensors to continuously monitor the surrounding environment. If the system detects an abnormal event, it will automatically start recording the scene, allowing the owner to understand the vehicle's status in a timely manner.
[0020] However, the application of sentry mode typically relies on geofencing technology. Users pre-define locations such as their home or workplace, and sentry mode is activated when a vehicle enters the area and deactivated when it leaves. This singular decision-making mechanism relies solely on location information and cannot analyze the vehicle's actual parking status or the risks of its surrounding environment, making it difficult to meet users' needs for precise security and energy management.
[0021] In view of this, this disclosure aims to propose a technical solution for achieving a refined recommendation sentinel mode based on multi-dimensional perception.
[0022] The technical solution first acquires an image of the vehicle's surrounding environment and the vehicle's remaining battery power if the vehicle is detected to be parked. Further, it analyzes the environmental risk type, the vehicle's parking pose normality, and the parking pose normality of other vehicles in the image to obtain environmental risk assessment results and parking risk assessment results. Finally, it performs fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended result for the sentry mode.
[0023] For example, when the system detects that the vehicle has shifted to park, the electronic parking brake is engaged, and the vehicle speed is zero, it determines that the vehicle has entered a parking state. It then activates the onboard surround-view camera to collect images of the vehicle's surrounding environment from the front, rear, left, and right sides, and reads the vehicle's remaining battery power from the vehicle's battery management system. Next, the system uses a Vision Language Model (VLM) deployed on the vehicle's local computing platform to perform semantic understanding and feature extraction on the surrounding environment images. On one hand, it identifies specific risk categories in the images, such as construction areas, narrow roads, blind spots, areas where pedestrians and vehicles share space, or large vehicles parked adjacent to each other, generating a quantitative environmental risk assessment result. On the other hand, it detects the relative positional relationship between ground markings and the vehicle's outline to determine whether the vehicle is crossing lines or parking in multiple spaces, while also identifying whether adjacent vehicles are parked on lines, occupying multiple parking spaces, or parked crookedly, thus comprehensively deriving a parking risk assessment result. Finally, the system performs a weighted calculation of the environmental risk assessment results and the parking risk assessment results to obtain a comprehensive risk level. If the comprehensive risk level is low and the vehicle's remaining battery power is sufficient to support the preset duration of operation, a recommendation to enable sentry mode is generated; if the comprehensive risk level is high or the vehicle's remaining battery power is below the threshold, a recommendation to disable sentry mode is generated.
[0024] Therefore, in the technical solution disclosed herein, by acquiring images of the vehicle's surrounding environment and the vehicle's remaining battery power when the vehicle is parked, the environmental risk type, the vehicle's parking pose normality, and the parking pose normality of other vehicles in the images of the surrounding environment are analyzed to obtain quantitative environmental risk assessment results and parking risk assessment results. Then, based on the vehicle's remaining battery power, environmental risk assessment results, and parking risk assessment results, a fusion reasoning is performed to obtain the sentinel mode recommendation result.
[0025] Through the above methods, the technical solution disclosed herein enables the system to perform fusion reasoning based on the vehicle's remaining battery power and the dual risk assessment results of the environment and parking during the sentry mode decision-making process, thereby balancing the protection needs of vehicle parking and range management, and improving the decision accuracy of activating sentry mode in complex parking scenarios.
[0026] The present disclosure will now be described through specific embodiments and in conjunction with specific application scenarios.
[0027] Please see Figure 1 , Figure 1 This is a flowchart illustrating an exemplary embodiment of a vehicle sentry mode recommendation method. The method may perform the following steps: Step 102: If the vehicle is detected to be parked, obtain an image of the vehicle's surrounding environment and the vehicle's remaining battery power.
[0028] For example, when the system detects that the vehicle has shifted to parking gear, the electronic parking brake has been engaged, and the vehicle speed is zero, it determines that the vehicle has entered parking mode. It then activates the onboard surround view camera to collect images of the vehicle's surrounding environment covering the front, rear, left, and right views, and reads the current remaining battery power from the vehicle's battery management system.
[0029] Among them, the vehicle-surrounding environment image represents the physical spatial distribution and traffic object status around the vehicle. The vehicle's remaining battery power represents the current energy reserve level of the power system and serves as a constraint indicator for subsequent evaluation of the energy consumption feasibility of the Sentinel Mode operation.
[0030] In practical applications, the system can also use non-visual sensors such as ultrasonic radar and millimeter-wave radar to replace or supplement the camera function for measuring vehicle distances, identifying obstacles, etc.
[0031] In some embodiments, after acquiring multi-view images of the vehicle's surrounding environment, the system can synthesize a bird's-eye view using image stitching technology to provide visual information about the vehicle's surrounding environment without blind spots.
[0032] Step 104: Analyze the environmental risk type, the parking pose normality of the vehicle and other vehicles in the vehicle surrounding environment image to obtain the environmental risk assessment results and parking risk assessment results.
