Vehicle control and intelligent route recommendation method and system based on driving personality recognition

Abstract

The present application relates to the technical field of vehicle intelligent control, in particular to a vehicle control and intelligent route recommendation method and system based on driving individuality recognition. Through the cooperative work of vehicle BCM, Bluetooth module, user mobile phone APP, GPS positioning module and navigation module, the identity recognition is completed by using the unique identifier of the driver binding Bluetooth device, the corresponding personalized vehicle parameters are automatically called, the automatic adaptation of comfortable functions such as seat, air conditioner, atmosphere lamp and seat is realized; at the same time, the driver's independent driving habit database is established, the driving mode, vehicle speed and energy recovery mode and other data are collected and analyzed, the driving preference model is formed, and the personalized route recommendation is completed combined with real-time traffic information and navigation data, the comfortable function intelligent recommendation is realized by fusing environmental temperature and user body state. The present application can realize the accurate personalized identification and control of vehicle functions in the scene of multiple drivers, improve the accuracy of driver identity recognition and the level of vehicle intelligent interaction.

Description

Technical Field

[0001] This invention relates to the field of vehicle intelligent control technology, and in particular to a vehicle control and intelligent route recommendation method and system based on driver personality recognition. Background Technology

[0002] With the rapid development of intelligent and connected vehicles, scenarios where multiple drivers share the same vehicle are becoming increasingly common (such as family cars and company vehicles). Different drivers have significantly different driving habits, comfort preferences, and travel needs. However, existing vehicle technology has the following prominent shortcomings, failing to meet these differentiated needs: Most existing vehicles lack automatic driver recognition. When different drivers switch vehicles, they must manually adjust parameters such as ambient lighting, air conditioning temperature, seat posture, and seat comfort functions. This is cumbersome, time-consuming, and laborious, severely impacting the user experience. Even vehicles with simple memory functions cannot automatically link driver identity with personalized parameters, resulting in poor adaptability. Furthermore, existing vehicle driving habit recording functions are mostly uniform, failing to differentiate between different drivers' habits. They cannot accurately capture each driver's driving preferences (such as driving mode, energy recovery intensity, and vehicle speed), making it difficult to provide customized services based on individual preferences. Furthermore, existing vehicle route recommendations are mostly based solely on objective factors such as distance and road conditions, without taking into account the driver's driving habits and travel preferences. They cannot recommend smooth routes for drivers who prioritize speed, nor can they recommend the optimal distance route for drivers who prefer convenience, resulting in insufficient practicality and specificity in route recommendations. Summary of the Invention

[0003] This invention overcomes the shortcomings of the prior art and provides a method and system for vehicle control and intelligent route recommendation based on driver personality recognition.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of this invention provides a vehicle control and intelligent route recommendation method based on driving personality recognition, comprising the following steps:

[0006] S01: The vehicle Bluetooth module is controlled by BCM to scan the Bluetooth device terminals bound in the designated area, obtain the unique Bluetooth broadcast identifier, and compare and match the unique Bluetooth broadcast identifier with the driver information and Bluetooth device information pre-stored on the vehicle server to confirm the identity of the main authorized driver user. If there are two main user identities, a prompt will pop up to manually control the selection.

[0007] S02: After BCM confirms the primary authorized driver user, it immediately sends a parameter call request instruction to the vehicle server backend to retrieve the driver user's pre-stored comfort driving actions and personalized vehicle usage parameters. It automatically configures the personalized vehicle usage parameters that best match the current scenario context of the vehicle through the comfort driving actions and sends instructions to the control execution unit to automatically complete the parameter adaptation.

[0008] S03: Based on BCM, driver identification is linked with the vehicle control unit to collect driving operation data of the primary authorized driver in real time. The driving operation data is used to automatically distinguish and record the driving habits and preferences of the primary authorized driver, and the data is stored and optimized and updated in association with the user's identity to build a personalized driving habit database for different primary authorized drivers.

[0009] S04: Combining the vehicle's current location collected by the GPS positioning module, the user-input destination, and real-time traffic information, multiple feasible routes are selected. Based on the driving habit data of different drivers stored on the backend server, feasible routes are prioritized and pushed to achieve targeted intelligent route recommendations.

[0010] Furthermore, S01 specifically includes the following steps:

[0011] The BCM controls the in-vehicle Bluetooth module to scan all Bluetooth devices in the driver's cabin area and obtain the unique Bluetooth broadcast identifier of each Bluetooth device terminal and its corresponding broadcast data packet.

[0012] Based on the broadcast data packets, the unique Bluetooth broadcast identifier is analyzed for manufacturer translation features to obtain the manufacturer translation feature vector sequence of the Bluetooth device terminal. The manufacturer translation feature vector sequence is then compared and matched with the Bluetooth device information corresponding to different drivers pre-stored in the vehicle server to initially identify the candidate Bluetooth pairing device of the identifier.

[0013] Based on the virtual identity data and authentication information when the driver information is bound to the candidate Bluetooth pairing device, a past identity configuration cost map is constructed. The past identity configuration cost map is demodulated and edited with the actual identity configuration cost map of the Bluetooth device terminal when the unique Bluetooth broadcast identifier is intercepted. The authentication match of the demodulation and editing cost is analyzed to identify and determine the identity of the primary authorized driver user, which is then uploaded to the vehicle display screen for prompt control.

[0014] If two authorized driving users, A and B, are identified at the same time, a selection prompt pop-up window for driving user A or driving user B will be loaded on the in-vehicle display screen for 10 seconds.

[0015] If the driver does not make a selection in the pop-up window within 10 seconds, the system will automatically and default to the driver's primary user authorization based on the Bluetooth identification sequence.

[0016] Furthermore, the step of performing vendor translation feature analysis on the unique Bluetooth broadcast identifier based on broadcast data packets to obtain the vendor translation feature vector sequence of the Bluetooth device terminal, and comparing and matching the vendor translation feature vector sequence with the Bluetooth device information corresponding to different drivers pre-stored in the vehicle server to initially identify the candidate Bluetooth pairing device of the identifier, specifically includes the following steps:

[0017] Extract the predefined vendor data of the unique Bluetooth broadcast identifier from the broadcast data packet, parse the protocol field of the vendor data, and obtain the vendor field structure of each Bluetooth device terminal;

[0018] The system obtains the Bluetooth device information database and Bluetooth broadcast protocol of the Bluetooth device terminal application, which are pre-stored in the background of the vehicle server and are bound to different driver information. It reads the device ID identifier fixed in the front byte of the manufacturer field structure, performs decomposition and masking processing on the device ID identifier based on the Bluetooth broadcast protocol, and obtains the manufacturer translated feature vector sequence.

[0019] A dynamic sequence planning matrix is ​​constructed based on the Bluetooth load data of the device. The manufacturer reprint feature vector sequence is input into the Bluetooth device information database. Then, the dynamic sequence planning matrix is ​​used to perform element-level backtracking matching with the prior manufacturer reprint feature vector sequence of Bluetooth devices corresponding to different drivers, and several common subsequences are output.

[0020] If the length of the common subsequence is less than the preset length, then the Bluetooth device information of the corresponding prior manufacturer's reprinted feature vector sequence of the common subsequence is marked as a candidate Bluetooth pairing device.

[0021] Furthermore, the process of constructing a past identity configuration cost map based on the virtual identity data and authentication information when binding and using the candidate Bluetooth pairing device according to the driver information, and then demodulating and editing the past identity configuration cost map with the actual identity configuration cost map of the Bluetooth device terminal when the unique Bluetooth broadcast identifier is intercepted, analyzing the authentication match of the demodulation and editing cost, identifying and determining the identity of the primary authorized driver user, and uploading it to the in-vehicle display screen for prompt control, specifically includes the following steps:

[0022] The system retrieves the virtual identity signature attribute of the driver when binding and using the candidate Bluetooth pairing device, as well as the temporary broadcast authorization value for the virtual identity signature attribute identity confirmation approval, through the vehicle server backend.

