Multi-passenger cooperative intelligent driving control system and method, vehicle and storage medium
By deconstructing driving control into multiple independent dimensions and combining multimodal perception and dynamic permission mapping, the central intelligent arbitrator handles conflicts, solving the balance between safety and efficiency in multi-occupant interaction and improving the human-machine co-driving experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN JIANGXIA CHUNENG AUTOMOBILE TECHNOLOGY R&D CO LTD
- Filing Date
- 2026-04-02
- Publication Date
- 2026-06-12
Smart Images
Figure CN122186185A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent connected vehicle technology, specifically to an intelligent driving control system, method, vehicle, and storage medium for multi-occupant collaboration. Background Technology
[0002] With the rapid development of autonomous driving technology and the widespread application of intelligent cockpit multi-screen interaction systems, vehicle interior space is undergoing a transformation from a single driver-dominated environment to a shared mobile space for multiple occupants. Multi-occupant collaboration aims to allow occupants in different positions within the vehicle to participate appropriately in the control and interaction process, thereby improving travel efficiency and passenger experience. Currently, various technical solutions exist in the industry for multi-occupant interaction. One common approach is to adopt a single control mode, where all driving-related control permissions, including core driving parameters such as following distance adjustment, driving mode switching, and navigation route setting, are limited to the driver's seat, with other occupants unable to intervene. Another technical solution attempts to distribute some control to non-driver occupants, such as enabling a complete transfer of driving control in emergencies, or opening up control of non-core functions like air conditioning and entertainment to rear-seat screens for independent operation.
[0003] However, the aforementioned technical solutions all have significant shortcomings in practical applications. Specifically, while the single control mode ensures a clear attribution of control, in scenarios such as family travel and business receptions, other passengers cannot offer helpful suggestions based on their own preferences or real-time road conditions, resulting in low efficiency in travel collaboration and limited passenger experience. Existing emergency takeover modes often involve a complete transfer of driving authority, representing an "all or nothing" switch of responsibility, posing significant safety risks. Solutions that distribute non-core functions across various screens only remain at a superficial interaction level, failing to achieve true human-machine co-driving value.
[0004] In summary, existing technologies consistently struggle to achieve an effective balance between ensuring basic driving safety and meeting the interactive needs of multiple occupants when handling multi-occupant interaction scenarios. How to improve the depth and efficiency of multi-occupant participation in driving-related interactions while ensuring the primary driver's ultimate control over the vehicle remains unaffected has become a key bottleneck restricting further improvements in the human-machine co-driving experience of intelligent driving. Summary of the Invention
[0005] In view of this, it is necessary to provide a multi-occupant collaborative intelligent driving control system, method, vehicle and storage medium to solve the technical problems of existing methods that make it difficult to balance driving safety and multi-occupant interaction efficiency, and limit the human-machine co-driving experience due to the single control authority.
[0006] To address the aforementioned technical problems, in a first aspect, the present invention provides a multi-occupant cooperative intelligent driving control system, comprising: The control control deconstruction module is used to deconstruct vehicle driving control into multiple independent driving parameter dimensions; A multimodal state perception network is used to perceive the identity and state of each occupant in the vehicle in real time and generate corresponding occupant state information. The dynamic permission mapping engine is used to dynamically determine, based on the real-time driving scenario and the occupant status information, to grant the target occupant permission to propose setting at least one of the driving parameter dimensions, so that the target occupant can initiate parameter setting proposals for the granted parameter dimensions. A central intelligent arbitrator is used to receive at least one of the parameter setting proposals, arbitrate the parameter setting proposals, and generate a proposal to be confirmed. The proposal confirmation and execution module is used to present the proposal to be confirmed, and after receiving the confirmation instruction from the main driver, to send the parameter setting instruction corresponding to the proposal to be confirmed to the vehicle control domain for execution.
[0007] In one possible implementation, the control deconstruction module deconstructs driving control into at least a longitudinal control domain, a lateral control domain, and a system configuration domain, with each domain including at least one independent driving parameter dimension.
[0008] In one possible implementation, the driving parameter dimensions of the longitudinal control domain include at least one of following distance, speed offset, power response mode, and energy recovery intensity. The driving parameter dimensions in the lateral control domain include at least one of navigation path preference, waypoint, lane selection preference, and lane change activism. The driving parameter dimension in the system configuration domain includes at least one of predictive driving mode, driving assistance sensitivity, and automation degradation strategy.
[0009] In one possible implementation, the dynamic permission mapping engine has a pre-defined policy rule base, which includes at least: Safety baseline rules are used to absolutely prohibit or lock the granting of setting proposal permissions to occupants based on vehicle speed, vehicle transient operating status, or vehicle malfunction status. Scene adaptation rules are used to establish a mapping relationship between different driving scenarios and the dimensions of available parameters; Crew role and status rules are used to determine target crew members based on their identity authentication results and crew status information, and to perform personalized allocation of open parameter dimensions to target crew members.
[0010] In one possible implementation, the occupant status information includes at least one of the following: occupant identification, gaze direction, attention state, and physical state.
[0011] In one possible implementation, the multi-occupant cooperative intelligent driving control system includes: The safety recovery module is configured to restore all driving parameters to the preset safety configuration in response to a recovery command triggered by the driver's seat.
[0012] On the other hand, the present invention also provides a multi-occupant cooperative intelligent driving control method, comprising: Deconstruct vehicle driving control into multiple independent driving parameter dimensions; Real-time perception of the identity and status of each passenger in the vehicle, generating corresponding passenger status information; Based on the real-time driving scenario and the occupant status information, dynamically determine the permission to grant the target occupant setting suggestion for at least one of the driving parameter dimensions; Receive parameter setting proposals from one or more occupants for open parameter dimensions, arbitrate the parameter setting proposals, and generate proposals to be confirmed; Upon receiving the confirmation instruction from the primary driver for the proposed confirmation, the parameter setting instruction corresponding to the proposed confirmation is sent to the vehicle control domain for execution.