[0033] For example, the system uses VLM deployed on the vehicle's local computing platform to perform semantic understanding and feature extraction on images of the vehicle's surrounding environment: on the one hand, it identifies specific risk categories such as construction areas, narrow roads, blind spots, mixed pedestrian and vehicle traffic areas, or large vehicles parked next to each other in the images, and generates quantitative environmental risk assessment results; on the other hand, it detects the relative positional relationship between ground markings and vehicle outlines, determines whether the vehicle is crossing the lines or crossing the space, and identifies whether neighboring vehicles are parked on the lines, occupying multiple parking spaces, with insufficient parking space width, or parked crookedly, and comprehensively obtains parking risk assessment results.
[0034] Among them, the environmental risk assessment results reflect the potential threat level of the current physical environment to vehicle safety, and the parking risk assessment results reflect the probability level of the accident caused by improper parking posture of the vehicle or other vehicles, resulting in a scratch or collision between the vehicle and other vehicles.
[0035] Please refer to Table 1, which shows examples of risk scenarios identified by VLM through semantic understanding of vehicle surrounding environment images.
[0036] Table 1 In practical applications, the system can also use traditional object detection models, such as the YOLO (You Only LookOnce) series and Faster R-CNN (Faster Region-Convolutional Neural Network), to replace VLM for environmental risk identification. Although the generalization ability and semantic understanding ability of traditional object detection models are slightly lower than those of VLM, their detection accuracy in specific risk scenarios (such as construction areas, obstacles, etc.) can achieve similar results.
[0037] In some embodiments, environmental risk types include not only those related to vehicle traffic conditions, but also vehicle parking slope (which can be obtained by a vehicle tilt sensor). If the slope of the vehicle or other vehicles around it is large, it also indicates that there is a significant safety risk.
[0038] Step 106: Based on the remaining battery power of the vehicle, the environmental risk assessment results, and the parking risk assessment results, perform fusion reasoning to obtain the recommended result of the Sentinel mode.
[0039] For example, the system weights the environmental risk assessment results and the parking risk assessment results to obtain a comprehensive risk level (including safe, low risk, medium risk, high risk, etc.). If the comprehensive risk level is low (such as safe or low risk) and the vehicle's remaining battery power is sufficient to support the preset duration of operation, a recommended result for turning on the sentry mode is generated. If the comprehensive risk level is high (such as medium risk or high risk) or the vehicle's remaining battery power is below the threshold, a recommended result for turning off the sentry mode is generated.
[0040] Among them, fusion reasoning overcomes the limitations of single-dimensional decision-making by balancing external security risks and internal energy supply capacity, and realizes intelligent decision-making that optimizes energy utilization efficiency while ensuring vehicle parking safety.
[0041] In some embodiments, in addition to using a preset algorithm, the fusion reasoning process can also use the vehicle's remaining battery power, environmental risk assessment results, and parking risk assessment results as input features to a pre-trained risk reasoning engine for nonlinear reasoning, so as to improve the decision-making adaptability to complex parking scenarios.
[0042] In some embodiments, when the system uses a VLM deployed on the vehicle's local computing platform to perform semantic understanding and feature extraction on images of the vehicle's surrounding environment, it also assesses whether the vehicle's parking posture affects other vehicles (e.g., the vehicle's parking posture affects other vehicles from leaving the parking space, the vehicle's parking posture occupies other vehicles' parking spaces, the vehicle's parking posture is too close to other parked vehicles, etc.). If the vehicle's parking posture affects other vehicles, the system directly suggests activating sentry mode without referring to the fusion inference results, to avoid disputes caused by a lack of monitoring when risks occur later.
[0043] Regarding the display method of the Sentinel Mode recommendation results, the system can detect whether the user is in the vehicle. If the user is in the vehicle, it will be displayed on the vehicle's central control screen; if the user is not in the vehicle, it will be displayed on the user's mobile phone or other terminal connected to the vehicle.
[0044] In practical applications, risk notifications and sentry mode recommendation results can also be displayed through wearable devices such as smartwatches and smart glasses connected to the vehicle by the user.
[0045] In one embodiment shown, the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result for the sentry mode includes: if any of the preset conditions are met, it is recommended to enable the sentry mode; otherwise, it is recommended to disable the sentry mode; wherein the preset conditions include: the vehicle's remaining battery power is not lower than a preset battery power threshold, and the comprehensive risk assessment result jointly indicated by the environmental risk assessment result and the parking risk assessment result is higher than a preset risk threshold.
[0046] For example, the system first combines the vehicle battery's historical discharge curve data with the battery's internal resistance characteristics at the current ambient temperature to calculate the minimum remaining charge required to ensure a cold start, and then adds a safety margin to determine a preset charge threshold (e.g., 20%). Simultaneously, the system assigns weights to environmental risks and parking risks based on the correlation between risk type and the vehicle's historical accidents (e.g., environmental risk weight 0.4, parking risk weight 0.6). Then, it performs a weighted summation of environmental risks (e.g., construction areas or narrow roads around the vehicle) and parking risks (e.g., vehicles crossing lane lines or excessively narrow spacing) to calculate a comprehensive risk assessment result. If either the vehicle's remaining charge is not lower than the preset charge threshold or the comprehensive risk assessment result is higher than a preset risk threshold (e.g., 60 points), then it recommends enabling the sentry mode; conversely, if neither the vehicle's remaining charge is lower than the preset charge threshold nor the comprehensive risk assessment result is higher than the preset risk threshold (e.g., 60 points) is met, then it recommends disabling the sentry mode.