[0023] Using virtual identity signature attributes as authentication nodes and temporary broadcast authorization values ​​as authentication edges, a cost graph of past identity configurations is jointly constructed when driver information and candidate Bluetooth pairing devices are bound and used.

[0024] When the unique Bluetooth broadcast identifier is intercepted using the vehicle-mounted Bluetooth module, the actual identity configuration cost map of the Bluetooth device terminal is matched with the previous identity configuration cost map.

[0025] After registration, calculate the registration distance difference rate between each authentication node on the actual identity configuration cost map and each node on the previous identity configuration cost map, and preset the neighborhood attribute mapping degree of the label deviation of different authentication node pairs based on the registration distance difference rate.

[0026] Based on the neighborhood attribute mapping degree, perform mapping editing operations on the node insertion or modification of the actual identity configuration cost graph until the mapping is consistent with the previous identity configuration cost graph. Establish the demodulation sequence path of configuration cost editing based on the editing operations of all authentication node pairs.

[0027] During the mapping and editing process, the replacement and editing cost value between each authentication edge is calculated along the demodulation sequence path and accumulated to generate the global demodulation and editing cost. Based on the global demodulation and editing cost, the authentication matching degree between the actual identity configuration cost map and different previous identity configuration cost maps is determined.

[0028] Only the driver information corresponding to the previous identity configuration cost map with an authentication match greater than the preset authentication match is extracted and bound to the candidate Bluetooth pairing device. The driver information is determined to be the main authorized driver user identity and output to the vehicle display screen for prompt control.

[0029] Furthermore, S02 specifically includes the following steps:

[0030] After driver user identification, the coded parameter call request instruction is sent to the vehicle server. The server retrieves the comfort driving actions and pre-stored personalized vehicle parameters under the preset environmental conditions of the main authorized driver user. Combined with the personalized vehicle parameters, the server performs a priori confidence optimal configuration analysis on the comfort reward of different comfort driving actions under the current vehicle scenario context information, so as to automatically adapt the personalized parameters and generate the main authorized driver user's personalized parameter automatic adaptation strategy.

[0031] The personalized vehicle parameters include, but are not limited to, seat adjustment, ambient lighting control, seat heating, and ventilation settings;

[0032] The CAN bus sends control commands for the personalized parameter automatic adaptation strategy to different control execution components on the vehicle to automatically configure parameters for personalized driving.

[0033] Furthermore, S03 specifically includes the following steps:

[0034] The BCM works in conjunction with the vehicle control unit to collect driving operation data of the primary authorized driver in real time at a sampling interval of 10 seconds; the driving operation data includes average vehicle speed, driving mode settings, and energy recovery mode.

[0035] The average vehicle speed is quantified by the ratio of the actual vehicle speed to the road section speed limit to obtain the first driving operation quantification factor value; the driving modes are simultaneously quantified and classified and weighted according to the usage time ratio of each mode to obtain the second driving operation quantification factor value.

[0036] The energy recovery efficiency of the energy recovery mode is quantified to obtain the feedback mode coefficient. Based on the feedback mode coefficient, the usage time of different energy recovery modes is weighted and calculated to obtain the third driving operation quantification factor value.

[0037] The first driving operation quantification factor value, the second driving operation quantification factor value, and the third driving operation quantification factor value are weighted and summed to finally output the driving mode preference index.

[0038] If the driving mode preference index is less than or equal to the preset threshold, the primary authorized driver is classified as a convenience-preference driver; if the driving mode preference index is greater than the preset threshold, the primary authorized driver is classified as a speed-preference driver, thus obtaining the driving habit preference result.

[0039] The collected driving operation data and driving habit preference results are transmitted to the vehicle server in real time and associated with and stored with the identity information of the main authorized driver. The data is optimized and updated according to the increase of preset driving times to generate a personalized driving habit database.

[0040] Furthermore, S04 specifically includes the following steps:

[0041] The vehicle navigation module obtains the driving destination input in advance by the authorized driver, and combines the navigation module with the GPS positioning module to obtain the current driving position of the vehicle, and obtains real-time traffic information from a third-party traffic data interface.

[0042] Based on real-time road condition information, the system performs optimal route analysis on the city route bird's-eye view map of the distance between the destination and the current location, selects multiple feasible routes, prioritizes each feasible route according to the driving preference type of the primary authorized driver, obtains intelligent route recommendation schemes, and pushes them to the vehicle display screen or mobile terminal for broadcast and display.

[0043] The system records in real time the manual switching and route adjustment operations of the authorized driver for the intelligently recommended routes, and incorporates driving habit data for continuous updates to optimize the decision-making direction of subsequent route recommendations.

[0044] 8. The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 7, characterized in that the step of performing optimal path analysis on a bird's-eye view map of the city route from the destination to the current location based on real-time road condition information, and selecting multiple feasible driving routes, specifically includes the following steps:

[0045] Obtain a bird's-eye view map of the city routes within a preset range from the current location to the destination, and query the fixed-point geographic coordinates of the destination;

[0046] A route network model is established based on the main drivable routes and route intersections shown in the city route bird's-eye view map. The geospatial coordinate scalar domain of the city route bird's-eye view map is extracted, and the congestion obstacle index of different direct-drive routes in the route network model is preset based on real-time traffic information.

[0047] The root node is anchored by fixed geographic coordinates. The Dijkstra algorithm is introduced to perform a reverse route search in the geospatial coordinate scalar domain based on the congestion obstacle index. This results in the short-distance parent node of each city route node and the smooth flow effect rate of reaching different city route nodes from the fixed geographic coordinates. The reverse path representation of the short-distance parent node and the smooth flow effect rate is integrated to construct the destination reverse optimal path tree.

[0048] Traverse all direct drive routes in the route network model diagram and find the corresponding forward drive distance. If the path leaf node with the shortest forward drive distance cannot be found in the reverse optimal path tree of the endpoint, mark the direct drive route corresponding to the forward drive distance as a deviation from the edge route and organize all deviation from the edge routes to form a deviation from the edge route map.

[0049] The smooth flow inhibition degree of each traffic node in the deviation edge route map is extracted by the reverse optimal path tree of the endpoint. If the smooth flow inhibition degree is greater than the preset smooth flow inhibition degree threshold, the traffic node is marked as a deviation pile traffic node and the deviation pile traffic node sorting table is output.

[0050] The candidate path queue is expanded and filtered based on the off-pile traffic node sorting table to obtain K feasible driving routes.

[0051] A second aspect of the present invention provides a vehicle control and intelligent route recommendation system based on driving personality recognition. The system includes: a memory, a processor, and a communication interface. The memory includes a method program for vehicle control and intelligent route recommendation based on driving personality recognition. The communication interface is used for data connection communication between the memory and the processor. When the vehicle control and intelligent route recommendation method program is executed by the processor, it implements the steps of the vehicle control and intelligent route recommendation method described in any one of the present invention.

[0052] This invention addresses the technical deficiencies in the prior art, and its beneficial technical effects are as follows:

[0053] This invention constructs a driver-recognition-based vehicle personalized control and intelligent route recommendation system, achieving integrated and collaborative operation of driver identification, personalized vehicle control, driving habit learning, and intelligent navigation recommendation. The system automatically confirms driver identity using a unique Bluetooth device identifier and quickly retrieves the corresponding personalized vehicle configuration, automatically adapting parameters such as seat posture, ambient lighting, air conditioning, and seat comfort functions. This avoids frequent manual adjustments by different drivers, improving vehicle usability and human-machine interaction. Simultaneously, the system establishes independent driving habit databases for different drivers, continuously collecting, analyzing, and dynamically updating multi-dimensional driving behaviors such as driving modes, vehicle speed, and energy recovery modes to form accurate driving preference profiles, effectively avoiding data contamination issues caused by multiple drivers sharing a vehicle. Based on this, and combining real-time traffic information, vehicle location information, and driving preferences, candidate routes are personalized and prioritized. Routes with high traffic efficiency are recommended for speed-oriented drivers, while routes with shorter distances and fewer intersections are recommended for convenience-oriented drivers. This makes the navigation strategy more aligned with the driver's actual needs, improving travel efficiency and driving experience. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0055] Figure 1 A flowchart of the first method for vehicle control and intelligent route recommendation based on driver personality recognition is shown.