[0013] In one possible implementation, arbitrating the parameter setting proposal to generate a proposal to be confirmed includes: The system detects whether there are multiple competing proposals for the same parameter dimension, and obtains the conflict detection results. If the conflict detection result indicates that a conflict exists, then according to a preset arbitration strategy, one of the multiple competing proposals is determined as the proposal to be confirmed. The preset arbitration strategy includes at least one of the following factors: occupant priority, occupant status information, occupant historical preference data, and the rationality assessment result of the proposal. If the conflict detection result is that there is no conflict, the received proposal will be directly used as the proposal to be confirmed.
[0014] Thirdly, the present invention also provides a vehicle including the aforementioned multi-occupant cooperative intelligent driving control system.
[0015] Fourthly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instructions, which, when executed by a processor, can implement the steps of the multi-occupant cooperative intelligent driving control method described in any of the above implementations.
[0016] The beneficial effects of this invention are as follows: The multi-occupant collaborative intelligent driving control system provided by this invention decomposes driving control rights into multiple independent dimensions through a control rights deconstruction module, achieving refined permission allocation; through the cooperation of a multimodal state perception network and a dynamic permission mapping engine, it ensures that only occupants with appropriate states can obtain proposal permissions in suitable scenarios; through a central intelligent arbitrator to handle multi-source proposal conflicts, it ensures the orderly integration of multi-occupant inputs; and through a proposal confirmation and execution module, it retains the final confirmation right of the primary driver, fundamentally guaranteeing driving safety. This achieves improved human-computer interaction efficiency and collaborative experience in multi-occupant scenarios while ensuring the primary driver's final control. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A schematic diagram of an embodiment of the multi-occupant cooperative intelligent driving control system provided by the present invention; Figure 2 A schematic flowchart of an embodiment of the intelligent driving control method for multi-occupant cooperation provided by the present invention; Figure 3 For the present invention Figure 2 A schematic diagram of an embodiment of S204. Detailed Implementation
[0019] In the description of the embodiments of the present invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.
[0020] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.
[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0022] This invention provides a multi-occupant collaborative intelligent driving control system, method, vehicle, and storage medium. The technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0023] Figure 1 This is a schematic diagram of an embodiment of the multi-occupant cooperative intelligent driving control system provided by the present invention, as shown below. Figure 1 As shown, the multi-occupant collaborative intelligent driving control system includes a control authority deconstruction module 101, a multimodal state perception network 102, a dynamic permission mapping engine 103, a central intelligent arbitrator 104, and a proposal confirmation and execution module 105, which are implemented as follows: The control control deconstruction module 101 is used to deconstruct the vehicle driving control into multiple independent driving parameter dimensions.
[0024] Specifically, traditional driving control is usually viewed as a whole, making it difficult to allocate control in a refined manner. This embodiment uses the control control deconstruction module 101 to deconstruct the driving control into multiple parameter dimensions that can be operated independently and do not interfere with each other, so as to subsequently assign setting proposal permissions to different target occupants.
[0025] Among them, the driving parameter dimension refers to the independently adjustable parameter unit related to vehicle driving control.
[0026] As an example, driving parameters may include following distance, speed offset, power response mode, navigation route preference, waypoints, and predictive driving mode.
[0027] A multimodal state perception network 102 is used to perceive the identity and state of each occupant in the vehicle in real time and generate corresponding occupant state information. This network can be deployed inside the vehicle cabin and achieves its perception function through the collaborative work of multiple sensors.
[0028] In some embodiments of the present invention, the occupant status information includes at least one of the following: occupant identification, gaze direction, attention state, and physical state.
[0029] Specifically, the multimodal state perception network 102 includes, but is not limited to, an in-cabin camera, a microphone array, a seat pressure sensor, and an account recognition module. The in-cabin camera can capture facial images of occupants for identification and gaze direction; the microphone array can be used to locate the source of voice and recognize voice commands; the seat pressure sensor can be used to determine the occupant's seating position and posture; and the account recognition module can confirm the occupant's identity through their logged-in account information. After the data collected by the above sensors is fused, occupant state information is generated, including occupant identification, attention level, fatigue state, and gaze direction.
[0030] Furthermore, the multimodal state-aware network 102 generates a real-time "state and capability vector" for each occupant based on their state information, which is used to determine whether to identify the occupant as the target occupant.
[0031] The dynamic permission mapping engine 103 is used to dynamically determine, based on real-time driving scenarios and occupant status information, to grant the target occupant permission to propose settings for at least one driving parameter dimension, so that the target occupant can initiate parameter setting proposals for the granted parameter dimensions.
[0032] Among them, real-time driving scenarios can be determined in real time by the scenario perception module by integrating GPS positioning, high-precision maps, radar and visual data, such as highway cruising scenarios, urban traffic jam following scenarios, parking lot location scenarios, etc.
[0033] Furthermore, the dynamic permission mapping engine 103 has a pre-set policy rule base. This rule base comprehensively judges the real-time driving scenario and occupant status information to determine the appropriate parameter dimensions to be opened in the current driving scenario and the target occupants qualified to make proposals. Based on the judgment results, the dynamic permission mapping engine 103 generates a corresponding list of proposeable permissions for each seat and sends it to the corresponding seat's interactive screen, enabling or disabling the relevant control components on that screen. Through this method, it is ensured that only occupants in suitable scenarios and in appropriate states can initiate setting proposals for specific driving parameter dimensions.