[0047] The preset battery threshold is used to prevent excessive battery drain in Sentry Mode from preventing the vehicle from starting normally. The comprehensive risk assessment quantifies the potential threat to vehicle safety posed by the current physical environment and the vehicle's parking posture. By jointly judging these two factors, the system prioritizes vehicle safety in high-risk or fully charged scenarios, while proactively avoiding unnecessary power loss in low-risk or low-battery scenarios, achieving a balance between vehicle safety and energy management.
[0048] In some implementations, considering that vehicles consume power quickly at low temperatures, a higher preset power threshold is required. The preset power threshold can be determined in conjunction with the vehicle's ambient temperature. For example, when the current temperature is below 10°C, the preset power threshold is set to 40%, and when the current temperature is not below 10°C, the preset power threshold is set to 20%.
[0049] In one embodiment shown, the method further includes: obtaining the current coordinates of the vehicle and uploading the current coordinates of the vehicle to a big data service, wherein the big data service queries historical accident statistics of the current coordinates of the vehicle; the step of performing fusion reasoning based on the remaining battery power of the vehicle, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result of the sentry mode includes: performing fusion reasoning based on the remaining battery power of the vehicle, the environmental risk assessment result, the parking risk assessment result, and the historical accident statistics to obtain the recommended result of the sentry mode.
[0050] For example, the system collects the vehicle's latitude and longitude coordinates through an onboard positioning system (GPS or BeiDou Navigation Satellite System) and uploads these coordinates to a cloud-based big data service via the onboard communication unit. Upon receiving the vehicle's coordinates, the cloud-based big data service executes coordinate point matching logic to determine the type of point of interest (POI) the vehicle is currently in (e.g., roadside parking space, shopping mall parking space). It then retrieves vehicle safety event records within a preset radius (default 200 meters, configurable by the user) around the POI within a preset time period (e.g., the last 90 days, configurable by the user). This retrieves statistical data on the frequency, type distribution, and peak time periods of events such as vehicle collisions, minor scrapes, and thefts. Finally, the cloud-based big data service calculates a big data risk score based on the retrieved statistical data, generating structured data that includes the POI type, the number of events in the last 90 days, the event type distribution, the big data risk score, and a comprehensive risk rating.
[0051] Subsequently, the system performs fusion inference of multi-source heterogeneous data, and uses the generated structured data, along with the vehicle's remaining battery power, environmental risk assessment results, and parking risk assessment results, as input features to the inference engine. The inference engine analyzes and evaluates the data in each dimension. If the overall risk probability exceeds the preset threshold and the battery power is still sufficient, a decision result recommending that the sentry mode be turned on is generated; if the overall risk probability is lower than the preset threshold or the battery power is insufficient, a decision result recommending that the sentry mode be turned off is generated.
[0052] The inference engine supports at least two optional model inference architectures.
[0053] The first method uses LLM (Large Language Model) to achieve fusion reasoning: The system converts the vehicle's remaining battery power, environmental risk assessment results, parking risk assessment results, and historical accident statistics into natural language, constructs standardized reasoning prompts to input into the LLM, and outputs sentinel mode recommendation results through semantic parsing and logical deduction of the LLM.
[0054] The second approach uses MLP (Multi-Layer Perceptron) to achieve fusion reasoning: The system constructs a feature vector from the vehicle's remaining battery power, environmental risk assessment results, parking risk assessment results, and historical accident statistics. The feature vector is then input into the MLP, which performs multi-layer nonlinear mapping operations to obtain the risk level probability distribution (e.g., 80% is medium risk and 20% is high risk), and then outputs the sentinel mode recommendation result.
[0055] In this embodiment, the newly added dimension of historical accident statistics assesses the safety of the current vehicle parking location from a statistical perspective. This can correct or enhance the judgment results based on visual perception, enabling the system to combine visual perception with big data analysis to improve the accuracy of sentry mode decision-making.
[0056] In some embodiments, if historical accident statistics show that there have been frequent cases of vehicle theft and malicious scratching in the parking area recently, the system can proactively increase the risk weight of the historical accident statistics portion when performing fusion inference of multi-source heterogeneous data, and tend to output a suggestion to enable sentinel mode; conversely, if historical accident statistics show that the accident rate in the parking area is extremely low in the past 90 days, the system can appropriately decrease the risk weight of the historical accident statistics portion when performing fusion inference of multi-source heterogeneous data, to avoid unnecessary activation of sentinel mode and resulting power consumption.
