Method and apparatus for determining vehicle control parameters

Abstract

Embodiments of the present disclosure provide a method and apparatus for determining a vehicle control parameter. In the method for determining a vehicle control parameter, vehicle control indication information input by a user is obtained; a vehicle control intention and a corresponding intention strength are determined according to the vehicle control indication information, the intention strength being used to represent a desired strength degree of the vehicle control intention; safety boundary information corresponding to the vehicle control intention is obtained, the safety boundary information being used to indicate a boundary of a safety constraint that should be observed by the vehicle; and a target parameter value of the vehicle control parameter associated with the vehicle control intention is determined according to at least the vehicle control intention, the intention strength, and the safety boundary information.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle control technology, and more specifically, to methods and apparatus for determining vehicle control parameters. Background Technology

[0002] Currently, the driving control functions of autonomous vehicles mainly rely on physical buttons or touch interfaces for parameter settings, such as setting cruise speed and following distance via steering wheel buttons or the central control screen. This method of parameter determination requires the driver to briefly deviate from the road and take their hands off the steering wheel, which is cumbersome and poses safety risks during driving.

[0003] With the development of voice interaction technology, voice control has been widely applied to non-driving functions such as in-vehicle air conditioning adjustment and music playback. However, since driving control parameters directly determine the longitudinal and lateral movement of the vehicle, the requirements for accuracy and safety are extremely high. Existing voice control solutions have not yet been effectively applied to driving control functions during driving due to problems such as misrecognition and semantic ambiguity. Although there are currently applications of voice-controlled automatic parking, parking scenarios involve low vehicle speeds and relatively controllable environments, making it difficult to directly apply voice control technology used in low-speed scenarios to normal medium- and high-speed driving scenarios. Therefore, a method is needed that can efficiently and safely determine vehicle control parameters based on the user's natural input during vehicle operation, thereby improving human-machine interaction efficiency and vehicle driving safety. Summary of the Invention

[0004] To address the aforementioned issues, this disclosure proposes a method and apparatus for determining vehicle control parameters, providing a way to efficiently and safely determine vehicle control parameters based on natural user input during vehicle operation, thereby improving human-computer interaction efficiency and vehicle driving safety.

[0005] According to one aspect of this disclosure, a method for determining vehicle control parameters is provided, comprising: acquiring vehicle control instruction information input by a user; determining a vehicle control intention and a corresponding intention intensity based on the vehicle control instruction information, the intention intensity being used to characterize the expected strength of the vehicle control intention; acquiring safety boundary information corresponding to the vehicle control intention, the safety boundary information being used to indicate the boundary of safety constraints that the vehicle should comply with; and determining target parameter values ​​of vehicle control parameters associated with the vehicle control intention based at least on the vehicle control intention, the intention intensity, and the safety boundary information.

[0006] According to one aspect of this disclosure, a method for controlling a vehicle is provided, comprising: acquiring a target parameter value of a vehicle control parameter associated with a vehicle control intention, determined according to the method described above for determining vehicle control parameters; outputting a vehicle control prompt message corresponding to the vehicle control intention based on the target parameter value; and controlling the vehicle according to the target parameter value in response to not receiving a cancellation operation message corresponding to the vehicle control prompt message within a preset time window.

[0007] According to one aspect of this disclosure, an apparatus for determining vehicle control parameters is provided, comprising: a control information acquisition module configured to acquire vehicle control instruction information input by a user; an intent determination module configured to determine a vehicle control intent and a corresponding intent intensity based on the vehicle control instruction information, the intent intensity being used to characterize the expected strength of the vehicle control intent; a safety information acquisition module configured to acquire safety boundary information corresponding to the vehicle control intent, the safety boundary information being used to indicate the boundary of safety constraints that the vehicle should comply with; and a parameter determination module configured to determine target parameter values ​​of vehicle control parameters associated with the vehicle control intent based at least on the vehicle control intent, the intent intensity, and the safety boundary information.