[0034] The central intelligent arbitrator 104 is used to receive at least one parameter setting proposal, arbitrate the parameter setting proposal, and generate a proposal to be confirmed.
[0035] Among them, parameter setting proposal refers to the structured data object generated after the crew operates on the open parameter dimensions on their interactive screen. It includes fields such as the proposal initiator information (such as crew ID), target parameter description, and context data when the proposal was generated.
[0036] Upon receiving a proposal, the central intelligent arbitrator 104 first performs conflict detection to determine if there are multiple pending proposals for the same parameter dimension. If no conflict exists, the proposal is directly designated as a proposal awaiting confirmation. If multiple competing proposals exist, a pre-defined arbitration strategy is used to select one as the proposal awaiting confirmation. The arbitration strategy can comprehensively consider factors such as occupant priority, occupant status information, occupant historical preference data, and the rationality assessment results of the proposal. For example, regarding the parameter dimension of following distance adjustment, when the driver and front passenger simultaneously propose different adjustment values, the central intelligent arbitrator 104 can select the driver's proposal based on the "driver's preference priority" principle, or select the proposal from the occupant whose gaze is always on the road based on the "proposer with better status priority" principle.
[0037] The proposal confirmation execution module 105 is used to present the proposal to be confirmed, and after receiving the confirmation instruction from the main driver, it sends the parameter setting instruction corresponding to the proposal to be confirmed to the vehicle control domain for execution.
[0038] As a preferred approach, the proposal confirmation execution module 105 is located in the driver's seat and can present the proposal to be confirmed through non-intrusive interaction channels such as a head-up display, digital instrument panel, or steering wheel light strip, avoiding distraction of the driver's attention. The driver can issue confirmation commands via dedicated physical buttons on the steering wheel, voice commands, or other means. Only upon receiving a clear confirmation command will the proposal to be confirmed be converted into an executable parameter setting command and sent to the vehicle's power control unit, chassis control unit, or driver assistance control unit for execution.
[0039] It should be noted that if the main driver rejects the proposal to be confirmed, or fails to respond within a certain period of time, the proposal confirmation execution module 105 will abandon the proposal to be confirmed, and the vehicle will maintain its original parameter settings.
[0040] Through the collaborative work of the aforementioned modules, this embodiment achieves safe and orderly collaborative adjustments of core driving parameters by multiple occupants. The control authority deconstruction module 101 decomposes driving control into multiple independent dimensions, achieving refined permission allocation. The cooperation of the multimodal state perception network 102 and the dynamic permission mapping engine 103 ensures that only occupants with appropriate states can obtain proposal permissions in suitable scenarios. The central intelligent arbitrator 104 handles multi-source proposal conflicts, ensuring the orderly fusion of multi-occupant inputs. The proposal confirmation and execution module 105 retains the final confirmation right of the primary driver, fundamentally guaranteeing driving safety. This embodiment, while ensuring the primary driver's final control, improves the efficiency and collaborative experience of human-machine interaction in multi-occupant scenarios, contributing to a higher level of human-machine co-driving.
[0041] In some embodiments of the present invention, the control control deconstruction module 101 deconstructs the driving control into at least a longitudinal control domain, a lateral control domain, and a system configuration domain, with each domain including at least one independent driving parameter dimension.
[0042] In some embodiments of the present invention, the driving parameter dimensions in the longitudinal control field include at least one of following distance, speed offset, power response mode, and energy recovery intensity. The driving parameter dimensions in the lateral control domain include at least one of navigation path preference, waypoint, lane selection preference, and lane change activism. The driving parameter dimension in the system configuration field includes at least one of the following: predictive driving mode, driving assistance sensitivity, and automation degradation strategy.
[0043] Specifically, the control deconstruction module 101 divides driving control into at least three control domains: longitudinal control domain, lateral control domain, and system configuration domain. Each domain contains multiple independently adjustable driving parameter dimensions, which constitute the basic unit for subsequent permission allocation. Through this hierarchical deconstruction approach, the originally holistic driving control is broken down into a set of discrete, describable micro-permissions, providing a structured foundation for refined permission allocation in multi-occupant scenarios.
[0044] Furthermore, the longitudinal control field involves parameters that directly affect the vehicle's forward and backward movement, mainly including following distance, speed deviation, power response mode, and energy recovery intensity.
[0045] The following distance refers to the time interval between the vehicle and the vehicle in front during adaptive cruise control, usually measured in seconds, such as 1.0 second, 1.5 seconds, or 2.0 seconds. Speed offset refers to a fixed offset value relative to the legal speed limit or set cruise speed, such as +0%, +5%, +10%, etc.; power response modes usually correspond to different accelerator pedal characteristic curves, which can be represented as "economy", "comfort", "sport" and other modes. The intensity of energy recovery affects the feeling of deceleration during gliding and braking, and is generally divided into weak, medium, and strong levels.
[0046] It should be noted that adjusting the above parameters will change the vehicle's longitudinal dynamics, directly affecting ride comfort and energy consumption.
[0047] Furthermore, the lateral control domain involves parameters that directly affect the vehicle's left and right movement, mainly including dimensions such as navigation path preference, waypoints, lane selection preference, and lane change activism.
[0048] Among them, navigation route preference refers to making a choice among multiple routes provided by the system, such as prioritizing routes with "shortest travel time", "avoiding tolls" or "avoiding highways". Route points refer to intermediate destinations that are inserted or deleted in a planned route, such as passing through a service area or tourist attraction; Lane selection preference refers to whether the system tends to drive in the left or right lane during cruise control; Lane change aggressiveness describes the system's strategy when overtaking or merging, which can be either conservative waiting or actively seeking opportunities.