[0057] In some embodiments, if the vehicle's vision sensor is not accurate enough and frequently captures interference information that is prone to triggering misjudgment, such as swaying tree branches or pedestrians passing by in the distance, the system can appropriately reduce the risk weight of the environmental risk assessment result and correspondingly increase the risk weight of other reliable data sources when performing fusion reasoning of multi-source heterogeneous data. This reduces the probability of false triggering of sentinel mode due to visual false alarms and improves the accuracy of recommendation results.
[0058] In some embodiments, a hard rule is also added as a fallback decision. For example, a battery safety threshold (e.g., 15%) is preset. When the remaining battery level of the vehicle is detected to be lower than this safety threshold, fusion reasoning is no longer performed (or although fusion reasoning is performed, the reasoning result is not directly adopted), and a decision to recommend turning off the sentry mode is directly output, or a prompt message (which may include the reasoning result) is pushed to the user, allowing the user to manually choose whether to turn on the sentry mode, to prevent the automatic operation of the sentry mode from causing excessive battery depletion.
[0059] In some embodiments, in addition to historical accident statistics, other data sources can be introduced, such as weather data (severe weather such as rain and fog increases risk), public security data (crime rate statistics in the area), and traffic flow data (higher risk during peak hours), to enrich the dimensions of risk assessment.
[0060] The historical accident statistics can be stored in a cloud-based big data service or in a lightweight risk database maintained locally on the vehicle. The latter is suitable for scenarios without network connectivity and has greater versatility.
[0061] In one embodiment shown, the method further includes: acquiring the ambient temperature around the vehicle, the vehicle model, and the operating mode of the Sentinel mode, to calculate the power consumption of the Sentinel mode per unit time based on the ambient temperature around the vehicle, the vehicle model, and the operating mode; the step of performing fusion reasoning based on the remaining battery power of the vehicle, the environmental risk assessment result, and the parking risk assessment result to obtain a recommended result for the Sentinel mode includes: performing fusion reasoning based on the remaining battery power of the vehicle, the environmental risk assessment result, the parking risk assessment result, and the power consumption of the Sentinel mode per unit time to obtain a recommended result for the Sentinel mode.
[0062] For example, when a vehicle is parked in an outdoor parking lot, the system obtains the current ambient temperature as -10℃, the vehicle model as a pure electric vehicle (model A), and the Sentinel mode operation mode as low-power mode. Then, based on this vehicle model, the system calls the baseline power consumption parameter corresponding to the low-power Sentinel mode, which is 1.5% / hour. This parameter represents the percentage of power consumption per unit time measured under standard test conditions (25℃) and standard power consumption Sentinel mode. Subsequently, the system introduces a temperature correction coefficient to correct the baseline power consumption parameter of 1.5% / hour. Since the current ambient temperature of -10℃ deviates from the standard test environment of 25℃, the corrected baseline power consumption parameter is determined to be 1.3 based on the battery temperature characteristic curve. Simultaneously, the system applies a weighting coefficient of 0.6 based on the low-power mode application mode (power consumption is 60% of standard power consumption). Finally, according to the formula 1.5% / hour × 1.3 × 0.6, the power consumption per unit time in low-power Sentinel mode under the current ambient temperature is calculated to be 1.17% / hour.
[0063] Then, the system further combines the power consumption per unit time of the low-power sentry mode, the vehicle's remaining power, environmental risk assessment results, and parking risk assessment results for fusion reasoning. For example, if the vehicle's current remaining power is 30%, the user's average parking time is 12 hours, and the preset minimum power reserve threshold is 10% (to ensure normal vehicle startup and basic function operation), based on the aforementioned calculation that the power consumption per unit time of the low-power sentry mode is 1.17% / hour, the system's expected power consumption for activating sentry mode within 12 hours is 14.04%. At this point, the system determines that the remaining power of 30% minus the expected consumption of 14.04% is 15.96%, which is higher than the minimum power reserve threshold of 10%, thus meeting the power requirements for activating sentry mode.
[0064] Furthermore, if the environmental risk assessment indicates a construction area near the current parking location (high risk level), and the parking risk assessment indicates the vehicle is crossing the lines and adjacent vehicles are parked incorrectly (medium risk level), the system tends to recommend activating Sentry Mode to ensure vehicle safety based on the comprehensive assessment of environmental and parking risks. Since the aforementioned battery level verification has passed, the system ultimately recommends activating Sentry Mode, maintaining a low-power operating mode to conserve battery power while ensuring basic safety.
[0065] In one embodiment shown, the method further includes: obtaining the vehicle parking time; the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the power consumption per unit time of the sentry mode to obtain the recommended result of the sentry mode includes: performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, the power consumption per unit time of the sentry mode, and the vehicle parking time to obtain the recommended result of the sentry mode.