[0008] According to one aspect of this disclosure, an apparatus for controlling a vehicle is provided, comprising: an apparatus for determining vehicle control parameters as described above; an information output module configured to output vehicle control prompt information corresponding to the vehicle control intention based on the target parameter value; and a vehicle control module configured to control the vehicle according to the target parameter value in response to not receiving cancellation operation information corresponding to the vehicle control prompt information within a preset time window or receiving confirmation operation information corresponding to the vehicle control prompt information within a preset time window.

[0009] According to one aspect of this disclosure, a processing apparatus is provided, comprising: a processor; and a memory storing instructions that, when executed by the processor, cause the processor to perform operations as described above in the method for determining vehicle control parameters and / or as described above in the method for controlling a vehicle.

[0010] According to one aspect of this disclosure, a driving assistance system is provided, comprising: a sensor configured to acquire environmental perception information; and a processing means as described above, wherein the safety boundary information is determined at least based on the environmental perception information.

[0011] According to one aspect of this disclosure, a machine-readable storage medium is provided that stores executable instructions, which, when executed, cause a processor to perform operations as described above in the method for determining vehicle control parameters and / or the method for controlling a vehicle as described above.

[0012] According to one aspect of this disclosure, a computer program product is provided, comprising executable instructions that, when executed, cause a processor to perform operations as described above in the method for determining vehicle control parameters and / or the method for controlling a vehicle as described above.

[0013] Embodiments of various aspects of this disclosure provide a way to efficiently and safely determine vehicle control parameters based on natural user input during vehicle operation. Other advantages of embodiments of this disclosure will be described below. Attached Figure Description

[0014] A further understanding of the nature and advantages of this application can be achieved by referring to the accompanying drawings. In the drawings, similar components or features may have the same reference numerals.

[0015] Figure 1 An exemplary application architecture for determining vehicle control parameters is shown according to one embodiment.

[0016] Figure 2 An exemplary process for determining vehicle control parameters according to an embodiment is shown.

[0017] Figure 3 Another exemplary process for determining vehicle control parameters according to an embodiment is shown.

[0018] Figure 4 An exemplary target parameter value determination process is shown in a method for determining vehicle control parameters according to an embodiment.

[0019] Figure 5 This illustrates yet another exemplary target parameter value determination process in a method for determining vehicle control parameters according to an embodiment.

[0020] Figure 6 Another exemplary target parameter value determination process is shown in the method for determining vehicle control parameters according to an embodiment.

[0021] Figure 7 An exemplary intent and intent intensity determination process are illustrated in a method for determining vehicle control parameters according to an embodiment.

[0022] Figure 8Another exemplary process for determining vehicle control parameters according to an embodiment is shown.

[0023] Figure 9 An exemplary flowchart of a method for determining vehicle control parameters according to an embodiment is shown.

[0024] Figure 10 An exemplary flowchart of a method for controlling a vehicle according to an embodiment is shown.

[0025] Figure 11 A schematic block diagram of a device for determining vehicle control parameters according to an embodiment is shown.

[0026] Figure 12 A schematic block diagram of a device for controlling a vehicle according to an embodiment is shown.

[0027] Figure 13 A schematic block diagram of a processing apparatus according to an embodiment is shown.

[0028] Figure 14 A schematic block diagram of a driving assistance system according to an embodiment is shown. Detailed Implementation

[0029] The subject matter described herein will be discussed below with reference to exemplary embodiments. It should be understood that these embodiments are discussed merely to enable those skilled in the art to better understand and implement the subject matter described herein, and are not intended to limit the scope, applicability, or examples set forth in the claims. The function and arrangement of the elements discussed may be changed without departing from the scope of the embodiments disclosed herein. Various processes or components may be omitted, substituted, or added as needed in the various examples. Furthermore, features described in some examples may be combined in other examples.