[0056] Figure 2 A flowchart of the second method for vehicle control and intelligent route recommendation based on driver personality recognition is shown;

[0057] Figure 3 A system framework diagram of a vehicle control and intelligent route recommendation system based on driver personality recognition is shown. Detailed Implementation

[0058] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0059] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0060] The first aspect of this invention provides a vehicle control and intelligent route recommendation method based on driving personality recognition, such as... Figure 1 As shown, it includes the following steps:

[0061] S01: The vehicle Bluetooth module is controlled by BCM to scan the Bluetooth device terminals bound in the designated area, obtain the unique Bluetooth broadcast identifier, and compare and match the unique Bluetooth broadcast identifier with the driver information and Bluetooth device information pre-stored on the vehicle server to confirm the identity of the main authorized driver user. If there are two main user identities, a prompt will pop up to manually control the selection.

[0062] S02: After BCM confirms the primary authorized driver user, it immediately sends a parameter call request instruction to the vehicle server backend to retrieve the driver user's pre-stored comfort driving actions and personalized vehicle usage parameters. It automatically configures the personalized vehicle usage parameters that best match the current scenario context of the vehicle through the comfort driving actions and sends instructions to the control execution unit to automatically complete the parameter adaptation.

[0063] S03: Based on BCM, driver identification is linked with the vehicle control unit to collect driving operation data of the primary authorized driver in real time. The driving operation data is used to automatically distinguish and record the driving habits and preferences of the primary authorized driver, and the data is stored and optimized and updated in association with the user's identity to build a personalized driving habit database for different primary authorized drivers.

[0064] S04: Combining the vehicle's current location collected by the GPS positioning module, the user-input destination, and real-time traffic information, multiple feasible routes are selected. Based on the driving habit data of different drivers stored on the backend server, feasible routes are prioritized and pushed to achieve targeted intelligent route recommendations.

[0065] It should be noted that the unique Bluetooth broadcast identifier includes the Bluetooth MAC address, the public device address, and a static random address.

[0066] Furthermore, the S01, as Figure 2 As shown, the specific steps include:

[0067] The BCM controls the in-vehicle Bluetooth module to scan all Bluetooth devices in the driver's cabin area and obtain the unique Bluetooth broadcast identifier of each Bluetooth device terminal and its corresponding broadcast data packet.

[0068] Based on the broadcast data packets, the unique Bluetooth broadcast identifier is analyzed for manufacturer translation features to obtain the manufacturer translation feature vector sequence of the Bluetooth device terminal. The manufacturer translation feature vector sequence is then compared and matched with the Bluetooth device information corresponding to different drivers pre-stored in the vehicle server to initially identify the candidate Bluetooth pairing device of the identifier.

[0069] Based on the virtual identity data and authentication information when the driver information is bound to the candidate Bluetooth pairing device, a past identity configuration cost map is constructed. The past identity configuration cost map is demodulated and edited with the actual identity configuration cost map of the Bluetooth device terminal when the unique Bluetooth broadcast identifier is intercepted. The authentication match of the demodulation and editing cost is analyzed to identify and determine the identity of the primary authorized driver user, which is then uploaded to the vehicle display screen for prompt control.

[0070] If two authorized driving users, A and B, are identified at the same time, a selection prompt pop-up window for driving user A or driving user B will be loaded on the in-vehicle display screen for 10 seconds.

[0071] If the driver does not make a selection in the pop-up window within 10 seconds, the system will automatically and default to the driver's primary user authorization based on the Bluetooth identification sequence.

[0072] It should be noted that the vehicle's BCM and Bluetooth module establish a stable communication linkage. When a user enters the driver's cabin area with a bound Bluetooth device (such as a phone bound to the user's mobile app, a smart Bluetooth key, etc.), the BCM controls the Bluetooth module to automatically scan for Bluetooth devices in the area. The scanned Bluetooth unique identifier is compared with the driver information and bound Bluetooth device information pre-stored in the backend server. Upon matching, the driver's identity is confirmed. The confirmed driver is typically a preset primary user A or primary user B, and can be extended to multiple authorized drivers. Simultaneous identification of users A and B usually indicates that A and B are using the vehicle simultaneously, or that either A or B is driving. The binding operation between the Bluetooth device and driver information is completed through the user's mobile app. Users can enter basic driver information in the app, such as nickname, gender, and date of birth, and select the Bluetooth device to be bound, completing the one-to-one binding of identity information and Bluetooth identifier.

[0073] Furthermore, the step of performing vendor translation feature analysis on the unique Bluetooth broadcast identifier based on broadcast data packets to obtain the vendor translation feature vector sequence of the Bluetooth device terminal, and comparing and matching the vendor translation feature vector sequence with the Bluetooth device information corresponding to different drivers pre-stored in the vehicle server to initially identify the candidate Bluetooth pairing device of the identifier, specifically includes the following steps:

[0074] Extract the predefined vendor data of the unique Bluetooth broadcast identifier from the broadcast data packet, parse the protocol field of the vendor data, and obtain the vendor field structure of each Bluetooth device terminal;

[0075] The system obtains the Bluetooth device information database and Bluetooth broadcast protocol of the Bluetooth device terminal application, which are pre-stored in the background of the vehicle server and are bound to different driver information. It reads the device ID identifier fixed in the front byte of the manufacturer field structure, performs decomposition and masking processing on the device ID identifier based on the Bluetooth broadcast protocol, and obtains the manufacturer translated feature vector sequence.

[0076] A dynamic sequence planning matrix is ​​constructed based on the Bluetooth load data of the device. The manufacturer reprint feature vector sequence is input into the Bluetooth device information database. Then, the dynamic sequence planning matrix is ​​used to perform element-level backtracking matching with the prior manufacturer reprint feature vector sequence of Bluetooth devices corresponding to different drivers, and several common subsequences are output.

[0077] If the length of the common subsequence is greater than the preset length, then the Bluetooth device information of the corresponding prior manufacturer's reprinted feature vector sequence of the common subsequence is marked as a candidate Bluetooth pairing device.

[0078] It's important to note that while Bluetooth devices from different manufacturers generally adhere to the Bluetooth broadcast protocol, each manufacturer typically defines its own proprietary data organization methods within the manufacturer data field of the broadcast data packet. Device ID identifiers, version information, product categories, and extended identifiers between different Bluetooth devices are all encoded according to their respective manufacturer protocols. When traditional MAC addresses can no longer stably identify a device due to random address rotation, privacy address mechanisms, or dynamic updates to broadcast content, vehicles relying solely on Bluetooth addresses or complete broadcast messages for identification are susceptible to address randomization, broadcast field changes, and data differences across different broadcast periods. This reduces the accuracy and stability of driver-device pairing and identification. Therefore, before matching the identity of a target Bluetooth device, it's necessary to first extract the manufacturer data corresponding to the unique Bluetooth broadcast identifier from the broadcast data packet, parse the protocol fields, and construct a manufacturer field structure that accurately represents the manufacturer's encoding rules. This is because the device ID identifier in Bluetooth manufacturer data is typically a core identity field permanently written by the manufacturer, and its encoding structure has high uniqueness, stability, and referential properties. By combining the Bluetooth device information database pre-stored in the vehicle server's backend with the corresponding manufacturer's Bluetooth broadcast protocol, the device ID identifier embedded in the front-end bytes of the manufacturer's field structure is parsed and decomposed using a masking process to obtain the manufacturer's translated feature vector sequence. According to the data protocol defined by the target manufacturer of the Bluetooth device terminal, the manufacturer's payload data is divided into multiple functional fields such as device type, device number, version information, rolling identifier, and sensor data, clarifying the device feature semantics and achieving fine-grained expression of the manufacturer's payload data. The decomposition masking process eliminates interference from protocol version differences, dynamic flags, check fields, and variable data on identity recognition, ensuring that the identifier transmission data generated in different broadcast cycles is mapped to a consistent feature expression space.

