Vehicle-mounted article memorizing and retrieving system based on user behaviors

Abstract

The invention relates to a vehicle-mounted article memorizing and retrieving system based on user behaviors, and the system comprises a behavior recognition module which is used for recognizing hiding behaviors of a user in a vehicle based on multi-sensor data; the position recording module is used for determining and recording the position information of the hidden article through the UWB technology when the hiding behavior is identified; the intention detection module is used for detecting a search intention of the user on the hidden articles; and the prompting module is used for providing a prompt of the position information to the user in a hidden mode when the searching intention is detected. According to the invention, intelligent identification of hiding behaviors of the user, accurate recording of article positions, timely detection of searching intentions and hidden prompt of position information are realized.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and more specifically to an in-vehicle item memory and retrieval system based on user behavior. Background Technology

[0002] With the intelligent upgrade of the automotive industry, cars are gradually transforming from simple means of transportation into a "third living space." Users' needs for the functionality and privacy of the car's interior space are becoming increasingly prominent, with the storage and management of personal belongings inside the car becoming a frequently used scenario. In daily life, users often hide high-value or private items such as cash, jewelry, and important documents in concealed locations inside the car for theft prevention or privacy protection. However, such hiding is often temporary and difficult to detect, and with the passage of time or memory fading, users are very likely to forget the exact location of the items.

[0003] Currently, vehicle internal storage space management is still in its early stages. Existing technologies have not yet effectively solved the core pain points of users such as "easily forgetting what they've stored" and "difficulty in finding what they've found." Specifically, there are the following shortcomings: First, they lack the ability to recognize and passively store information about a user's intention to "hide" items. Traditional vehicles only provide basic storage space and cannot detect a user's hiding behavior, nor do they have the function of recording key information such as the location and time of storage. They rely entirely on the user's memory, making it difficult to retrieve items once forgotten.

[0004] Secondly, storage information is isolated from the vehicle's intelligent system. The status of the in-vehicle storage space is not linked to the vehicle's intelligent modules such as navigation, calendar, and voice assistant, making it impossible to achieve contextual association and intelligent interaction, and unable to proactively provide assistance based on user scenarios.

[0005] Third, existing related technical solutions have obvious limitations and none of them specifically address the aforementioned core pain points: In-vehicle item inventory solutions: Some vehicle models generate an in-vehicle item list by scanning with a built-in camera. This can only answer "What items are in the car?", but cannot record the specific location of the items, nor does it have intuitive retrieval guidance capabilities. It fails to solve the problems of "Where are the items?" and "How to find them?" Smart locker solutions: These solutions enhance the security of items through electronic locks, biometrics, or password unlocking. They focus on "theft prevention" rather than "prevention of forgetting," and there is an additional risk that users may forget their passwords and be unable to retrieve their items. Item finder products: Users need to attach the anti-loss device to the target item in advance. They are only applicable to preset anti-loss scenarios and are completely unsuitable for users' spontaneous hiding behavior, and cannot meet the sudden privacy storage needs.

[0006] Furthermore, the problems caused by users forgetting extend further: when lending, repairing, or selling a vehicle, if users forget that there are still private items stored in the car, it can easily lead to the permanent loss of the items, resulting in privacy leaks or property losses.

[0007] In summary, there is a lack of in-vehicle systems in the current technology that can intelligently understand a user's intention to "hide," automatically record the hiding location and related contextual information, and assist users in retrieving items in a natural, intuitive, and private manner. This technological gap urgently needs to be filled.

[0008] The methods described in this section are not necessarily methods that had been previously conceived or adopted. Unless otherwise specified, no method described in this section should be assumed to be prior art simply because it is included in this section. Similarly, unless otherwise specified, the issues mentioned in this section should not be considered to be accepted in any prior art. Summary of the Invention

[0009] This invention aims to solve the problem that users easily forget the location of items they have hidden in their vehicles, and provides a vehicle-mounted item memory and retrieval system based on user behavior. The specific technical solution is as follows: A vehicle-mounted item memory and retrieval system based on user behavior includes a behavior recognition module, a location recording module, an intent detection module, and a prompting module. The behavior recognition module is used to identify users' hiding behavior inside the vehicle based on multi-sensor data; The location recording module is used to determine and record the location information of the hidden items using UWB technology when the act of hiding is detected. The intent detection module is used to detect the user's intent to find the hidden item; and The prompting module is used to provide the user with the location information in a covert manner when a search intent is detected.

[0010] This solution achieves a complete closed loop from behavior recognition to prompt retrieval, automatically capturing user intent and providing accurate prompts, fundamentally solving the core problem of users forgetting the location of items, and improving the intelligence and practicality of the system.