[0030] As used herein, the term "comprising" and its variations are open terms meaning "including but not limited to". The term "based on" means "at least partially based on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other definitions, whether explicit or implicit, may be included below. Unless explicitly indicated by the context, the definition of a term shall remain consistent throughout the specification.

[0031] The technical scope of the term "autonomous driving" can include, but is not limited to, all types of autonomous driving technologies that rely on a perception-decision-control closed loop to achieve autonomous vehicle control, such as driver assistance (Level 1 or Level 2, such as adaptive cruise control, lane keeping assist, etc.), intelligent driving (Level 3 and above, such as conditional autonomous driving, etc.), and fully driverless driving (Level 4 or Level 5).

[0032] The flowcharts used in this disclosure illustrate operations implemented according to some embodiments of this disclosure. It should be clearly understood that the operations in the flowcharts may not be implemented sequentially. Instead, the operations may be implemented in reverse order or simultaneously. Furthermore, one or more additional operations may be added to the flowcharts. One or more operations may be removed from the flowcharts.

[0033] Exemplary methods and apparatus for determining vehicle control parameters according to embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings.

[0034] Figure 1 An exemplary application architecture 100 for determining vehicle control parameters is shown according to one embodiment.

[0035] like Figure 1 As shown, vehicle 110 may include microphone 111, sensor 112, electronic control unit (ECU) 113, advanced driver assistance systems (ADAS) 114, etc. In some examples, microphone 111 can acquire vehicle control instruction information input by the user. In some examples, ECU 113 can determine the vehicle control intention and the corresponding intention strength based on the vehicle control instruction information. In some examples, ECU 113 can acquire safety boundary information corresponding to the vehicle control intention. In one example, the safety boundary information can be determined based on sensor 112 (such as lidar, ultrasonic radar, vision sensor, wheel speed sensor, etc.). In some examples, ECU 113 can determine the target parameter value of the vehicle control parameters associated with the vehicle control intention, at least based on the vehicle control intention, intention strength, and safety boundary information.

[0036] The method for determining vehicle control parameters according to embodiments of this disclosure can be executed by an electronic control unit 113. It is understood that... Figure 1 The application scenarios shown only illustrate components relevant to the embodiments of this disclosure. In actual implementations, these application scenarios may include more or fewer components. For example, they may also include touch sensors or in-vehicle cameras for acquiring vehicle control instruction information. These examples are for illustrative purposes only. Figure 1The illustration shows a specific number of microphones 111, sensors 112, electronic control units 113, etc., but in actual implementations, the number of these components is not fixed. It is also understood that the embodiments disclosed herein are not limited to the exemplary scenarios described above, but can be applied to any variations of these exemplary scenarios and any other applicable scenarios.

[0037] Figure 2 An exemplary process 200 for determining vehicle control parameters according to an embodiment is shown.

[0038] like Figure 2 As shown, in step S210, the vehicle control instruction information 201 input by the user can be obtained. In this embodiment, the vehicle control instruction information 201 can be content information input by the user through any interactive method, used to express their expectations for vehicle driving control. In some examples, the vehicle control instruction information 201 can be the original input signal, or it can be structured data obtained after recognition processing. In some examples, the vehicle control instruction information 201 can be used to reflect the user's expectations for longitudinal vehicle control (e.g., acceleration, deceleration, cruise speed setting, following distance adjustment, etc.) and / or lateral control (e.g., steering, lane changing, overtaking, parking, etc.).

[0039] In some examples, vehicle control instruction information 201 can be voice control information, such as raw voice signals, natural language text obtained through speech recognition, or natural language text obtained through speech recognition combined with acoustic features extracted from the voice signals. In one example, a user can say things like "faster," "accelerate," "change cruise speed to 120km / h," "slow down," "too close," "keep your distance," "hurry up and catch up," "get a little closer," "find an opportunity to change lanes to the right," "hurry up and overtake it," or "change lanes to the left," etc. The above voice signals can be collected by a microphone and used as vehicle control instruction information 201; or, the text obtained by recognizing the above voice signals can be directly obtained from the speech recognition module, and the obtained text can be used as vehicle control instruction information 201.