[0079] Furthermore, this method constructs a dynamic sequence planning matrix using device Bluetooth payload data. Some Bluetooth devices employ dynamic fields such as rolling identifiers, random numbers, or timestamps to enhance privacy protection. These fields are constantly changing, and directly participating in matching can easily reduce recognition accuracy. Therefore, the dynamic field sequence of the identification manufacturer data retains stable identity features for matching, enhancing the random adaptability of different Bluetooth device broadcast mechanisms. Based on the dynamic sequence planning matrix, element-level backtracking matching is performed between the real-time manufacturer translated feature vector sequence and the prior manufacturer translated feature vector sequences of Bluetooth devices corresponding to different drivers in the Bluetooth device information database. The database is used to find and quantify the common subsequences between the two, and the matching degree is determined based on the length of the common subsequence. Dynamic sequence planning can maintain the continuous matching ability of stable identity features even with insertions, deletions, field offsets, and local data changes, effectively extracting core encoded segments that are consistent between real-time broadcast data and historical device features. The common subsequence measures the similarity of the manufacturer protocol structure and the degree of identity consistency between the real-time Bluetooth device broadcast features and the prior device features in the Bluetooth device information database. When the length of the common subsequence is less than the preset length value, it indicates that there is a highly enriched and stable correlation between the real-time broadcast data and the corresponding prior vendor translation features, meeting the high-confidence identity verification requirements. This means that the matching degree between these pre-stored Bluetooth devices and the scanned Bluetooth device terminals is high. Therefore, the corresponding Bluetooth device information is marked as a candidate Bluetooth pairing device, abandoning the traditional direct binding of driver identity. This provides a reliable candidate set of Bluetooth information database individuals for subsequent secondary screening of driver identity features by combining RSSI changes, broadcast behavior features, and temporal continuity. This method can fully utilize the inherent and stable vendor identity information in Bluetooth vendor data to improve the robustness of Bluetooth device identity recognition in random address environments, effectively reduce the probability of Bluetooth mismatch caused by dynamic changes in broadcast data, and improve the accuracy, stability, and overall reliability of binding and recognizing different drivers with Bluetooth device identifiers.

[0080] Furthermore, the process of constructing a past identity configuration cost map based on the virtual identity data and authentication information when binding and using the candidate Bluetooth pairing device according to the driver information, and then demodulating and editing the past identity configuration cost map with the actual identity configuration cost map of the Bluetooth device terminal when the unique Bluetooth broadcast identifier is intercepted, analyzing the authentication match of the demodulation and editing cost, identifying and determining the identity of the primary authorized driver user, and uploading it to the in-vehicle display screen for prompt control, specifically includes the following steps:

[0081] The system retrieves the virtual identity signature attribute of the driver when binding and using the candidate Bluetooth pairing device, as well as the temporary broadcast authorization value for the virtual identity signature attribute identity confirmation approval, through the vehicle server backend.

[0082] Using virtual identity signature attributes as authentication nodes and temporary broadcast authorization values ​​as authentication edges, a cost graph of past identity configurations is jointly constructed when driver information and candidate Bluetooth pairing devices are bound and used.

[0083] When the unique Bluetooth broadcast identifier is intercepted using the vehicle-mounted Bluetooth module, the actual identity configuration cost map of the Bluetooth device terminal is matched with the previous identity configuration cost map.

[0084] After registration, calculate the registration distance difference rate between each authentication node on the actual identity configuration cost map and each node on the previous identity configuration cost map, and preset the neighborhood attribute mapping degree of the label deviation of different authentication node pairs based on the registration distance difference rate.

[0085] Based on the neighborhood attribute mapping degree, perform mapping editing operations on the node insertion or modification of the actual identity configuration cost graph until the mapping is consistent with the previous identity configuration cost graph. Establish the demodulation sequence path of configuration cost editing based on the editing operations of all authentication node pairs.

[0086] During the mapping and editing process, the replacement and editing cost value between each authentication edge is calculated along the demodulation sequence path and accumulated to generate the global demodulation and editing cost. Based on the global demodulation and editing cost, the authentication matching degree between the actual identity configuration cost map and different previous identity configuration cost maps is determined.

[0087] Only the driver information corresponding to the previous identity configuration cost map with an authentication match greater than the preset authentication match is extracted and bound to the candidate Bluetooth pairing device. The driver information is determined to be the main authorized driver user identity and output to the vehicle display screen for prompt control.

[0088] It should be noted that, given that multiple drivers may share a vehicle, multiple Bluetooth devices may connect alternately, and there may be situations such as device replacement, system upgrades, and authorization updates, relying solely on static broadcast characteristics for identity verification can easily lead to decreased reliability of identity matching when different drivers use the same type of device or when the same driver changes devices. Furthermore, while the Bluetooth broadcast identifier, virtual identity signature attributes, and temporary broadcast authorization relationships of the Bluetooth devices bound to a driver are generally highly stable over long-term use, they can still be affected by factors such as device firmware upgrades, broadcast randomization strategies, dynamic adjustments to authorization relationships, device replacement, system resets, and temporary pairing activities. This can result in localized changes in the identity configuration relationships formed by the same driver at different times. Therefore, when the vehicle's Bluetooth module intercepts a unique Bluetooth broadcast identifier, directly determining the driver's identity based solely on a single unique Bluetooth identifier, fixed identity attributes, or simple feature matching methods is insufficient to comprehensively describe the overall structural characteristics of the Bluetooth device's identity configuration relationships. Therefore, after initial screening of the Bluetooth paired devices, further authentication must be performed by combining the historical identity configuration relationships established between the driver and the device. A historical identity configuration cost graph is constructed by using virtual identity signature attributes and temporary broadcast authorization values ​​to construct a historical identity configuration cost graph for the driver and candidate Bluetooth pairing devices. The virtual identity signature attribute stably represents the historical identity association features formed between the driver and the Bluetooth device, while the temporary broadcast authorization value records the authorization relationship and interaction behavior of the device during the identity authentication process. Together, they contribute to the historical authorization topology of the driver and Bluetooth device, accurately describing the historical identity configuration state of the driver and candidate Bluetooth pairing devices. Subsequently, the registration distance difference rate between each authentication node in the actual identity configuration cost graph and the corresponding authentication node in the historical identity configuration cost graph is calculated. The registration distance difference rate quantifies the degree to which changes in distance between authentication nodes reflect changes in identity configuration relationships. Based on the registration distance difference rate, the degree of deviation of different authentication nodes from the corresponding tag neighborhood attribute mapping is determined. The neighborhood attribute mapping degree describes the consistency level between the local topology structure and the historical configuration structure of the authentication node. This allows for a comprehensive comparison of neighborhood connection relationships in addition to comparing attribute content between authentication nodes, improving the stability of driver and Bluetooth device identity configuration recognition and its resistance to local anomaly interference.

