Search

CN-122009018-A - Vehicle-mounted object memorizing and retrieving system based on user behaviors

CN122009018ACN 122009018 ACN122009018 ACN 122009018ACN-122009018-A

Abstract

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

Inventors

  • WANG SHIRU
  • LI GAO
  • DING GUOLIANG

Assignees

  • 武汉江夏楚能汽车技术研发有限公司

Dates

Publication Date
20260512
Application Date
20260205

Claims (10)

  1. 1. A vehicle-mounted item memorizing and retrieving system based on user behavior, comprising: a behavior recognition module for recognizing a user's hidden behavior in the vehicle based on the 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 hidden behavior is identified; An intention detecting module for detecting the intention of the user to search for the hidden article, and And the prompt module is used for providing prompts of the position information for the user in a hidden mode when the searching intention is detected.
  2. 2. The system according to claim 1, characterized in that said behavior recognition module is specifically configured to: The method comprises the steps of collecting user behavior data through a plurality of sensor units arranged in a vehicle, wherein the plurality of sensor units at least comprise a camera in the vehicle, a seat pressure distribution sensor and a vehicle-mounted positioning device; And calculating confidence coefficient of the hiding behavior based on the user behavior data, and judging the hiding behavior when the confidence coefficient exceeds a preset threshold value.
  3. 3. The system of claim 2, wherein said calculating a confidence level for the hidden behavior based on the user behavior data comprises: Analyzing the in-vehicle camera data, identifying the gesture of the user, and determining a vision concealment score based on the body shielding degree and the head rotation frequency degree of the user; analyzing the data of the in-vehicle camera and the seat pressure distribution sensor, identifying the actions and the operation time of the user, and determining an action tardiness score and an operation time standardization score; analyzing the data of the vehicle-mounted positioning device, identifying the vehicle state, and determining scene abnormality scores; and carrying out weighted summation on the visual concealment score, the action delay score, the operation duration standardization score and the scene anomaly score, and calculating the confidence level of the concealed behavior.
  4. 4. A system according to any one of claims 1 to 3, wherein the location recording module employs an ultra wideband positioning system comprising a plurality of UWB anchors preset in the vehicle; When the hiding behavior is identified, starting each UWB anchor point and configuring the UWB anchor points into a radar working mode; Taking one UWB anchor point as a transmitter, taking the other UWB anchor points as receivers to carry out coherent scanning on key positions in a carriage, and synchronously recording real-time CIR data on all receiving and transmitting paths; Performing differential processing on the real-time CIR data and the previous scanning result, detecting the data subjected to the differential processing, and judging whether abnormal points exceeding a preset threshold exist or not; estimating the distance between the abnormal point and the transmitter or the receiver according to the signal time delay, and calculating the position of the abnormal point in the vehicle space by combining a receiving and transmitting path; If the estimated position is matched with the preset hidden position, the abnormal point is marked and recorded as the position information of the hidden article.
  5. 5. The system of claim 1, wherein the intent detection module is configured to detect seek semantics in a user's voice instructions by voice recognition and/or to detect user's turn-over behavior in a vehicle by behavioral analysis.
  6. 6. The system of claim 1, wherein the prompt module includes a privacy control unit for user authentication prior to providing the prompt, and wherein the location information is displayed only after the authentication is passed.
  7. 7. The system of claim 6, wherein the prompting module is configured to provide the prompt by at least one of 3D car model highlighting of a center control screen, voice broadcasting, an augmented reality head-up display, or a cell phone application.
  8. 8. The system of claim 1, further comprising a risk early warning module for sending a remote alert to the vehicle owner when the vehicle is in a sensitive geographic location or a non-vehicle owner operation is detected.
  9. 9. The system of claim 1, further comprising a context trigger module for automatically triggering a prompt based on a time condition comprising a concealed item storage time exceeding a preset threshold or an event condition comprising a vehicle repair, maintenance or account switch.
  10. 10. The system of claim 1, wherein the location recording module is further configured to store location information encrypted with a time stamp, a user voice description as a hidden event, and bound with a unique code.

Description

Vehicle-mounted object memorizing and retrieving system based on user behaviors Technical Field The invention relates to the technical field of automobiles, in particular to a vehicle-mounted object memory and retrieval system based on user behaviors. Background With the intelligent upgrade of the automobile industry, automobiles are gradually transformed from pure vehicles to a third living space, and the demands of users on the functionality and privacy of the space in the automobiles are increasingly prominent, wherein the storage and management of personal objects in the automobiles become a high-frequency use scene. In daily life, users often hide high-value or private articles such as cash, jewelry, important files and the like in hidden positions in a vehicle due to the anti-theft or privacy protection requirements, but the hiding behavior often has the temporary property and the concealment property, and the time lapse or the memory is fuzzy, so that the users are very easy to forget the specific storage positions of the articles. At present, the management of the vehicle to the internal storage space is still in the primary stage, and the prior art cannot effectively solve the core pain point of the user that the stored objects are easy to forget and the object is difficult to find, and has the following defects: First, there is a lack of recognition and passive storage memory capabilities for the user's "hidden" intent. The traditional vehicle only provides basic storage space, can not sense hiding behaviors of users, has no function of recording key information such as storage positions and time of articles, completely depends on self memory of the users, and is difficult to retrieve after forgetting. And secondly, the storage information is mutually isolated from the intelligent system of the vehicle. The state of the storage space in the vehicle does not form linkage with vehicle intelligent modules such as navigation, calendars, voice assistants and the like, and can not realize situational association and intelligent interaction and can not actively provide assistance based on the use situation of a user. Thirdly, the prior related technical proposal has obvious limitation, and the core pain points are not solved pertinently: The vehicle-mounted object list scheme is that part of vehicle types are scanned by the built-in cameras to generate a vehicle-mounted object list, so that only the 'what objects are in the vehicle' can be answered, the specific storage positions of the objects cannot be recorded, the visual searching guide capability is not provided, and the problems of 'where the objects are and how to find' cannot be solved; the intelligent storage box scheme is characterized in that the safety of the article is improved through an electronic lock, a biological identification or password unlocking mode, the core is focused on theft prevention rather than forgetting prevention, and the additional risk that the article cannot be taken out due to the fact that a user forgets a password exists; The article finder product is characterized in that a user is required to attach the anti-lost device to a target article in advance, the article finder product is only suitable for a preset anti-lost scene, is not suitable for temporary intentional hiding behaviors of the user, and cannot meet sudden privacy storage requirements. In addition, the problem caused by forgetting of users is further extended, namely, if the users forget to store the privacy articles in the vehicle when lending the vehicle to other people, repairing or selling, the articles are easy to permanently lose, and privacy disclosure or property loss is caused. In summary, there is a lack of a vehicle-mounted system in the prior art that can intelligently understand the "hiding" behavior intention of the user, automatically record the hiding position and related context information, and assist the user to complete article retrieval in a natural, intuitive and private manner, and the related art blank needs to be filled. The approaches described in this section are not necessarily approaches that have been previously conceived or pursued. Unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section. Similarly, the problems mentioned in this section should not be considered as having been recognized in any prior art unless otherwise indicated. Disclosure of Invention The invention aims to solve the problem that a user can forget the position easily after hiding an article in a vehicle, and provides a vehicle-mounted article memorizing and retrieving system based on user behaviors, which comprises the following specific technical scheme: A vehicle-mounted object memorizing and retrieving system based on user behaviors comprises a behavior recognition module, a position recording module, an intention detection modul