[0040] In some examples, vehicle control instruction information 201 can be gesture control information, such as including but not limited to at least one of the following: hand posture, motion trajectory, and gesture type captured by a camera. For example, the gesture trajectory of a user waving their hand forward can be used as vehicle control instruction information 201. In some examples, vehicle control instruction information 201 can be touch operation information, such as including but not limited to at least one of the following: click, swipe, and long press operations on a touchscreen, as well as the pressure value, swipe speed, and touch coordinates corresponding to the above operations. For example, the swipe operation of a user quickly sliding the speed acceleration bar on a touchscreen and the corresponding swipe speed value can be used as vehicle control instruction information 201. In some examples, vehicle control instruction information 201 can be button control information, such as including but not limited to at least one of the following: the number of times a physical button or virtual button is pressed, the pressing frequency, and the pressing duration. For example, the button sequence of a user quickly clicking the "accelerate" button three times consecutively and the click interval time sequence can be used as vehicle control instruction information 201. In some examples, user gestures can be captured by in-vehicle visual sensing devices (such as cameras), user operations can be converted into vehicle control instruction information 201 by the touch sensor of in-vehicle touch display devices (such as central touch screens), and user voice or touch commands can be obtained as vehicle control instruction information 201 by using the communication connection between a mobile terminal (such as a smartphone APP) and the vehicle (such as Bluetooth or V2X communication).

[0041] In step S220, the vehicle control intention 202 and the corresponding intention intensity 203 can be determined based on the vehicle control instruction information 201. In this embodiment, the vehicle control intention 202 may include longitudinal control intentions (e.g., acceleration intention, deceleration intention, cruise speed setting intention, cruise speed adjustment intention, following distance setting intention, following distance adjustment intention, etc.) and / or lateral control intentions (e.g., left turn intention, right turn intention, lane change intention, overtaking intention, etc.). In some examples, the vehicle control intention 202 can be determined through various methods such as semantic parsing, gesture recognition, and preset intention correspondence rules. For example, if the user says "faster, faster," the vehicle control intention 202 can be determined as "acceleration intention"; if the user says "closer," the vehicle control intention 202 can be determined as "follow distance reduction intention"; if the user says "overtake quickly," the vehicle control intention 202 can be determined as a composite intention of "acceleration and lane change." For example, if a user quickly waves their hand forward, the vehicle control intention 202 can be determined as "acceleration intention"; if a user waves their hand downward twice, the vehicle control intention 202 can be determined as "reducing following distance intention"; if a user waves their hand to the left, the vehicle control intention 202 can be determined as "changing lanes to the left".

[0042] In this embodiment, the intent intensity 203 can be used to characterize the strength of the desired vehicle control intent 202. In some examples, the intent intensity 203 can be a discrete level value. In one example, the user's desired intensity can be divided into three levels: low, medium, and high, corresponding to intent intensities 203 of 1, 2, and 3, respectively. In some examples, the intent intensity 203 can be a continuous value. In one example, it can continuously take values ​​within the range of 0 to 1, where the closer the intent intensity 203 is to 0, the weaker the desired intensity, and the closer it is to 1, the stronger the desired intensity. In some examples, the intent intensity 203 can be determined through semantic parsing, preset intent value correspondence rules, etc. For example, if the user says "too slow," the intent intensity can be determined to be "high" or a relatively high value (e.g., 0.8) based on the degree adverb "too." Another example is that the intent intensity 203 can be determined to be "high" or a relatively high value (e.g., 0.9) based on the user's pressure when sliding the accelerator bar being greater than 5N; the intent intensity 203 can also be determined based on the frequency with which the user clicks the accelerator button.