[0049] These parameters primarily affect the vehicle's path planning and lateral movement strategies, and are closely related to navigation and driving style.
[0050] Furthermore, the system configuration domain involves setting parameters that do not directly output motion commands but indirectly affect the vehicle's dynamic behavior by changing controller strategies. These parameters mainly include dimensions such as predictive driving modes, driver assistance sensitivity, and automation degradation strategies.
[0051] Among them, the predictive driving mode refers to a comprehensive control strategy package preset for specific road conditions, such as "mountain road mode", "snow mode" and "congestion mode", which usually includes the linkage adjustment of longitudinal, lateral and chassis parameters. Driver assistance sensitivity refers to the trigger threshold of functions such as lane departure warning and forward collision warning, which can be set to different levels such as "early", "medium" and "late"; The automation degradation strategy defines whether to prioritize "requesting takeover" or "executing the least risk strategy" (such as pulling over) when the autonomous driving system is about to disengage.
[0052] Although these parameters do not directly control the vehicle, they have a significant impact on the driving experience and safety boundaries.
[0053] It is worth mentioning that, in order to provide an objective and clear basis for the logical judgment of the central intelligent arbitrator 104, this embodiment further defines parameter attributes for each driving parameter dimension. The parameter attributes include at least one or more of the following: value type, safety level, effective delay, and correlation.
[0054] Specifically, value types describe the form in which a parameter can take a value. Value types include continuous values, discrete enumerated values, and Boolean switches. Continuous values refer to parameters that can take any value within a continuous range, such as following distance which can be set to any value such as 1.2 seconds, 1.5 seconds, or 1.8 seconds; discrete enumerated values refer to parameters that select a value from a predefined finite set, such as the power response mode which can only be selected from "Economy," "Comfort," and "Sport"; Boolean switches refer to parameters that only have two states: "on" or "off," such as enabling or disabling the automatic high beam function. By clearly defining the value type, the central intelligent arbitrator 104 can accurately analyze whether the target value proposed by the occupant is legal and valid.
[0055] The safety level is used to assess the impact of parameter adjustments on driving safety. Safety levels are divided into three categories: high, medium, and low. High-safety-level parameters directly affect vehicle safety control, such as lane-changing aggression and automation degradation strategies; adjustments to these parameters require careful consideration. Medium-safety-level parameters primarily affect ride comfort and energy consumption, such as following distance and power response modes. Low-safety-level parameters only affect convenience and have minimal impact on safety, such as navigation route preferences and waypoint selection. This safety level classification provides an important reference for the permission opening strategy of the dynamic permission mapping engine 103; for example, stricter opening conditions can be set for high-safety-level parameters.
[0056] The term "effectiveness delay" describes the timing of parameter adjustments taking effect. Effectiveness delays include immediate effect, effect in the next driving cycle, and effect upon reaching a specific area. Immediate effect means the adjusted parameters are applied to vehicle control immediately; for example, switching power response modes typically requires immediate effect. Effect in the next driving cycle means the adjusted parameters take effect after the current trip ends and the vehicle is started again; for example, some personalized settings may be delayed until the next driving cycle. Effect upon reaching a specific area means the adjusted parameters only take effect when the vehicle reaches a specific geographical location; for example, adding waypoints requires the vehicle to reach the corresponding area before it can be triggered. By clearly defining the effectiveness delay characteristics, the central intelligent arbitrator 104 can rationally plan the timing of parameter command issuance.
[0057] Among these, correlation describes the coupling relationship between parameter dimensions and other parameters. Correlation can manifest as parameter linkage or mutual exclusion. For example, when an occupant suggests switching the power response mode to "Sport Mode," the system can automatically adjust the following distance to a shorter value and the lane change responsiveness to a higher level to match the overall driving style. By defining the correlation between parameters, the central intelligent arbitrator 104 can simultaneously generate adjustment suggestions or prompts for related parameters when confirming a parameter suggestion, improving the overall coordination of parameter settings.
[0058] Through the definition of the above parameter attributes, the control authority deconstruction module 101 not only realizes the dimensional deconstruction of driving control authority, but also assigns metadata that can be used for logical judgment to each parameter dimension. This attribute information provides a quantitative basis for the scenario adaptation rule formulation of the dynamic permission mapping engine 103 and the conflict arbitration decision of the central intelligent arbitrator 104, enabling the system to adopt differentiated processing strategies according to the characteristics of different parameters, further improving the accuracy and safety of multi-occupant collaborative control.
[0059] This embodiment, through the aforementioned multi-dimensional deconstruction, transforms complex driving control rights into a set of clearly structured and attributed parameter units. Each parameter unit has an independent value type (such as continuous value, discrete enumerated value, Boolean value), security level, and effective rules. This provides a quantifiable data foundation for the rule judgment of the subsequent dynamic permission mapping engine 103 and the conflict handling of the central intelligent arbitrator 104, realizing the transformation of driving control rights from "holistic" to "component-based." This allows for the precise opening of the right to propose corresponding parameters according to specific needs during multi-occupant collaboration, avoiding the security risks brought about by the overall permission transfer and improving the flexibility and targeting of the interaction.
[0060] In some embodiments of the present invention, the dynamic permission mapping engine 103 has a preset policy rule base, which includes at least: Safety baseline rules are used to absolutely prohibit or lock the granting of setting proposal permissions to occupants based on vehicle speed, vehicle transient operating status, or vehicle malfunction status. Scene adaptation rules are used to establish a mapping relationship between different driving scenarios and the dimensions of available parameters; Crew role and status rules are used to determine the target crew member based on the crew member's identity authentication result and crew member status information, and to perform personalized allocation of the target crew member with open parameter dimensions.