[0066] For example, when a vehicle enters an open-air parking lot and engages the parking brake, the system first obtains the user's subsequent travel plan through the vehicle's navigation system. After analyzing the travel plan, it estimates that the user will restart the vehicle in 8 hours, thus determining the parking duration to be 8 hours. Simultaneously, the system obtains the vehicle's remaining battery level as 40% and, based on the aforementioned steps, calculates the power consumption per hour in Sentinel Mode under the current environment and low-power operation mode to be 1.2%. Based on this, the system calculates the total estimated power consumption for activating Sentinel Mode to be 9.6% (1.2% / hour × 8 hours). At this point, the system determines that if Sentinel Mode is activated throughout the journey, the remaining battery level when the vehicle restarts will be approximately 30.4% (40% - 9.6%), higher than the preset minimum reserve battery threshold (e.g., 20%), thus meeting the battery requirements for activating Sentinel Mode.
[0067] The system then further refines its recommendations by combining environmental risk assessment results with parking risk assessment results. For example, if the environmental risk assessment indicates the area is low-risk (e.g., inside a gated community), but the parking risk assessment shows the vehicle is parked next to a main road, posing a risk of scratches, the system ultimately recommends activating Sentinel Mode, maintaining a low-power operating mode to conserve battery power while ensuring basic safety.
[0068] The system can determine the Sentinel Mode operating mode for the current power consumption calculation based on the user's historical usage habits, the previous Sentinel Mode configuration, or user-preset parameters. For special scenarios such as long parking times, high-risk parking areas, or remaining battery power at critical thresholds, the system can simultaneously calculate the unit power consumption and total power consumption of Sentinel Mode in both low-power and standard operating modes, providing data support for fusion inference. In low-power mode, Sentinel Mode retains only basic functions, such as abnormal proximity detection and push notifications of abnormal information to the user's mobile phone. In standard mode, Sentinel Mode retains all functions, including features such as real-time remote viewing and multi-location backups.
[0069] In another scenario, assuming the vehicle is parked for 2 hours, but the environmental risk assessment indicates construction vibration in the vicinity, and the parking risk assessment indicates a neighboring vehicle is being moved, resulting in an extremely high overall risk level. At this point, the vehicle's remaining battery power is 20%, and the system calculates a total estimated power consumption of 3% (assuming the Sentinel Mode consumes 1.5% / hour × 2 hours in standard operation mode). The system determines that if Sentinel Mode is enabled throughout the entire process, the remaining battery power upon restarting the vehicle will be approximately 17% (20% - 3%), exceeding the preset minimum reserve battery threshold of 10%. Based on this, the system ultimately recommends enabling Sentinel Mode while using standard operation mode. This ensures vehicle safety through short-term, high-intensity monitoring, and the brief increase in energy consumption will not jeopardize the vehicle's normal starting and driving.
[0070] In this embodiment, by introducing the parking duration as a time dimension parameter, the system can distinguish between different parking scenarios such as short-term high-risk and long-term low-risk scenarios, thereby achieving dynamic matching between the security level of Sentinel Mode and energy consumption management, and taking into account both the parking safety protection effect and the vehicle's range economy.
[0071] Please see Figure 2 , Figure 2 This is a block diagram illustrating a vehicle sentinel mode recommendation architecture as an exemplary embodiment. Figure 2 As shown, the vehicle sentry mode recommendation architecture is divided into a data acquisition layer, a risk inference engine layer, and an interaction output and execution layer.
[0072] The data acquisition layer comprises a visual perception module, a data acquisition module, and a state management module. The visual perception module is responsible for collecting image information of the vehicle's surroundings and generating environmental risk assessment results and parking risk assessment results. The data acquisition module is used to acquire navigation information and positioning information to calculate the vehicle's parking duration. The state management module collects the vehicle's slope and remaining battery power in real time, and calculates the power consumption per unit time under different operating modes of Sentinel Mode. All data is then aggregated and sent down to the risk inference engine.
[0073] The risk inference engine layer consists of a feature fusion layer, an inference analysis layer, and a level judgment layer. The feature fusion layer receives multi-source information from the data acquisition layer and integrates the vehicle's remaining battery power, environmental risk assessment results, parking risk assessment results, power consumption per unit of Sentinel mode, and parking duration into unified input features. The inference analysis layer performs multi-dimensional information fusion calculations to comprehensively assess environmental risks, parking safety risks, and power consumption. The level judgment layer combines preset rules to output recommended Sentinel mode activation / deactivation results and appropriate operating modes.
[0074] The interaction output and execution layer includes the user interface, mobile application, and sentry mode controller. The user interface and mobile application are used to display recommended content to the user and receive user selection commands; the sentry mode controller receives the user's final decision command and starts or stops the sentry mode according to the user's selected operating mode, realizing a closed loop of parking security functions.
[0075] The data acquisition layer aggregates various types of data and then transmits them uniformly to the feature fusion layer of the risk inference engine. The feature fusion layer organizes and encapsulates multi-source heterogeneous data to construct a comprehensive risk assessment data structure. The data structure includes at least the following key fields: environmental risk score (0-100), number of risk events (discrete value), risk type (discrete value), parking standardization score (0-100, the inverse value is used for risk calculation in actual use), vehicle remaining battery percentage (0-100), sentry mode unit time power consumption percentage (0-100), estimated parking time (0-100), and big data risk score (0-100). The inference analysis layer can use LLM for inference.