[0011] Preferably, the behavior recognition module is specifically used for: User behavior data is collected through a variety of sensor units installed inside the vehicle; the various sensor units include at least an in-vehicle camera, a seat pressure distribution sensor, and an on-board positioning device. The confidence level of the concealment behavior is calculated based on the user behavior data, and when the confidence level exceeds a preset threshold, it is determined to be concealment behavior.

[0012] By fusing multi-source data and determining confidence levels, the system can capture user behavior characteristics more comprehensively, reduce misjudgments caused by environmental interference, and improve the accuracy and reliability of behavior recognition.

[0013] Preferably, the step of calculating the confidence level of concealment behavior based on the user behavior data includes: By analyzing data from in-vehicle cameras, user posture is identified, and a visual concealment score is determined based on the degree of body occlusion and the frequency of head turning. By analyzing data from in-vehicle cameras and seat pressure distribution sensors, user actions and operation durations are identified, and action hesitation scores and operation duration standardization scores are determined. Analyze data from the vehicle positioning device to identify the vehicle's status and determine the scenario anomaly score; The confidence level of the concealment behavior is calculated by weighting and summing the visual concealment score, action hesitation score, operation time standardization score, and situational anomalousness score.

[0014] Preferably, the location recording module adopts an ultra-wideband positioning system, including multiple UWB anchor points pre-set in the vehicle; Upon detecting concealment behavior, activate each UWB anchor point and configure it to radar operating mode; Using one UWB anchor point as a transmitter and the other UWB anchor points as receivers, coherent scanning is performed on key locations inside the carriage, and real-time CIR data on all transmission and reception paths is recorded simultaneously. The real-time CIR data is differentially processed with the previous scan result, and the differentially processed data is detected to determine whether there are any anomalies exceeding a preset threshold. Based on the signal delay, the distance between the anomaly point and the transmitter or receiver is estimated, and combined with the transmission and reception path, the location of the anomaly point in the vehicle interior is deduced. If the calculated location matches a preset hiding location, the anomaly is marked and recorded as the location information of the hidden item.

[0015] Preferably, the intent detection module is used to detect the searching semantics in the user's voice commands through speech recognition, and / or to detect the user's rummaging behavior inside the vehicle through behavior analysis. This mechanism covers both voice and behavior as means of intent expression, enabling the system to flexibly respond to different user habits and improving the timeliness of detection and user experience.

[0016] Preferably, the prompting module includes a privacy control unit for authenticating the user before providing the prompt, and only displays the location information after successful authentication. Only authorized users can access sensitive information, effectively preventing privacy leaks and enhancing system security and user trust.

[0017] Preferably, the prompting module is used to provide prompts through at least one of the following methods: a 3D car model highlighted on the central control screen, voice broadcast, augmented reality head-up display, or mobile application.

[0018] Multiple prompting methods provide users with flexible choices to adapt to different usage scenarios (such as while driving or when stationary). Among them, AR-HUD and 3D car model provide intuitive guidance, improving retrieval efficiency and concealment.

[0019] Preferably, the system also includes a risk warning module, which sends a remote alert to the vehicle owner when the vehicle is located in a sensitive geographical location or when unauthorized operation is detected. This proactive warning capability enables the system to provide alerts in high-risk scenarios such as vehicle repairs or loan out, preventing item loss or privacy breaches and enhancing the system's proactive protection capabilities.

[0020] Preferably, the system also includes a scenario triggering module for automatically triggering prompts based on time conditions or event conditions. The time conditions include the storage time of the hidden items exceeding a preset threshold, and the event conditions include vehicle repair, maintenance, or account switching.

[0021] Through an automatic triggering mechanism, the system can provide timely reminders without user intervention, preventing items from being permanently lost due to prolonged forgetfulness, thus demonstrating the system's intelligence and proactiveness.

[0022] Preferably, the location recording module is also used to encrypt and store the location information, timestamp, and user voice description as a hidden event, and bind it with a unique code to ensure the security and traceability of the data. At the same time, the rich contextual information facilitates subsequent retrieval and improves the overall reliability of the system.