[0061] Specifically, the dynamic permission mapping engine 103 has a pre-set policy rule base, which includes safety baseline rules, scene adaptation rules, and occupant filtering rules. Through the synergistic effect of these three types of rules, the dynamic permission mapping engine 103 can generate a corresponding list of proposed permissions for each seat and send it to the corresponding seat's interactive screen to enable or disable the relevant control components on that screen.
[0062] Furthermore, safety baseline rules are used to make judgments based on the vehicle's operating status to absolutely prohibit or lock the granting of setting proposal permissions to occupants.
[0063] In this embodiment, the vehicle operating status includes, but is not limited to, vehicle speed, vehicle transient operating status, and vehicle fault status.
[0064] Among them, the transient operating state of a vehicle refers to the dynamic maneuvering process that the vehicle is performing, such as automatic lane changing, automatic parking, and emergency braking.
[0065] Specifically, when the vehicle speed exceeds a preset threshold (such as 80 km / h), the safety baseline rules can prohibit operations such as modifying navigation route preferences or adding waypoints to prevent safety accidents caused by driver distraction or passenger interference. When the vehicle is performing an automatic lane change operation, the safety baseline rules can lock the parameter proposal rights of all non-driver occupants to ensure centralized control during transient operations. When the vehicle fault diagnosis system alarms or the confidence level of the autonomous driving system is lower than a preset threshold, the safety baseline rules fully lock the proposal rights of non-driver occupants to ensure that control is not interfered with when the vehicle is in an abnormal state.
[0066] It should be noted that the safety baseline rule has the highest priority, and its judgment result can cover the output of other rules, fundamentally ensuring the bottom line of driving safety.
[0067] Furthermore, scenario adaptation rules are used to establish mapping relationships between different driving scenarios and available parameter dimensions. Driving scenarios can be determined in real time by the scenario perception module by fusing GPS data, high-precision map information, visual sensor data, and radar data.
[0068] As an example, driving scenarios may include, but are not limited to, highway cruising scenarios, urban traffic jam following scenarios, parking lot location scenarios, and winding mountain road scenarios. As shown in Table 1, Table 1 is a mapping table between driving scenarios and available parameter dimensions.
[0069] Table 1 Mapping between driving scenarios and available parameter dimensions
[0070] In highway cruising scenarios, due to relatively stable road conditions and high vehicle speeds, passengers are more sensitive to comfort. The scenario adaptation rules can open up parameters such as following distance, speed deviation, power response mode, and lane selection preference. In urban traffic congestion scenarios, due to low vehicle speeds and frequent acceleration and deceleration, the safety tolerance is relatively low. The scenario adaptation rules can allow for more directional adjustment of the following distance and the economy or comfort mode of the power response mode. In parking lot location scenarios, due to extremely low vehicle speeds and strong spatial exploration requirements, scenario adaptation rules can open up parameter dimensions such as destination fine-tuning, automatic parking mode selection, and panoramic view control. In scenarios involving winding mountain roads, the scenario adaptation rules can enable a predictive driving mode, such as suggesting switching to mountain road mode to optimize vehicle dynamics in advance.
[0071] Furthermore, the crew role and status rules are used to determine the target crew based on the crew's identity authentication results and crew status information, and to perform personalized allocation of open parameter dimensions to the target crew.
[0072] Identity authentication can be achieved through biometric technology (such as facial recognition, fingerprint recognition, and voiceprint recognition) or account login, and is used to distinguish between authenticated passengers (such as permanent family members registered in the system) and unauthenticated passengers (such as temporary visitors).
[0073] For certified occupants, the occupant role and state rules grant them the right to propose all open dimensions in the current scenario; for uncertified occupants, they are only granted the right to propose lower security level parameters, such as navigation path preferences. Simultaneously, the occupant role and state rules also perform real-time filtering based on occupant state information.
[0074] The occupant status information is generated by the multimodal state perception network 102, and includes at least the occupant's attention concentration, gaze direction, fatigue state, and interaction intention. When the system detects that an occupant's gaze has left the road for more than a preset time (e.g., 3 seconds) or that their fatigue level is high, the occupant role and status rules can temporarily disable all their suggestion rights. When the system detects that an occupant's gesture points to the central control screen and is accompanied by a related voice command (e.g., "Adjust the following distance further away"), the occupant role and status rules can proactively highlight the relevant control items on their nearest screen to guide the interaction.
[0075] Furthermore, the permission mapping process is as follows: First, real-time driving scenario tags, vehicle status (vehicle speed, system mode), and occupant status information are collected.
[0076] Secondly, based on the safety baseline rules, absolutely prohibited dimensions are excluded; based on the scenario adaptation rules, a set of open dimensions for the scenario is generated; based on the occupant role and status rules, the set is filtered and allocated to generate a personalized list of suggested permissions for each seat.
[0077] Finally, the "permission list" for each seat is sent to the corresponding seat's interactive screen. The screen UI dynamically enables or disables relevant control components (such as sliders and buttons turning gray) based on this list. Only passengers in that seat can initiate a valid proposal based on the parameters in the list.
[0078] Through the synergistic effect of the three rules mentioned above, the dynamic permission mapping engine 103 achieves a refined allocation from overall control to micro-permissions. The safety baseline rule defines inviolable safety boundaries, the scenario adaptation rule provides a basic set of permissions for scenario-adaptive behavior, and the occupant role and state rules then perform personalized filtering and allocation based on this. The superimposed output of these three layers of rules generates a differentiated list of proposeable permissions for each seat, ensuring that only when the occupant is in a suitable scenario and in a suitable state can a setting proposal be made for specific driving parameter dimensions. This enhances the flexibility and relevance of multi-occupant interactions while ensuring safety.