[0076] To enable LLM to recognize and understand multi-dimensional data, the system pre-configures a standardized text inference input template. This template serves as a unified data transformation carrier, converting structured evaluation data into a coherent and readable natural language context, fully conveying all decision-making dimension information. This information is then input into the LLM deployed on the vehicle for inference. The inference input template can be: "The current vehicle parking environment analysis is as follows: Environmental risk score: (0-100), identified risk factors include: risk type (discrete value), parking standardization score (0-100), number of safety events that occurred at this point of interest in the past 90 days, big data risk score (0-100), vehicle remaining battery percentage (0-100), percentage of power consumption per unit time in sentry mode (0-100), estimated parking time (0-100). Please combine the above information to determine the current parking risk level (safe / low risk / medium risk / high risk) and output the recommended result of sentry mode (no need to turn on / recommend to turn on manually / automatically turn on when leaving the vehicle, low power mode / standard mode)."
[0077] In one embodiment, the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result for the Sentinel Mode includes: performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result and the reason for the recommendation for the Sentinel Mode; the method further includes: displaying the recommended result and the reason for the recommendation for the Sentinel Mode to the vehicle user, so that the vehicle user can select the recommended result, and determine whether the Sentinel Mode is enabled based on the selection.
[0078] For example, after a vehicle is parked, the inference engine, based on the vehicle's remaining battery power, environmental risk assessment results, and parking risk assessment results, performs multi-dimensional fusion inference and outputs a comprehensive risk level (e.g., safe, low risk, medium risk, high risk) and a sentry mode recommendation result (e.g., no need to activate, manual activation recommended, automatic activation upon leaving the vehicle). Simultaneously, the inference engine extracts the features most influential on this decision, such as narrow road passage in environmental risk, vehicle crossing lines in parking risk, and remaining battery power, and converts this feature data into user-readable natural language text to form the recommendation reason. After obtaining the sentry mode recommendation result and recommendation reason output by the inference engine, the system displays on the in-vehicle interface to the user: "Three minor collisions have occurred at the current parking location in the past 90 days, and the vehicle has sufficient remaining battery power. It is recommended to automatically activate sentry mode upon leaving the vehicle." The interface also includes "Confirm Activation" and "Cancel Activation" options. Users can combine the risk details displayed by the system with their own judgment to independently confirm or reject the action. The system ultimately determines the final activation / deactivation status of the sentry mode based on the user's selection.
[0079] The recommended content is displayed on the vehicle's central control screen and the user's linked in-vehicle mobile terminal, such as a smartphone. When the vehicle is parked or the user unlocks and gets back in, the central control screen displays a risk notification interface. This interface may include risk level indicators and icons, a list of triggering factors, and recommended actions. The risk level indicators can be visually displayed using color coding (e.g., yellow - low risk, orange - medium risk, red - high risk). The list of triggering factors highlights the key factors that led to the current risk assessment. When the risk level reaches medium or high risk, the system will also push a notification to the user via the mobile application. The notification includes a title, risk level, key risk factors, and options for the user to choose between enabling Sentry Mode, disabling Sentry Mode, or Do Not Disturb (to avoid notifications for the current time and a future period) on the mobile application.
[0080] In some embodiments, the in-vehicle interactive interface can use differentiated visual identifiers to distinguish decision results. A red identifier indicates that the Sentinel Mode is recommended to be turned on, and a green identifier indicates that the Sentinel Mode is recommended to be turned off. The corresponding reasons for the recommendation are fully displayed in paragraph text form, clearly listing the decision basis such as environmental hazards, parking risks, and battery status.
[0081] Please see Figure 3 , Figure 3 This is a schematic diagram illustrating a sentinel mode recommendation interface as shown in an exemplary embodiment. Figure 3As shown, this interface is a parking risk warning pop-up triggered after the vehicle is parked. The title "Parking Risk Warning" is displayed in the center of the top of the pop-up, and a close button (X) is located in the upper right corner for users to manually close the warning. The risk warning content is displayed in the middle of the pop-up in text form: "The current parking location is in a fire lane, and there have been 3 minor collisions in this shopping mall in the past 7 days. It is recommended to activate Sentry Mode to protect your vehicle." Below the text is a 3D rendering of a silver sedan to enhance the visual warning effect. There are three operation buttons at the bottom of the pop-up: "Confirm Activation," "Do Not Activate for Now," and "Do Not Disturb This Time" from left to right. Users can choose the appropriate operation based on their own judgment, and the system will execute the subsequent logic according to the user's choice. The design of this warning pop-up interface takes into account the efficiency of information acquisition and ease of operation for users in parking scenarios. Through clear risk warnings and intuitive interactive options, it helps users make quick decisions and effectively improves parking safety.
[0082] In one embodiment shown, the method further includes: generating and recording the vehicle user's preference based on the vehicle user's selection of the recommendation result; the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommendation result of the sentry mode includes: performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the vehicle user's preference to obtain the recommendation result of the sentry mode.