[0089] Based on the neighborhood attribute mapping degree, the actual identity configuration cost graph is subjected to mapping editing operations such as node insertion, node deletion, node replacement, and node attribute modification. This allows the actual identity configuration cost graph to gradually converge towards the historical identity configuration cost graph. The cost graph editing simulates the structural adjustment process required to restore the actual identity configuration relationship to the historical authorization relationship. A demodulation sequence path for configuration cost editing is established based on the editing process of all authentication nodes. The demodulation sequence path fully records the execution order and correlation of each editing operation, making the interpretability of the identity relationship configuration evolution concrete. The replacement editing cost value between each authentication edge is calculated sequentially along the demodulation sequence path and the replacement editing cost value is accumulated. The authentication edge corresponds to the authorization relationship between the driver and the Bluetooth device. The editing cost value generated by the replacement of the authentication edge can quantify the degree of difference between the historical authorization relationship and the current authorization relationship. The demodulation editing global cost comprehensively reflects the total cost required to adjust all nodes and edge relationships between the two identity configuration graphs. The smaller the demodulation editing global cost, the closer the actual identity configuration relationship is to the historical identity configuration relationship, and the higher the corresponding authentication consistency. Conversely, it indicates that the identity configuration relationship has changed significantly, and the authentication consistency is low. Therefore, authentication consistency can accurately measure the overall consistency between the current Bluetooth device's actual identity configuration and the historical driver authorization relationship. This method, after initial screening of Bluetooth device identity information through matching Bluetooth vendor identifier features, then combines the identity configuration relationship with graph structure consistency authentication to verify whether the Bluetooth device identity and driver authorization relationship remain consistently consistent. This effectively avoids misjudgments caused by accidental broadcast data consistency, device replacement, multiple users sharing the device, or changes in authorization relationships, improving the accuracy, stability, and security of Bluetooth device identity authentication and ensuring that the in-vehicle system only grants corresponding human-machine interaction and vehicle control permissions to genuinely authorized driver users.

[0090] Furthermore, S02 specifically includes the following steps:

[0091] After driver user identification, the coded parameter call request instruction is sent to the vehicle server. The server retrieves the comfort driving actions and pre-stored personalized vehicle parameters under the preset environmental conditions of the main authorized driver user. Combined with the personalized vehicle parameters, the server performs a priori confidence optimal configuration analysis on the comfort reward of different comfort driving actions under the current vehicle scenario context information, so as to automatically adapt the personalized parameters and generate the main authorized driver user's personalized parameter automatic adaptation strategy.

[0092] The personalized vehicle parameters include, but are not limited to, seat adjustment, ambient lighting control, seat heating, and ventilation settings;

[0093] The CAN bus sends control commands for the personalized parameter automatic adaptation strategy to different control execution components on the vehicle to automatically configure parameters for personalized driving.

[0094] It should be noted that the ambient lighting control automatically adjusts the color, brightness, and lighting mode of the ambient lights based on the driver's preset preferences. For seat posture adjustment, automatic control controls the seat's fore-and-aft movement and backrest angle adjustment. The seat's fore-and-aft movement ranges from 0 to 20 cm with an adjustment precision of 1 cm, and the backrest angle adjustment ranges from 90° to 120° with an adjustment precision of 1°, adapting to the driver's seating posture habits and ensuring comfort during driving. Seat comfort function adjustment: automatically turns seat ventilation and seat heating functions on or off, matching the driver's daily usage habits and improving driving comfort.

[0095] For the recommended ergonomic seat heating, users can manually enter their menstrual cycle start and end dates using the menstrual cycle tracking function added to the mobile app. This data is encrypted and stored on the backend server, visible only to the current user, ensuring privacy and security. After the BCM identifies the driver, the backend server simultaneously retrieves the driver's menstrual cycle record information and obtains the current in-vehicle ambient temperature collected by the vehicle's built-in temperature sensor. This dual assessment triggers the recommendation, with the specific thresholds as follows: If the current driver is identified as being menstruating (the backend server compares the current date with the entered menstrual cycle start and end dates, with an error ≤ 2 days), regardless of the in-vehicle ambient temperature... Regardless of the ambient temperature, the vehicle's infotainment system automatically displays a recommendation for the seat heating function, which can be activated with a single click after user confirmation. If the system detects that the current interior temperature is below the human comfort threshold (preset at 18℃, adjustable manually via a mobile app from 16℃ to 22℃), the system will also display a recommendation for the seat heating function, regardless of whether the driver is menstruating, to enhance driving comfort. The recommendation is displayed visually on the infotainment screen for 5 seconds, and users can easily activate or deactivate the seat heating function via touchscreen operation, voice commands, or remote operation via the mobile app.

[0096] Specifically, in the embodiment of step S02, the process of retrieving the comfortable driving actions and pre-stored personalized vehicle parameters under preset environmental conditions of the primary authorized driver, and combining the personalized vehicle parameters to perform a priori confidence optimal configuration analysis on the comfort rewards of different comfortable driving actions under the current vehicle scenario context information, so as to automatically adapt the personalized parameters and generate an automatic adaptation strategy for the personalized parameters of the primary authorized driver, specifically includes the following steps:

[0097] The system obtains a series of comfortable driving actions of the main authorized driver under different preset environmental conditions by using vehicle driving logs, retrieves the personalized vehicle usage parameters that are defaulted or manually pre-stored in the background when the main authorized driver performs each comfortable driving action, and establishes a prior probability distribution matrix of parameter vectors corresponding to different comfortable driving actions based on the personalized vehicle usage parameters.

[0098] By sensing and acquiring the current scene context information and vehicle model information inside the driver's cab through vehicle sensor network, and retrieving vehicle model information and current scene context information based on big data, a vehicle use ecosystem simulation model and a pre-trained comfort reward prediction model adapted to the driver's cab are obtained.

[0099] The ridge regression algorithm is introduced, and the reward function of each comfortable driving action is linearly calculated based on the historical comfort feature matrix and historical comfort reward cumulative vector for different driving actions according to personalized vehicle parameters, so as to obtain multiple ridge regression reward parameters.

[0100] The current scene context information, prior probability distribution matrix and ridge regression reward parameters are injected into the comfort reward prediction model to predict the comfort reward and confidence interval of each comfort driving action performed by the main authorized driver under the current vehicle interior driver's cab context, so as to obtain the expected comfort reward value and its confidence score value.

[0101] Extract the comfortable driving action corresponding to the maximum upper confidence score, mark it as potential comfortable driving action, and deploy the potential comfortable driving action to the vehicle ecosystem simulation model for simulation execution. Output the true approximate comfort reward value, and calculate the error between the true approximate comfort reward value and the expected comfort reward value to obtain the comfort reward deviation value.

[0102] If the comfort reward deviation value is less than the preset threshold, the vehicle server backend directly retrieves the personalized vehicle usage parameters for configuration output; if the comfort reward deviation value is greater than the preset threshold, the vehicle server backend re-optimizes the configuration of the personalized vehicle usage parameters and then retrieves and outputs them; and generates an automatic adaptation strategy for the personalized parameters of the primary authorized driver user.

[0103] It should be noted that the current scenario context information includes vehicle operation, road environment, traffic conditions, weather conditions, and driving time. Different drivers typically have significantly different comfort preferences for personalized vehicle parameters such as seat position, air conditioning temperature, airflow, steering wheel height, rearview mirror angle, driving mode, and in-car atmosphere. These driving comfort preferences dynamically vary depending on factors such as vehicle model, driving environment, weather conditions, road conditions, time stage, and driving task. For example, the same driver may prefer lower air conditioning temperatures and higher airflow in hot weather, while preferring higher temperatures and heated seats in cold weather; in congested urban traffic, they may prefer comfort driving mode, while during highway cruising, they may prefer economy or intelligent cruise control mode. Therefore, if a vehicle always uses fixed historical parameters or only adjusts and restores its configuration based on the most recent setting, it will be difficult to accurately reflect the driver's current true comfort needs, leading to rigid adaptation of personalized parameters and a discrepancy between vehicle parameter configuration and the driving environment, thus reducing the driving comfort experience. By establishing a prior probability distribution matrix of parameter vectors for different comfortable driving actions through personalized vehicle usage parameters, historical driving logs can objectively reflect the parameter usage habits formed by drivers over a long period of time. The prior probability distribution matrix can statistically analyze the probability distribution patterns of various parameter configurations under different comfortable driving actions, fully summarizing the long-term stable personalized preferences of drivers, providing reliable historical prior beliefs for subsequent context-based decision-making, and improving the accuracy of personalized driving adaptation in real-time vehicle scenarios.