[0079] In some embodiments of the present invention, the multi-occupant cooperative intelligent driving control system further includes: The safety recovery module is configured to restore all driving parameters to the preset safety configuration in response to a recovery command triggered by the driver's seat.
[0080] Specifically, the safety recovery module is connected to a physical interaction device in the driver's seat, which can be located on the steering wheel, center console, or other locations easily accessible to the driver.
[0081] As a preferred approach, this physical interaction device employs a dedicated hardware button with the highest hardware priority, ensuring it can be recognized and responded to in any system state. When the primary driver triggers the recovery command, the safety recovery module immediately takes over control, unconditionally restoring all driving parameters to the preset safety configuration.
[0082] It should be noted that the preset safety configuration refers to the primary driver's personal default configuration, which is isolated from the parameter settings effective in collaborative mode. In this embodiment, the system uses a software sandbox mechanism to store the collaboratively effective parameter configuration and the primary driver's personal default configuration in different storage areas or distinguish them using different identifiers. All parameter settings from non-driver occupants and confirmed by the primary driver are marked as "collaborative mode" configurations; while the primary driver's personal default configuration is saved independently as a "safety baseline". When the safety recovery module receives a recovery command, the system directly reads the personal default configuration and overwrites all current driving parameters, achieving rapid configuration switching.
[0083] As an example, after taking over the vehicle, the primary driver may find that parameters such as following distance and power response mode have been adjusted by passengers to values that do not match their personal driving habits. In this case, the primary driver simply needs to press the preset "one-button reset" button on the steering wheel. The safety recovery module will respond to this button's trigger, restoring all driving parameters to their default settings, including following distance to 1.5 seconds, power response mode to "Comfort" mode, and navigation route preference to "shortest time." The entire recovery process is completed instantly, without requiring the primary driver to manually search for and adjust each parameter.
[0084] Furthermore, the safety recovery module has a higher priority than all other control modules. When a recovery command is triggered, the safety recovery module can directly send a parameter reset command to the vehicle control domain, and simultaneously send a reset signal to the dynamic permission mapping engine 103 and the central intelligent arbitrator 104, clearing all pending proposals and temporary permission assignments. This design ensures that the primary driver can unconditionally and immediately reclaim full control at any time and in any scenario, returning to a fully autonomous driving environment.
[0085] In this way, this embodiment provides a quick and direct intervention method when the main driver is dissatisfied with the cooperation results or perceives potential risks, thereby enhancing the robustness and safety of the system.
[0086] Figure 2 This is a schematic flowchart of an embodiment of the multi-occupant cooperative intelligent driving control method provided by the present invention, as shown below. Figure 2 As shown, the multi-occupant cooperative intelligent driving control method includes: S201, Deconstruct vehicle driving control into multiple independent driving parameter dimensions; S202. Real-time perception of the identity and status of each passenger in the vehicle, generating corresponding passenger status information; S203. Based on real-time driving scenarios and occupant status information, dynamically determine the permission to grant the target occupant setting proposals for at least one driving parameter dimension. S204. Receive parameter setting proposals from one or more crew members for open parameter dimensions, arbitrate the parameter setting proposals, and generate proposals to be confirmed. S205. After receiving the confirmation instruction from the main driver for the proposal to be confirmed, the parameter setting instruction corresponding to the proposal to be confirmed is sent to the vehicle control domain for execution.
[0087] In step S201, the vehicle's control authority is first deconstructed dimensionally, breaking down the originally holistic driving control into a set of independently describable and operable basic units. These basic units are the driving parameter dimensions, each corresponding to an adjustable driving-related parameter.
[0088] As an example, the deconstructed driving parameter dimensions may include following distance, speed offset, and power response mode in the longitudinal control dimension, navigation path preference and lane selection preference in the lateral control dimension, and predictive driving mode in the system configuration dimension.
[0089] It should be noted that each parameter dimension after deconstruction has an independent attribute definition, including value type, security level, effective delay, and correlation, etc. This attribute information will be used for logical judgment in subsequent steps.
[0090] In step S202, various types of data related to the occupants are continuously collected through a multimodal sensor network deployed in the cockpit.
[0091] After fusion processing, the aforementioned multi-source data generates an occupant state vector containing information such as occupant identification, attention level, fatigue status, gaze direction, and interaction intent. This state vector reflects the real-time state of each occupant at the current moment, providing a quantitative basis for determining whether an occupant is suitable to initiate parameter setting suggestions in subsequent steps.
[0092] In step S203, the real-time driving scenario can be determined in real time by the scenario perception module by fusing GPS data, high-precision maps, visual sensor information, and radar data. Examples include highway cruising scenarios, urban congestion following scenarios, parking lot location scenarios, or winding mountain road scenarios. Based on a preset policy rule base, the driving scenario and occupant status information are comprehensively judged to determine the set of parameter dimensions suitable for opening in the current scenario, and target occupants qualified to make proposals are selected from this set.
[0093] In step S204, the parameter setting proposal refers to the structured data object generated after the occupant operates on the open parameter dimensions on their interactive screen. It includes fields such as the proposal initiator information, the target parameter description, and the context data when the proposal was generated.
[0094] In a specific example, an occupant is identified as the target occupant and performs an action on the available controls on their screen (such as dragging a slider, clicking a button, or giving a voice command). The cockpit domain controller captures this raw interaction event and attaches metadata: seat ID, interaction method (touch / voice), timestamp, and generates a parameter setting proposal.
[0095] Upon receiving a parameter setting proposal, a conflict detection process is first performed to determine if there are multiple pending proposals for the same parameter dimension. If no conflict exists, the proposal is directly designated as a proposal awaiting confirmation. If multiple competing proposals exist, a pre-defined arbitration strategy is used to select one proposal from among them as the proposal awaiting confirmation.