[0083] For example, after viewing the Sentry Mode recommendation results and risk reasons pushed by the system on the vehicle's central control screen, the user manually chooses to confirm or refuse to enable Sentry Mode. After user confirmation, the system retains the complete context of this interaction, including the current remaining battery level, environmental risk assessment results, parking risk assessment results, current vehicle location, operation timestamp, and user operation tags. The system continuously accumulates multiple rounds of interaction records, summarizes user behavior characteristics through statistical analysis, generates user preferences, and persistently stores them in a local or cloud-based user profile database. If multiple data points show that the user refuses to enable Sentry Mode even in medium-to-high-risk scenarios when the battery level is below 30%, the user is marked as battery-sensitive; if the user frequently manually enables Sentry Mode even in low-risk parking lots, the user is marked as safety-priority. In subsequent vehicle parking-triggered fusion inference, the system will retrieve this user preference as an adjustment parameter for calculation. For battery-sensitive users, the system will lower the battery threshold and more conservatively push recommendations to enable Sentinel Mode in scenarios with equal risk. For safety-first users, the risk assessment threshold will be lowered, making it easier to output recommendations to enable Sentinel Mode, ultimately generating personalized recommendations that fit the user's daily habits.
[0084] The preferences of vehicle users can be stored in a local vehicle or cloud-based user database and maintained in a structured data format, including preference types, sensitive parameter thresholds, and weight correction coefficients. User preference records are dynamically updated with each new user interaction, ensuring real-time tracking of user driving habits.
[0085] During the fusion inference process, vehicle user preferences are quantified into specific parameters or weights and integrated into the fusion inference computation framework: if a rule-based inference method is used, vehicle user preferences are reflected as a judgment threshold correction for specific conditions; if a neural network or large model-based inference method is used, vehicle user preferences are encoded as feature vectors or contextual cue words, which are input into the model along with feature vectors of vehicle remaining battery power, environmental risk assessment results, and parking risk assessment results. The model learns the nonlinear relationship between user preferences and other risk features through attention mechanisms or specific fully connected layers, and outputs personalized sentinel mode recommendation results.
[0086] In this embodiment, the remaining battery power of the vehicle, the results of environmental risk assessment, the results of parking risk assessment, and user habits are integrated and reasoned to achieve personalized recommendations, taking into account both parking safety protection and the differentiated needs of users, thereby improving the accuracy of the Sentinel Mode recommendations and the user experience.
[0087] Corresponding to the above-described vehicle sentry mode recommendation method, this disclosure also provides an embodiment of a vehicle sentry mode recommendation device.
[0088] Please see Figure 4 , Figure 4 This is an exemplary embodiment illustrating the hardware structure of an electronic device. At the hardware level, the device includes a processor 402, an internal bus 404, a network interface 406, memory 408, and non-volatile memory 410, and may also include other necessary hardware. One or more embodiments of this disclosure can be implemented in software, for example, the processor 402 reads the corresponding computer program from the non-volatile memory 410 into memory 408 and then runs it. Of course, besides software implementation, one or more embodiments of this disclosure do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.
[0089] Please see Figure 5 , Figure 5 This is a block diagram illustrating an exemplary embodiment of a vehicle sentry mode recommendation device 500. This vehicle sentry mode recommendation device 500 can be applied to, for example... Figure 4 The illustrated electronic device is used to implement the technical solution of this disclosure. The device includes: The first acquisition unit 502 is used to acquire images of the vehicle's surrounding environment and the vehicle's remaining battery power if the vehicle is detected to be parked. Analysis unit 504 is used to analyze the environmental risk type, the parking pose normality of the vehicle and the parking pose normality of other vehicles in the vehicle surrounding environment image, and obtain environmental risk assessment results and parking risk assessment results. The reasoning unit 506 is used to perform fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the sentry mode.
[0090] In some embodiments, the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result for the Sentinel Mode includes: If any of the preset conditions are met, it is recommended to enable the sentry mode. Otherwise, it is recommended to turn off the Sentinel mode; The preset conditions include: the remaining battery power of the vehicle is not lower than a preset battery power threshold, and the comprehensive risk assessment result jointly indicated by the environmental risk assessment result and the parking risk assessment result is higher than a preset risk threshold.
[0091] In some embodiments, the apparatus further includes: The query unit 508 is used to obtain the current coordinates of the vehicle and upload the current coordinates of the vehicle to the big data service, wherein the big data service queries the historical accident statistics data of the current coordinates of the vehicle. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the historical accident statistics.
[0092] In some embodiments, the apparatus further includes: The second acquisition unit 510 is used to acquire the ambient temperature around the vehicle, the vehicle model, and the operating mode of the sentry mode, so as to calculate the power consumption of the sentry mode per unit time based on the ambient temperature around the vehicle, the vehicle model, and the operating mode. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the power consumption per unit time in the Sentinel Mode.