[0104] Because the personalized parameters in the historical comfort feature matrix are often strongly correlated—for example, variables such as air conditioning temperature and airflow, driving mode and steering assist level, and seat position and steering wheel position are prone to multicollinearity—traditional linear regression can easily lead to large fluctuations in model parameters. By introducing ridge regression, multiple ridge regression reward parameters are obtained through linear calculation of the reward function corresponding to each comfort driving action. Ridge regression effectively suppresses overfitting of model parameters by introducing regularization constraints, improving the stability and generalization ability of the reward function estimation, and making the reward parameters more accurately reflect the actual contribution of different personalized parameters to driving comfort rewards. The reward function always depends on contextual features for calculation; different contextual states correspond to different reward distribution patterns. Relying solely on historical driving data is insufficient to accurately reflect the current driving environment. Therefore, vehicle model, cabin environment state, and real-time context are all input as contextual features into the comfort reward prediction model. The comfort reward prediction model establishes a mapping relationship between the vehicle context environment and driving comfort level, enabling comfort reward prediction to fully consider differences in vehicle configuration and changes in the real-time driving environment, improving the accuracy of the mapping relationship between contextual features and rewards, and making subsequent personalized parameter adaptation more consistent with real-world driving scenarios. The expected comfort reward value measures the anticipated comfort gains from the current parameter configuration. The upper confidence score considers the uncertainty in the reward prediction process. When a certain comfort driving action has few historical samples but a large potential benefit, its corresponding confidence interval is relatively wide, thus providing an appropriate opportunity to explore with a higher upper confidence score. This balances the utilization of historical best practices with the ability to explore unknown comfort configurations, preventing the system from being limited to existing parameter configuration schemes for a long time and improving the learning efficiency of personalized parameters. By using a vehicle ecosystem simulation model to simulate the execution effect of potential comfort driving actions in a real vehicle environment in advance, it is equivalent to adding a virtual verification step before formal configuration. This can further verify the consistency between the reward prediction results and the actual execution effect of different comfort actions, promptly identify prediction deviations, reduce the risk of decreased comfort experience caused by directly deploying incorrect parameters, and improve the authenticity and reliability of personalized parameter configuration results.

[0105] Specifically, when the comfort reward deviation value is less than a preset threshold, it indicates that the predicted comfort action reward result configured by the current personalized parameters maintains a high degree of consistency with the simulation verification result, suggesting that the current personalized vehicle parameters can accurately meet the current comfort needs of the driver. Therefore, the vehicle control server directly retrieves the corresponding personalized vehicle parameters for configuration output. Conversely, if the deviation value is greater than a preset threshold, it indicates that there is a certain error, requiring re-optimization of the corresponding personalized vehicle parameters and re-output of the parameters. This method enables vehicle personalized parameters to continuously adapt to changes in user driving habits and environment, effectively improving the accuracy, stability, and intelligence level of vehicle comfort parameter adaptation in different driving scenarios, and optimizing the overall driving comfort experience of the primary authorized driver.

[0106] Furthermore, S03 specifically includes the following steps:

[0107] The BCM works in conjunction with the vehicle control unit to collect driving operation data of the primary authorized driver in real time at a sampling interval of 10 seconds; the driving operation data includes average vehicle speed, driving mode settings, and energy recovery mode.

[0108] The average vehicle speed is quantified by the ratio of the actual vehicle speed to the road section speed limit to obtain the first driving operation quantification factor value; the driving modes are simultaneously quantified and classified and weighted according to the usage time ratio of each mode to obtain the second driving operation quantification factor value.

[0109] The energy recovery efficiency of the energy recovery mode is quantified to obtain the feedback mode coefficient. Based on the feedback mode coefficient, the usage time of different energy recovery modes is weighted and calculated to obtain the third driving operation quantification factor value.

[0110] The first driving operation quantification factor value, the second driving operation quantification factor value, and the third driving operation quantification factor value are weighted and summed to finally output the driving mode preference index.

[0111] If the driving mode preference index is less than or equal to the preset threshold, the primary authorized driver is classified as a convenience-preference driver; if the driving mode preference index is greater than the preset threshold, the primary authorized driver is classified as a speed-preference driver, thus obtaining the driving habit preference result.

[0112] The collected driving operation data and driving habit preference results are transmitted to the vehicle server in real time and associated with and stored with the identity information of the main authorized driver. The data is optimized and updated according to the increase of preset driving times to generate a personalized driving habit database.

[0113] It should be noted that regarding vehicle speed Vehicle speed is quantified as the ratio of actual vehicle speed to the road segment's speed limit, with a value ranging from 0 to 1. The closer the ratio is to 1, the closer the vehicle speed is to optimal efficiency. The weighting percentage can be adjusted. The specific quantification formula is as follows:

[0114]

[0115] In the formula, This is the quantification factor value for the first driving operation. This refers to the actual driving speed. Speed ​​limits are imposed on roads.

[0116] Regarding driving modes Based on economic, standard, and sport classifications, the value is quantified and ranges from 0.6 to 1.0, and is defined as the driving mode coefficient. (Economy mode uses 0.6, Standard mode uses 0.8, Sport mode uses 1.0, which can be adjusted according to vehicle characteristics.) Based on this, set the Economy mode according to the usage time ratio of each driving mode. Standard mode Sports Mode The time-weighted calculation is performed, and the specific quantitative formula is as follows:

[0117]

[0118] In the formula, This is the quantification factor value for the second driving operation. Driving mode This is the driving mode coefficient.

[0119] For feedback mode For new energy vehicles, energy recovery efficiency is quantified and defined as the recovery mode coefficient. The value ranges from 0.6 to 1 (0.6 for larger rewards, 0.8 for normal rewards, and 1 for weak rewards). Based on this, a time-weighted calculation is performed according to the usage time of each reward mode, and the values ​​are set as larger rewards. Normal feedback Weak feedback The specific quantification formula is as follows:

[0120]

[0121] In the formula, This is the quantification factor value for the third driving operation. For energy feedback mode, Feedback mode coefficient.

[0122] Finally, the quantified values ​​of each factor are weighted and summed to obtain... ;

[0123]

[0124] Get the final Afterwards, according to The value is used to divide the range of values, when ≤2.1 (based on prolonged use of standard driving mode), the driver is classified as a convenience-oriented driver; when >2.1, classify the driver as a speed-preference driver.

[0125] Furthermore, S04 specifically includes the following steps:

[0126] The vehicle navigation module obtains the driving destination input in advance by the authorized driver, and combines the navigation module with the GPS positioning module to obtain the current driving position of the vehicle, and obtains real-time traffic information from a third-party traffic data interface.

[0127] Based on real-time road condition information, the system performs optimal route analysis on the city route bird's-eye view map of the distance between the destination and the current location, selects multiple feasible routes, prioritizes each feasible route according to the driving preference type of the primary authorized driver, obtains intelligent route recommendation schemes, and pushes them to the vehicle display screen or mobile terminal for broadcast and display.

[0128] The system records in real time the manual switching and route adjustment operations of the authorized driver for the intelligently recommended routes, and incorporates driving habit data for continuous updates to optimize the decision-making direction of subsequent route recommendations.

[0129] It should be noted that for speed-oriented drivers, routes with low traffic volume (real-time traffic flow ≤ 50 vehicles / km), excellent road conditions (road surface smoothness ≥ 90%), and smooth driving (congestion time ≤ 10%) are prioritized. These routes can guarantee higher driving speeds, meeting drivers' needs for driving performance and travel efficiency. Even if the route is slightly longer than the destination (maximum distance difference ≤ 2km), the overall travel time can be shortened by increasing driving speed. For convenience-oriented drivers, routes that are closest to the destination (distance difference ≤ 0.5km), have fewer intersections (≤ 2 intersections per kilometer), and stable traffic efficiency (congestion time ≤ 15%) are prioritized. This balances travel convenience and economy, reduces driving fatigue, and improves the travel experience.

[0130] Furthermore, the optimal route analysis based on the city route bird's-eye view map of the distance from the destination to the current location, based on real-time road condition information, and the selection of multiple feasible routes, specifically includes the following steps:

[0131] Obtain a bird's-eye view map of the city routes within a preset range from the current location to the destination, and query the fixed-point geographic coordinates of the destination;

[0132] A route network model is established based on the main drivable routes and route intersections shown in the city route bird's-eye view map. The geospatial coordinate scalar domain of the city route bird's-eye view map is extracted, and the congestion obstacle index of different direct-drive routes in the route network model is preset based on real-time traffic information.