[0096] As an example, when the front passenger and rear passengers simultaneously suggest different adjustment values for the following distance, this method can select the suggestion that is closer to the driver's historical preferences based on the "driver's preference priority" principle, or select the suggestion from the passenger whose eyes are always on the road based on the "suggestor's better condition priority" principle. Through arbitration, multiple inputs are orderly merged into a single suggestion to be confirmed, avoiding control conflicts caused by simultaneous operation by multiple passengers.
[0097] In step S205, the proposal to be confirmed is presented through a dedicated interaction channel in the driver's seat. For example, the proposal content is displayed in a non-intrusive way, such as through a head-up display, digital instrument panel, or steering wheel light strip, to avoid distracting the driver's attention.
[0098] The driver can issue confirmation commands via dedicated physical buttons on the steering wheel or voice commands.
[0099] It should be understood that the pending confirmation proposal will only be converted into an executable parameter setting instruction and sent to the vehicle's power control unit, chassis control unit, or driver assistance control unit for execution upon receiving an explicit confirmation instruction.
[0100] It should be noted that if the primary driver rejects the proposal to be confirmed, or fails to respond within a certain period of time, the method abandons the proposal, and the vehicle maintains its original parameter settings. By retaining the primary driver's final confirmation right, driving safety is fundamentally guaranteed.
[0101] This embodiment achieves refined permission allocation through parameter deconstruction; ensures the relevance and real-time nature of permission allocation through occupant status awareness and dynamic permission mapping; achieves orderly fusion of multi-occupant inputs through arbitration processing; and retains the final anchor point of safety responsibility through driver confirmation. Thus, it improves human-computer interaction efficiency and collaborative experience in multi-occupant scenarios while ensuring the driver's ultimate control.
[0102] In some embodiments of the present invention, such as Figure 3 As shown, step S204 arbitrates the parameter setting proposal and generates a proposal to be confirmed, including: S301. Detect whether there are multiple competing proposals for the same parameter dimension, and obtain the conflict detection result; S302. If the conflict detection result indicates that a conflict exists, then according to the preset arbitration strategy, one of the multiple competing proposals is determined as the proposal to be confirmed. The preset arbitration strategy includes at least one of the following factors: crew priority, crew status information, crew historical preference data, and the rationality assessment result of the proposal. S303. If the conflict detection result is that there is no conflict, the received proposal will be directly used as a proposal to be confirmed.
[0103] Specifically, the process of arbitrating the parameter setting proposal and generating a proposal to be confirmed in step S204 can be further implemented in the manner of steps S301 to S303 to achieve orderly processing of proposals initiated by multiple crew members at the same time.
[0104] In step S301, after receiving one or more parameter setting proposals, the central intelligent arbitrator first parses the proposals and extracts the parameter dimension information for each proposal. Then, the central intelligent arbitrator groups the proposals according to the parameter dimensions and counts the number of proposals under each parameter dimension. If the number of proposals corresponding to a certain parameter dimension is greater than 1, it is determined that there are multiple competing proposals for that parameter dimension; if the number of proposals is equal to 1, it is determined that there is no conflict.
[0105] Through this detection method, the system can quickly identify conflict scenarios that require further arbitration, as well as conflict-free scenarios that can be passed directly.
[0106] As an example, when the front passenger proposes to adjust the following distance to 1.8 seconds, and the rear passenger proposes to adjust the following distance to 2.2 seconds, the central intelligent arbitrator 104 detects that both proposals are related to the parameter dimension of "following distance", and therefore determines that there are competing proposals.
[0107] In step S302, the preset arbitration strategy can be judged based on a combination of factors, including but not limited to at least one of the following: passenger priority, passenger status information, passenger historical preference data, and the rationality assessment results of the proposal.
[0108] Among them, occupant priority refers to the preset priority level based on the occupant's identity or seat position. For example, the driver's seat occupant can be set to the highest priority, followed by the front passenger seat occupant, and then the rear passengers; or it can be differentiated according to account type, such as the owner's account having a higher priority than the visitor's account. When multiple competing proposals are initiated, the central intelligent arbitrator can prioritize the proposal initiated by the occupant with the higher priority.
[0109] The occupant state information is derived from occupant state vectors generated by a multimodal state perception network, including occupant attention level, gaze direction, and fatigue level. The central intelligent arbitrator can evaluate the real-time state of each proposal initiator and prioritize proposals from occupants in better states (such as focused attention and gaze on the road), ensuring that the proposers are in a suitable state to participate in driving interactions.
[0110] Among these, occupant historical preference data refers to the system's record of the occupant's settings for various driving parameters during past trips. The central intelligent arbitrator can match the current proposal with the occupant's historical preferences, selecting the proposal that is more consistent with historical preferences to enhance the personalized experience. For example, if an occupant has consistently preferred to set a longer following distance in previous trips, then when they propose adjusting the following distance, the proposal is likely to better align with their long-term preferences.
[0111] The rationality assessment result of the proposal refers to the quantitative evaluation of the reasonableness of the proposed target value in the current scenario. The central intelligent arbitrator can evaluate the rationality score of each proposal based on factors such as scenario safety boundaries and vehicle dynamics constraints, and select the proposal with higher rationality. For example, in a high-speed cruise scenario, proposing a following distance of too small a value (such as 0.8 seconds) may be evaluated as having low rationality, while proposing 1.5 seconds is considered to have high rationality.