[0093] In some embodiments, the apparatus further includes: The third acquisition unit 512 is used to acquire the vehicle parking time. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, the parking risk assessment results, and the power consumption per unit time in Sentinel Mode to obtain the recommended result for Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, the power consumption per unit time in Sentinel Mode, and the vehicle parking duration.
[0094] In some embodiments, the step of performing fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result to obtain the recommended result for the Sentinel Mode includes: The device further includes: a fusion reasoning process based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results and reasons for the Sentinel Mode; The display unit 514 is used to display the recommended results and reasons for the Sentinel Mode to the vehicle user, so that the vehicle user can select the recommended results and determine whether the Sentinel Mode is enabled based on the selection.
[0095] In some embodiments, the apparatus further includes: The generation unit 516 is used to generate and record the vehicle user's preferences based on the vehicle user's selection of the recommendation results; The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the vehicle user's preferences.
[0096] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.
[0097] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0098] The systems, devices, modules, or units described in the above embodiments can be implemented by a computer or entity, or by a product with a certain function. A typical implementation device is a computer, which can be a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.
[0099] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0100] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.
[0101] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.
[0102] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant regions, and corresponding operation portals shall be provided for users to choose to authorize or refuse.
[0103] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0104] The foregoing has described specific embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0105] The terminology used in one or more embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this disclosure. The singular forms “a,” “the,” and “the” as used in one or more embodiments of this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items.
[0106] It should be understood that although the terms first, second, third, etc., may be used to describe various information in one or more embodiments of this disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of one or more embodiments of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information.
[0107] The above description is merely a preferred embodiment of one or more embodiments of this disclosure and is not intended to limit the scope of one or more embodiments of this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this disclosure should be included within the scope of protection of one or more embodiments of this disclosure.
Claims
1. A vehicle sentry mode recommendation method, characterized in that, The method includes: If the vehicle is detected as parked, obtain images of the vehicle's surrounding environment and the vehicle's remaining battery power; The environmental risk type, the parking pose normality of the vehicle and the parking pose normality of other vehicles in the vehicle surrounding environment image are analyzed to obtain environmental risk assessment results and parking risk assessment results. The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, and the parking risk assessment result.
2. The method according to claim 1, characterized in that, The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: If any of the preset conditions are met, it is recommended to enable the sentry mode. Otherwise, it is recommended to turn off the Sentinel mode; The preset conditions include: the remaining battery power of the vehicle is not lower than a preset battery power threshold, and the comprehensive risk assessment result jointly indicated by the environmental risk assessment result and the parking risk assessment result is higher than a preset risk threshold.
3. The method according to claim 1, characterized in that, The method further includes: Obtain the current coordinates of the vehicle and upload them to a big data service. The big data service then queries the historical accident statistics data of the vehicle's current coordinates. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the historical accident statistics.
4. The method according to claim 1, characterized in that, The method further includes: The ambient temperature around the vehicle, the vehicle model, and the operating mode of the Sentinel mode are obtained to calculate the power consumption of the Sentinel mode per unit time based on the ambient temperature around the vehicle, the vehicle model, and the operating mode. The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the power consumption per unit time in the Sentinel Mode.
5. The method according to claim 4, characterized in that, The method also includes: obtaining the vehicle parking duration; The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, the parking risk assessment results, and the power consumption per unit time in Sentinel Mode to obtain the recommended result for Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, the power consumption per unit time in Sentinel Mode, and the vehicle parking duration.
6. The method according to claim 1, characterized in that, The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The method further includes: merging and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results and reasons for the Sentinel Mode; The vehicle user is shown the recommended results and reasons for the Sentry Mode, so that the vehicle user can select from the recommended results and determine whether the Sentry Mode is enabled based on the selection.
7. The method according to claim 6, characterized in that, The method further includes: Based on the vehicle user's selection of the recommendation results, the vehicle user's preferences are generated and recorded; The step of fusing and reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode includes: The recommended result for the Sentinel Mode is obtained by fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment result, the parking risk assessment result, and the vehicle user's preferences.
8. A vehicle sentry mode recommendation device, characterized in that, The device includes: The acquisition unit is used to acquire images of the vehicle's surrounding environment and the vehicle's remaining battery power if the vehicle is detected to be parked. The analysis unit is used to analyze the environmental risk type, the parking pose normality of the vehicle and the parking pose normality of other vehicles in the vehicle surrounding environment image, and to obtain environmental risk assessment results and parking risk assessment results. The reasoning unit is used to perform fusion reasoning based on the vehicle's remaining battery power, the environmental risk assessment results, and the parking risk assessment results to obtain the recommended results for the Sentinel Mode.
9. An electronic device, characterized in that, It includes a communication interface, a processor, a memory, and a bus, wherein the communication interface, the processor, and the memory are interconnected via the bus; The memory stores machine-readable instructions, and the processor executes the method according to any one of claims 1 to 7 by invoking the machine-readable instructions.
10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores machine-readable instructions, which, when invoked and executed by a processor, implement the method described in any one of claims 1 to 7.