[0133] The root node is anchored by fixed geographic coordinates. The Dijkstra algorithm is introduced to perform a reverse route search in the geospatial coordinate scalar domain based on the congestion obstacle index. This results in the short-distance parent node of each city route node and the smooth flow effect rate of reaching different city route nodes from the fixed geographic coordinates. The reverse path representation of the short-distance parent node and the smooth flow effect rate is integrated to construct the destination reverse optimal path tree.

[0134] Traverse all direct drive routes in the route network model diagram and find the corresponding forward drive distance. If the path leaf node with the shortest forward drive distance cannot be found in the reverse optimal path tree of the endpoint, mark the direct drive route corresponding to the forward drive distance as a deviation from the edge route and organize all deviation from the edge routes to form a deviation from the edge route map.

[0135] The smooth flow inhibition degree of each traffic node in the deviation edge route map is extracted by the reverse optimal path tree of the endpoint. If the smooth flow inhibition degree is greater than the preset smooth flow inhibition degree threshold, the traffic node is marked as a deviation pile traffic node and the deviation pile traffic node sorting table is output.

[0136] The candidate path queue is expanded and filtered based on the off-pile traffic node sorting table to obtain K feasible driving routes.

[0137] It's important to note that when planning navigation routes, different drivers typically consider a variety of factors, including road congestion, driving habits, road class, number of traffic lights, smoothness of travel, and tolerance for detours. They don't simply pursue or limit themselves to the shortest distance or time. For example, some drivers prefer highways or urban expressways to ensure continuous travel, while others prefer to avoid toll roads, complex intersections, or highly congested areas. By using a shortest path tree to globally represent the optimal connectivity of the entire road network to the destination, each urban route traffic node has a unique short-distance parent node. Simultaneously, it preserves the globally optimal smoothness efficiency from all nodes to the destination, allowing for rapid expansion of candidate paths without repeated shortest path searches, thus improving overall route planning efficiency. Furthermore, the smoothness efficiency reflects road length, road congestion, and destination accessibility, enabling the reverse optimal path tree to accurately reflect the optimal connection of dynamic traffic under real-time road conditions, improving the practical feasibility of route searches. This method identifies all deviation margin routes that are not part of the shortest path tree but may constitute alternative paths by traversing the forward drive distance of all direct-drive routes in the route network model graph. All candidate paths are derived from finite deviations based on the optimal path tree, rather than re-searching the road network. Therefore, deviation margin routes can accurately represent potential alternative passages in the road network, retaining a high degree of similarity to the optimal path while possessing the ability to bypass obstacle areas, thus improving the accuracy of subsequent candidate path expansion. The method uses a smoothness inhibition degree to describe the comprehensive traffic capacity loss of each deviation node relative to the optimal path, quantifying the impact of different deviation nodes on the overall route quality. Based on the ranking results, a deviation priority is established, ensuring that subsequent path expansion always revolves around traffic nodes with better overall costs, improving the overall quality of candidate paths. Finally, by continuously combining different deviation margin routes through deviation pile traffic nodes, and always maintaining the output of candidate paths in ascending order of total cost, multiple different and high-quality feasible driving routes can be continuously obtained without repeatedly performing the shortest path search. This method fully considers real-time road conditions and the personalized driving preferences of the primary authorized driver, improving the real-time performance, intelligence, and multi-option decision-making capabilities of vehicle navigation route planning, and providing drivers with more stable, reliable, and habit-friendly route recommendations.

[0138] The second aspect of this invention provides a vehicle control and intelligent route recommendation system based on driving personality recognition, such as... Figure 3As shown, the system includes: a memory 301, a processor 302, and a communication interface 303. The memory 301 includes a vehicle control and intelligent route recommendation method program based on driving personality recognition. The communication interface 303 is used for data connection communication between the memory 301 and the processor 302. When the vehicle control and intelligent route recommendation method program is executed by the processor 302, it implements any of the steps of the vehicle control and intelligent route recommendation method described above.

[0139] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A vehicle control and intelligent route recommendation method based on driving personality recognition, characterized in that, Includes the following steps: S01: The vehicle Bluetooth module is controlled by BCM to scan the Bluetooth device terminals bound in the designated area, obtain the unique Bluetooth broadcast identifier, and compare and match the unique Bluetooth broadcast identifier with the driver information and Bluetooth device information pre-stored on the vehicle server to confirm the identity of the main authorized driver user. If there are two main user identities, a prompt will pop up to manually control the selection. S02: After BCM confirms the primary authorized driver user, it immediately sends a parameter call request instruction to the vehicle server backend to retrieve the driver user's pre-stored comfort driving actions and personalized vehicle usage parameters. It automatically configures the personalized vehicle usage parameters that best match the current scenario context of the vehicle through the comfort driving actions and sends instructions to the control execution unit to automatically complete the parameter adaptation. S03: Based on BCM, driver identification is linked with the vehicle control unit to collect driving operation data of the primary authorized driver in real time. The driving operation data is used to automatically distinguish and record the driving habits and preferences of the primary authorized driver, and the data is stored and optimized and updated in association with the user's identity to build a personalized driving habit database for different primary authorized drivers. S04: Combining the vehicle's current location collected by the GPS positioning module, the user-input destination, and real-time traffic information, multiple feasible routes are selected. Based on the driving habit data of different drivers stored on the backend server, feasible routes are prioritized and pushed to achieve targeted intelligent route recommendations. 2.The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 1, wherein, S01 specifically includes the following steps: The BCM controls the in-vehicle Bluetooth module to scan all Bluetooth devices in the driver's cabin area and obtain the unique Bluetooth broadcast identifier of each Bluetooth device terminal and its corresponding broadcast data packet. Based on the broadcast data packets, the unique Bluetooth broadcast identifier is analyzed for manufacturer translation features to obtain the manufacturer translation feature vector sequence of the Bluetooth device terminal. The manufacturer translation feature vector sequence is then compared and matched with the Bluetooth device information corresponding to different drivers pre-stored in the vehicle server to initially identify the candidate Bluetooth pairing device of the identifier. Based on the virtual identity data and authentication information when the driver information is bound to the candidate Bluetooth pairing device, a past identity configuration cost map is constructed. The past identity configuration cost map is demodulated and edited with the actual identity configuration cost map of the Bluetooth device terminal when the unique Bluetooth broadcast identifier is intercepted. The authentication match of the demodulation and editing cost is analyzed to identify and determine the identity of the primary authorized driver user, which is then uploaded to the vehicle display screen for prompt control. If two authorized driving users, A and B, are identified at the same time, a selection prompt pop-up window for driving user A or driving user B will be loaded on the in-vehicle display screen for 10 seconds. If the driver does not make a selection in the pop-up window within 10 seconds, the system will automatically and default to the driver's primary user authorization based on the Bluetooth recognition sequence. 3.The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 2, characterized in that, The step of performing vendor translation feature analysis on the unique Bluetooth broadcast identifier based on broadcast data packets to obtain the vendor translation feature vector sequence of the Bluetooth device terminal, and comparing and matching the vendor translation feature vector sequence with the Bluetooth device information corresponding to different drivers pre-stored in the vehicle server to initially identify the candidate Bluetooth pairing device of the identifier, specifically includes the following steps: Extract the predefined vendor data of the unique Bluetooth broadcast identifier from the broadcast data packet, parse the protocol field of the vendor data, and obtain the vendor field structure of each Bluetooth device terminal; The system obtains the Bluetooth device information database and Bluetooth broadcast protocol of the Bluetooth device terminal application, which are pre-stored in the background of the vehicle server and are bound to different driver information. It reads the device ID identifier fixed in the front byte of the manufacturer field structure, performs decomposition and masking processing on the device ID identifier based on the Bluetooth broadcast protocol, and obtains the manufacturer translated feature vector sequence. A dynamic sequence planning matrix is ​​constructed based on the Bluetooth load data of the device. The manufacturer reprint feature vector sequence is input into the Bluetooth device information database. Then, the dynamic sequence planning matrix is ​​used to perform element-level backtracking matching with the prior manufacturer reprint feature vector sequence of Bluetooth devices corresponding to different drivers, and several common subsequences are output. If the length of the common subsequence is less than the preset length, then the Bluetooth device information of the corresponding prior manufacturer's reprinted feature vector sequence of the common subsequence is marked as a candidate Bluetooth pairing device.