[0112] The central intelligent arbitrator can use one of the aforementioned factors for arbitration alone, or it can weight and fuse multiple factors to calculate a comprehensive score for each proposal, ultimately selecting the proposal with the highest score as the one to be confirmed. Through this multi-factor arbitration mechanism, the system can make the optimal choice that balances safety, personalization, and rationality when multiple passengers simultaneously submit adjustment requests.
[0113] In step S303, when there is only one proposal for a certain parameter dimension, no complex arbitration process is required. The central intelligent arbitrator can directly treat the proposal as a proposal to be confirmed and proceed to the subsequent confirmation process. This approach simplifies the processing flow in conflict-free scenarios and improves system response efficiency.
[0114] Through the above steps, this embodiment enables proposals to quickly proceed to the confirmation stage in conflict-free scenarios. In conflict scenarios, the system can intelligently select the optimal proposal based on multi-dimensional factors, avoiding disorderly competition from multiple input sources. This arbitration mechanism not only ensures the sense of participation in multi-occupant collaboration but also ensures that the proposals submitted to the primary driver for confirmation have high rationality and acceptability, thereby improving overall collaboration efficiency and user experience.
[0115] The present invention also provides a vehicle. The vehicle includes the aforementioned multi-occupant cooperative intelligent driving control system and other necessary components required for the vehicle. However, it should be understood that it is not required to implement all of the shown components, and more or fewer components may be implemented instead.
[0116] Accordingly, this application also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions of the multi-occupant cooperative intelligent driving control method provided in the above-described method embodiments.
[0117] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.
[0118] The above provides a detailed description of the multi-occupant cooperative intelligent driving control system, method, vehicle, and storage medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A multi-occupant cooperative intelligent driving control system, characterized in that, include: The control control deconstruction module is used to deconstruct vehicle driving control into multiple independent driving parameter dimensions; A multimodal state perception network is used to perceive the identity and state of each occupant in the vehicle in real time and generate corresponding occupant state information. The dynamic permission mapping engine is used to dynamically determine, based on the real-time driving scenario and the occupant status information, to grant the target occupant permission to propose setting at least one of the driving parameter dimensions, so that the target occupant can initiate parameter setting proposals for the granted parameter dimensions. A central intelligent arbitrator is used to receive at least one of the parameter setting proposals, arbitrate the parameter setting proposals, and generate a proposal to be confirmed. The proposal confirmation and execution module is used to present the proposal to be confirmed, and after receiving the confirmation instruction from the main driver, to send the parameter setting instruction corresponding to the proposal to be confirmed to the vehicle control domain for execution.
2. The multi-occupant cooperative intelligent driving control system according to claim 1, characterized in that, The control control deconstruction module deconstructs driving control into at least a longitudinal control domain, a lateral control domain, and a system configuration domain, with each domain including at least one of the driving parameter dimensions.
3. The multi-occupant cooperative intelligent driving control system according to claim 2, characterized in that, The driving parameter dimensions in the longitudinal control domain include at least one of following distance, speed deviation, power response mode, and energy recovery intensity; The driving parameter dimensions in the lateral control domain include at least one of navigation path preference, waypoint, lane selection preference, and lane change activism. The driving parameter dimension in the system configuration domain includes at least one of predictive driving mode, driving assistance sensitivity, and automation degradation strategy.
4. The multi-occupant cooperative intelligent driving control system according to claim 1, characterized in that, The dynamic permission mapping engine has a pre-defined policy rule base, which includes at least the following: Safety baseline rules are used to absolutely prohibit or lock the granting of setting proposal permissions to occupants based on vehicle speed, vehicle transient operating status, or vehicle malfunction status. Scene adaptation rules are used to establish a mapping relationship between different driving scenarios and the dimensions of available parameters; Crew role and status rules are used to determine target crew members based on their identity authentication results and crew status information, and to perform personalized allocation of open parameter dimensions to target crew members.
5. The multi-occupant cooperative intelligent driving control system according to claim 1, characterized in that, The occupant status information includes at least one of the following: occupant identification, gaze direction, attention state, and physical state.
6. The multi-occupant cooperative intelligent driving control system according to claim 1, characterized in that, include: The safety recovery module is configured to restore all driving parameters to the preset safety configuration in response to a recovery command triggered by the driver's seat.
7. A method for intelligent driving control oriented towards multi-occupant collaboration, characterized in that, include: Deconstruct vehicle driving control into multiple independent driving parameter dimensions; Real-time perception of the identity and status of each passenger in the vehicle, generating corresponding passenger status information; Based on the real-time driving scenario and the occupant status information, dynamically determine the permission to grant the target occupant setting suggestion for at least one of the driving parameter dimensions; Receive parameter setting proposals from one or more occupants for open parameter dimensions, arbitrate the parameter setting proposals, and generate proposals to be confirmed; Upon receiving the confirmation instruction from the primary driver for the proposed confirmation, the parameter setting instruction corresponding to the proposed confirmation is sent to the vehicle control domain for execution.
8. The method according to claim 7, characterized in that, The arbitration of the parameter setting proposal to generate a proposal to be confirmed includes: The system detects whether there are multiple competing proposals for the same parameter dimension, and obtains the conflict detection results. If the conflict detection result indicates that a conflict exists, then according to a preset arbitration strategy, one of the multiple competing proposals is determined as the proposal to be confirmed. The preset arbitration strategy includes at least one of the following factors: occupant priority, occupant status information, occupant historical preference data, and the rationality assessment result of the proposal. If the conflict detection result is that there is no conflict, the received proposal will be directly used as the proposal to be confirmed.
9. A vehicle, characterized in that, The system includes an intelligent driving control system for multi-occupant collaboration as described in any one of claims 1 to 6.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the intelligent driving control method for multi-occupant cooperation as described in claim 7 or 8.