[0023] The beneficial effects of this invention are as follows: This invention provides a vehicle-mounted item memory and retrieval system based on user behavior, achieving intelligent recognition of user concealment behavior, accurate recording of item location, timely detection of search intent, and concealed location information prompts. The overall solution not only addresses the pain points of users easily forgetting what they've hidden and the difficulty of finding items, but also enhances the vehicle's intelligence level through multi-module collaborative work. The system possesses high privacy protection capabilities, adaptability, and user-friendly experience, upgrading the car from a mere means of transportation into an intelligent mobile companion. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of a vehicle-mounted item memory and retrieval system based on user behavior, provided as an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] In the description of this application, 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 indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0027] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0028] This invention provides a vehicle-mounted item memory and retrieval system based on user behavior. Its core lies in achieving an intelligent closed loop from user concealment behavior recognition to item retrieval prompts through the collaborative work of multiple modules. For example... Figure 1 As shown ( Figure 1 (This is a schematic diagram of the system structure of the present invention). The system can be integrated into the vehicle's electronic control unit (ECU) or in-vehicle infotainment system, implemented through a combination of hardware and software. The system mainly includes a behavior recognition module, a location recording module, an intent detection module, and a prompting module. Each module interacts with data via a vehicle bus (such as a CAN bus) or wireless communication (such as Bluetooth or Wi-Fi). In practical applications, the system can be deployed on an in-vehicle processor and connected to various in-vehicle sensor units to form a distributed or centralized architecture.

[0029] Specifically, upon system startup, an initial self-test is performed to ensure that all sensor units (such as the in-vehicle camera, seat pressure distribution sensor, UWB anchor points, etc.) are functioning correctly. The system defaults to a low-power monitoring mode to conserve energy; upon detecting a user entering the vehicle, it automatically switches to active mode. The implementation method is described in detail below with reference to the accompanying drawings and modules.

[0030] The behavior recognition module is used to identify a user's concealment behavior inside the vehicle based on multi-sensor data. The implementation of this module relies on various sensor units pre-installed in the vehicle, including but not limited to in-vehicle cameras, seat pressure distribution sensors, and onboard positioning devices (such as GPS or inertial navigation systems). These sensor units are connected to the behavior recognition module via data acquisition circuitry, and the sampling frequency can be adjusted according to the scenario; for example, the camera acquires video streams at 30fps, and the pressure sensor acquires pressure distribution data at 100Hz.

[0031] In practice, the behavior recognition module calculates the confidence level of concealment behavior through the following steps: Data Acquisition and Preprocessing: In-vehicle cameras capture user posture video streams, seat pressure distribution sensors monitor seat pressure changes in real time, and on-board positioning devices acquire vehicle position and motion status. Raw data undergoes filtering and noise reduction to minimize environmental interference.

[0032] Feature extraction and score calculation: Visual camouflage score: By analyzing data from in-vehicle cameras, a pre-trained convolutional neural network (CNN) model or pose recognition model (such as OpenPose) is used to identify the user's posture. For example, the system detects whether the user's body obstructs the operating area (such as the glove box or under the seat), and the frequency of head turning (frequent turning may indicate alertness). The score is normalized to a range of 0-1 based on the proportion of obstructed area and the frequency of head turning.

[0033] Action hesitation score: Combining camera and pressure sensor data, the smoothness of user actions is identified. For example, the speed of action is analyzed by optical flow algorithm. If the action is slow or has pauses, the score is higher. At the same time, pressure sensor data is used to detect whether the user frequently adjusts their sitting posture, indicating hesitant behavior. The value range is [0-1].

[0034] Standardized score for operation duration: Record the time from the start to the end of the user's operation. The operation duration is calculated as (t_actual - t_min) / (t_max - t_min). t_actual is the actual operation time, and t_min and t_max are empirical thresholds (e.g., set to 2 seconds and 15 seconds). If the time is too short or too long, it does not meet the concealment feature.

[0035] Scenario Anomaly Score: Based on data from the vehicle positioning device, the vehicle's location is determined and a value is assigned according to the vehicle's location. For example, if the vehicle is located in a home garage (within the GPS fence) and the engine is off, the score is 0.9; if it is located in a public parking lot, the score is 0.6; if the vehicle is in motion, the score is 0. The value range is [0,1].

[0036] Confidence calculation: The above scores are weighted and summed. The weights can be optimized by machine learning algorithms (such as support vector machines). The default settings are 40% for visual concealment, 30% for action hesitation, 20% for situational abnormality, and 10% for operation time. When the confidence exceeds the preset threshold (such as 0.7), the module determines it as a concealment behavior and triggers the location recording module.

[0037] This implementation method improves the accuracy of behavior recognition and reduces false positives by fusing multi-source data. For example, in testing, the system achieved an accuracy rate of over 95% in recognizing 100 simulated hiding behaviors.

[0038] The location recording module uses ultra-wideband (UWB) technology to determine and record the location information of concealed items. This module includes multiple pre-installed UWB anchor points within the vehicle, typically placed in key locations within the passenger compartment (such as doors, under seats, and the ceiling), forming a positioning network. The UWB anchor points utilize low-power chips (such as Decawave DW1000) and support radar operation. In this embodiment, at least five UWB anchor points are pre-installed.