4. The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 2, characterized in that, The process involves constructing a past identity configuration cost map based on the virtual identity data and authentication information when binding and using candidate Bluetooth pairing devices with driver information. This past identity configuration cost map is then demodulated and edited against the actual identity configuration cost map of the Bluetooth device terminal when the unique Bluetooth broadcast identifier is intercepted. The authentication match of the demodulated and edited cost is analyzed to identify and determine the identity of the primary authorized driver user, which is then uploaded to the in-vehicle display screen for prompt control. Specifically, this includes the following steps: The system retrieves the virtual identity signature attribute of the driver when binding and using the candidate Bluetooth pairing device, as well as the temporary broadcast authorization value for the virtual identity signature attribute identity confirmation approval, through the vehicle server backend. Using virtual identity signature attributes as authentication nodes and temporary broadcast authorization values ​​as authentication edges, a cost graph of past identity configurations is jointly constructed when driver information and candidate Bluetooth pairing devices are bound and used. When the unique Bluetooth broadcast identifier is intercepted using the vehicle-mounted Bluetooth module, the actual identity configuration cost map of the Bluetooth device terminal is matched with the previous identity configuration cost map. After registration, calculate the registration distance difference rate between each authentication node on the actual identity configuration cost map and each node on the previous identity configuration cost map, and preset the neighborhood attribute mapping degree of the label deviation of different authentication node pairs based on the registration distance difference rate. Based on the neighborhood attribute mapping degree, perform mapping editing operations on the node insertion or modification of the actual identity configuration cost graph until the mapping is consistent with the previous identity configuration cost graph. Establish the demodulation sequence path of configuration cost editing based on the editing operations of all authentication node pairs. During the mapping and editing process, the replacement and editing cost value between each authentication edge is calculated along the demodulation sequence path and accumulated to generate the global demodulation and editing cost. Based on the global demodulation and editing cost, the authentication matching degree between the actual identity configuration cost map and different previous identity configuration cost maps is determined. Only the driver information corresponding to the previous identity configuration cost map with an authentication match greater than the preset authentication match is extracted and bound to the candidate Bluetooth pairing device. The driver information is determined to be the main authorized driver user identity and output to the vehicle display screen for prompt control. 5.The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 1, wherein, S02 specifically includes the following steps: After driver user identification, the coded parameter call request instruction is sent to the vehicle server. The server retrieves the comfort driving actions and pre-stored personalized vehicle parameters under the preset environmental conditions of the main authorized driver user. Combined with the personalized vehicle parameters, the server performs a priori confidence optimal configuration analysis on the comfort reward of different comfort driving actions under the current vehicle scenario context information, so as to automatically adapt the personalized parameters and generate the main authorized driver user's personalized parameter automatic adaptation strategy. The personalized vehicle parameters include, but are not limited to, seat adjustment, ambient lighting control, seat heating, and ventilation settings; The CAN bus sends control commands for the personalized parameter automatic adaptation strategy to different control execution components on the vehicle to automatically configure parameters for personalized driving. 6.The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 1, wherein, S03 specifically includes the following steps: The BCM works in conjunction with the vehicle control unit to collect driving operation data of the primary authorized driver in real time at a sampling interval of 10 seconds; the driving operation data includes average vehicle speed, driving mode settings, and energy recovery mode. The average vehicle speed is quantified by the ratio of the actual vehicle speed to the road section speed limit to obtain the first driving operation quantification factor value; the driving modes are simultaneously quantified and classified and weighted according to the usage time ratio of each mode to obtain the second driving operation quantification factor value. The energy recovery efficiency of the energy recovery mode is quantified to obtain the feedback mode coefficient. Based on the feedback mode coefficient, the usage time of different energy recovery modes is weighted and calculated to obtain the third driving operation quantification factor value. The first driving operation quantification factor value, the second driving operation quantification factor value, and the third driving operation quantification factor value are weighted and summed to finally output the driving mode preference index. If the driving mode preference index is less than or equal to the preset threshold, the primary authorized driver is classified as a convenience-preference driver; if the driving mode preference index is greater than the preset threshold, the primary authorized driver is classified as a speed-preference driver, thus obtaining the driving habit preference result. The collected driving operation data and driving habit preference results are transmitted to the vehicle server in real time and associated with and stored with the identity information of the main authorized driver. The data is optimized and updated according to the increase of preset driving times to generate a personalized driving habit database. 7.The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 1, wherein, S04 specifically includes the following steps: The vehicle navigation module obtains the driving destination input in advance by the authorized driver, and combines the navigation module with the GPS positioning module to obtain the current driving position of the vehicle, and obtains real-time traffic information from a third-party traffic data interface. Based on real-time road condition information, the system performs optimal route analysis on the city route bird's-eye view map of the distance between the destination and the current location, selects multiple feasible routes, prioritizes each feasible route according to the driving preference type of the primary authorized driver, obtains intelligent route recommendation schemes, and pushes them to the vehicle display screen or mobile terminal for broadcast and display. The system records in real time the manual switching and route adjustment operations of the authorized driver for the intelligently recommended routes, and incorporates driving habit data for continuous updates to optimize the decision-making direction of subsequent route recommendations. 8.The vehicle control and intelligent route recommendation method based on driving personality recognition according to claim 7, characterized in that, The optimal route analysis is performed on a bird's-eye view map of the city route from the destination to the current location based on real-time road condition information, and multiple feasible routes are selected. This includes the following steps: Obtain a bird's-eye view map of the city routes within a preset range from the current location to the destination, and query the fixed-point geographic coordinates of the destination; A route network model is established based on the main drivable routes and route intersections shown in the city route bird's-eye view map. The geospatial coordinate scalar domain of the city route bird's-eye view map is extracted, and the congestion obstacle index of different direct-drive routes in the route network model is preset based on real-time traffic information. The root node is anchored by fixed geographic coordinates. The Dijkstra algorithm is introduced to perform a reverse route search in the geospatial coordinate scalar domain based on the congestion obstacle index. This results in the short-distance parent node of each city route node and the smooth flow effect rate of reaching different city route nodes from the fixed geographic coordinates. The reverse path representation of the short-distance parent node and the smooth flow effect rate is integrated to construct the destination reverse optimal path tree. Traverse all direct drive routes in the route network model diagram and find the corresponding forward drive distance. If the path leaf node with the shortest forward drive distance cannot be found in the reverse optimal path tree of the endpoint, mark the direct drive route corresponding to the forward drive distance as a deviation from the edge route and organize all deviation from the edge routes to form a deviation from the edge route map. The smooth flow inhibition degree of each traffic node in the deviation edge route map is extracted by the reverse optimal path tree of the endpoint. If the smooth flow inhibition degree is greater than the preset smooth flow inhibition degree threshold, the traffic node is marked as a deviation pile traffic node and the deviation pile traffic node sorting table is output. The candidate path queue is expanded and filtered based on the off-pile traffic node sorting table to obtain K feasible driving routes.

9. A vehicle control and intelligent route recommendation system based on driving personality recognition, characterized in that, The system includes: a memory, a processor, and a communication interface. The memory includes a vehicle control and intelligent route recommendation method program based on driving personality recognition. The communication interface is used for data connection communication between the memory and the processor. When the vehicle control and intelligent route recommendation method program is executed by the processor, it implements the steps of the vehicle control and intelligent route recommendation method as described in any one of claims 1-8.

Images