[0039] Upon detecting concealment behavior, the module initiates the following process: Anchor point configuration and scanning: Each UWB anchor point is activated and configured in coherent radar mode. Using one anchor point as a transmitter and the others as receivers, a preset area within the vehicle (such as storage compartments or seat gaps) is scanned. The scanning cycle is once per second, simultaneously recording real-time channel impulse response (CIR) data on all transceiver paths.

[0040] Differential processing and anomaly detection: Real-time CIR data is differentially processed with the previous scan results to highlight areas of change. Threshold detection algorithms (such as peak detection) are used to determine if there are anomalies where signal strength changes exceed a preset threshold (such as 3dB). These anomalies may correspond to newly introduced items.

[0041] Location estimation and verification: Based on signal delay, the distance between the anomaly point and the anchor point is estimated using the Time Difference of Arrival (TDOA) algorithm. Combining multipath data, the coordinates of the anomaly point in the vehicle's three-dimensional space are calculated using triangulation. The system has a built-in location map that can be hidden (e.g., based on a CAD model). If the calculated location matches the map (e.g., the error is less than 5cm), the point is marked and its location information is recorded, along with a timestamp.

[0042] For example, in experiments, the UWB system achieved centimeter-level positioning accuracy for small items such as keys. The location information is encrypted and stored in the vehicle's onboard storage unit, and is linked to the user's identity to ensure security.

[0043] The intent detection module detects the user's intent to find hidden items, supporting both voice and behavioral detection methods. This module is integrated into in-vehicle voice assistants or behavior analysis systems.

[0044] 1) Speech Recognition: Employs an offline speech recognition engine (such as a deep learning-based model) to monitor user voice commands in real time. The system predefines relevant keywords for searching (such as "find something" or "where is my wallet"), and triggers a prompt module when a matching semantic is detected. The recognition process is handled locally to avoid privacy leaks.

[0045] 2) Behavioral Analysis: Detect user searching behavior using in-car cameras and pressure sensors. For example, use motion recognition algorithms (such as OpenPose) to analyze whether the user frequently bends over or rummages through storage compartments within a preset time period (e.g., 120 seconds); if the behavioral pattern's similarity to a preset searching template exceeds a threshold, it is determined to be a searching intent.

[0046] This module supports multimodal fusion; for example, when voice commands are ambiguous, it combines behavioral data to improve detection reliability. In implementation, the module adopts an event-driven mechanism, operates with low power consumption, and activates the system only when confidence is high.

[0047] The notification module is used to provide location information to users discreetly. This module includes a privacy control unit that implements user authentication (such as fingerprint or facial recognition), and displays information only after successful authentication.

[0048] Prompt methods: Supports multiple outputs, such as a 3D car model highlighted on the central control screen (highlighting the hidden location through the rendering engine), voice broadcast (played only through the car's speakers), augmented reality head-up display (AR-HUD, overlaying the prompt information onto the real-world view), or mobile application (pushing notifications via Bluetooth). Users can customize their preferred method.

[0049] Privacy protection: All notification data is transmitted in encrypted form and logs are automatically cleared after each session. For example, in a car-sharing scenario, the system requires the car owner's authorization to display historical records.

[0050] This implementation method ensures the timeliness and discreetness of the notification, avoiding attracting the attention of others.

[0051] The system can also be equipped with a risk warning module and a scenario triggering module to enhance proactive protection.

[0052] Risk warning module: Utilizing in-vehicle positioning devices and biometric sensors, this module detects whether the vehicle is located in a sensitive area (such as a repair shop) or whether it is being operated by someone other than the owner. If a risk is detected, the module sends an encrypted alert to the owner's mobile phone via cellular network, including a location snapshot and operation logs.

[0053] Scenario-triggered module: Automatically triggers prompts based on time conditions (such as items stored for more than 7 days) or event conditions (such as a vehicle being sent for repair). Event conditions are obtained through the vehicle bus (such as a maintenance mode signal), and the system reminds the user in advance to prevent forgetting.

[0054] While embodiments or examples of this disclosure have been described with reference to the accompanying drawings, it should be understood that the methods, systems, and devices described above are merely exemplary embodiments or examples, and the scope of the invention is not limited by these embodiments or examples, but only by the granted claims and their equivalents. Various elements in the embodiments or examples may be omitted or replaced by their equivalents. Furthermore, the steps may be performed in a different order than that described in this disclosure. Further, various elements in the embodiments or examples may be combined in various ways. Importantly, as the technology evolves, many elements described herein can be replaced by equivalents that appear after this disclosure.

[0055] The personal image and identity recognition technologies involved in this embodiment strictly comply with the provisions of the "Personal Information Protection Law of the People's Republic of China" and Article 5 of the "Patent Law of the People's Republic of China," and do not violate any laws, social ethics, or harm public interests. The personal images and identity recognition information collected by this technology do not originate from image collection or personal identification devices in public places, and explicit consent has been obtained from each data collection subject before information collection. The design and implementation of this technology does not involve mass image collection or identity recognition scenarios in public places, and does not require the application of special regulations for the installation of equipment in public places, but it still strictly complies with the general national requirements for personal information protection, fully guaranteeing the data collection subjects' right to know, right to consent, and information security. The collected information is used only for the legal purposes stipulated in the solution, without any design for abuse or illegal transfer, and will not harm the legitimate rights and interests of the public or the normal social order, complying with the requirements of public order and good morals and the protection of public interests.

Claims

1. A vehicle-mounted item memory and retrieval system based on user behavior, characterized in that, include: The behavior recognition module is used to identify users' hiding behavior inside the vehicle based on multi-sensor data; The location recording module is used to determine and record the location information of the hidden items using UWB technology when the act of hiding is detected. The intent detection module is used to detect the user's intent to find the hidden item; as well as The prompting module is used to provide the user with the location information in a covert manner when a search intent is detected.

2. The system according to claim 1, characterized in that, The behavior recognition module is specifically used for: User behavior data is collected through a variety of sensor units installed inside the vehicle; the various sensor units include at least an in-vehicle camera, a seat pressure distribution sensor, and an on-board positioning device. The confidence level of the concealment behavior is calculated based on the user behavior data, and when the confidence level exceeds a preset threshold, it is determined to be concealment behavior.

3. The system according to claim 2, characterized in that, The calculation of the confidence level of concealed behavior based on the user behavior data includes: By analyzing data from in-vehicle cameras, user posture is identified, and a visual concealment score is determined based on the degree of body occlusion and the frequency of head turning. By analyzing data from in-vehicle cameras and seat pressure distribution sensors, user actions and operation durations are identified, and action hesitation scores and operation duration standardization scores are determined. Analyze data from the vehicle positioning device to identify the vehicle's status and determine the scenario anomaly score; The confidence level of the concealment behavior is calculated by weighting and summing the visual concealment score, action hesitation score, operation time standardization score, and situational anomalousness score.

4. The system according to any one of claims 1 to 3, characterized in that, The location recording module uses an ultra-wideband positioning system, including multiple UWB anchor points pre-set in the vehicle; Upon detecting concealment behavior, activate each UWB anchor point and configure it to radar operating mode; Using one UWB anchor point as a transmitter and the other UWB anchor points as receivers, coherent scanning is performed on key locations inside the carriage, and real-time CIR data on all transmission and reception paths is recorded simultaneously. The real-time CIR data is differentially processed with the previous scan result, and the differentially processed data is detected to determine whether there are any anomalies exceeding a preset threshold. Based on the signal delay, the distance between the anomaly point and the transmitter or receiver is estimated, and combined with the transmission and reception path, the location of the anomaly point in the vehicle interior is deduced. If the calculated location matches a preset hiding location, the anomaly is marked and recorded as the location information of the hidden item.

5. The system according to claim 1, characterized in that, The intent detection module is used to detect the searching semantics in the user's voice commands through speech recognition, and / or to detect the user's rummaging behavior inside the vehicle through behavior analysis.

6. The system according to claim 1, characterized in that, The prompting module includes a privacy control unit, which is used to authenticate the user before providing the prompt, and only displays the location information after successful authentication.

7. The system according to claim 6, characterized in that, The prompting module is used to provide prompts through at least one of the following methods: a 3D car model highlighted on the central control screen, voice broadcast, augmented reality head-up display, or mobile application.

8. The system according to claim 1, characterized in that, It further includes a risk warning module, which sends a remote alert to the vehicle owner when the vehicle is located in a sensitive geographical location or when non-owner operation is detected.

9. The system according to claim 1, characterized in that, It further includes a scenario triggering module for automatically triggering prompts based on time conditions or event conditions. The time conditions include the storage time of the hidden item exceeding a preset threshold, and the event conditions include the vehicle being sent for repair or maintenance, or the account being switched.

10. The system according to claim 1, characterized in that, The location recording module is also used to encrypt and store location information, timestamps, and user voice descriptions as hidden events, and bind them with a unique code.

Images