[0043] In step S230, safety boundary information 204 corresponding to vehicle control intention 202 can be obtained. In this embodiment, safety boundary information 204 can be used to indicate the boundaries of safety constraints that the vehicle should comply with. In some examples, safety boundary information 204 can refer to a set of safety constraint conditions corresponding to vehicle control intention 202, used to constrain the execution range of vehicle control parameters. In some examples, safety boundary information 204 can be used to ensure that the vehicle is controlled within a safe range allowed by the current traffic environment, vehicle dynamics conditions, and interactions with other traffic participants. In some examples, safety boundary information 204 can include at least one of longitudinal control safety boundaries, lateral control safety boundaries, and interactive safety boundaries. In some examples, longitudinal safety boundaries can include, but are not limited to, at least one of the following: maximum permissible acceleration, maximum permissible deceleration, minimum following distance, and target speed limit. In some examples, lateral safety boundaries can include, but are not limited to, at least one of the following: maximum permissible steering angle, maximum permissible yaw rate, maximum permissible lateral acceleration, and minimum lane change longitudinal safety distance. In some examples, the interactive safety boundary may include, but is not limited to, at least one of the following: a time-to-collision (TTC) threshold, a minimum lateral safety clearance, and a confidence threshold for a vehicle's intention to cut in. In some examples, the specific form of the safety boundary information 204 may be adapted to the vehicle control intention 202. In one example, when the vehicle control intention 202 is an acceleration intention, the corresponding safety boundary information 204 may include the maximum permissible acceleration at the current vehicle speed (e.g., 2.5 m / s²). 2The information includes the following distance and the current speed limit (e.g., 120 km / h). In one example, the current speed limit can be determined based on the road speed limit. For example, the current speed limit can be set to the same value as the road speed limit. Alternatively, the current speed limit can be a value determined based on the road speed limit, adjusted according to weather (e.g., rain, fog, strong winds, visibility) and road conditions (e.g., road curvature radius, road surface adhesion coefficient). Similarly, the maximum permissible acceleration can be adjusted accordingly based on weather and road conditions. In one example, when the vehicle control intention 202 is a following distance adjustment intention, the corresponding safety boundary information 204 can include the minimum safe following distance (e.g., the distance value corresponding to a 1.5s interval). In one example, when the confidence level of a vehicle in an adjacent lane cutting into this lane is higher than 0.7, the minimum safe following distance in the corresponding safety boundary information 204 can be dynamically increased to a 2.5s interval. In one example, when the vehicle control intention 202 is a lane change intention, the corresponding safety boundary information 204 may include the maximum allowable steering angle at the current vehicle speed (e.g., 8°), the minimum longitudinal safety distance for lane change (e.g., the TTC between the vehicle and the vehicle behind in the target lane is greater than 5s), the lateral distance (e.g., greater than 1.5m), and the maximum allowable lateral acceleration (e.g., 0.3g).

[0044] In some examples, multimodal perception data can be acquired first, and then safety boundary information 204 can be determined based on the multimodal perception data. In one example, multimodal perception data may include, but is not limited to, at least one of the following: current vehicle speed, distance to the vehicle in front, distance to the vehicle behind, relative speed to nearby vehicles in the same lane and / or adjacent lanes, lane speed limit signs, motor vehicle lane signs, traffic light status, road curvature radius, weather visibility, road surface adhesion coefficient, and the cutting intention of vehicles in adjacent lanes. In some examples, the determined safety boundary information 204 can also be obtained directly from the corresponding onboard computing module.

[0045] In step S240, the target parameter value 208 of the vehicle control parameters associated with the vehicle control intention 202 can be determined at least based on the vehicle control intention 202, the intention strength 203, and the safety boundary information 204. In this embodiment, the target parameter value 208 of the vehicle control parameters associated with the vehicle control intention 202 can be determined by combining at least the vehicle control intention 202, the intention strength 203, and the safety boundary information 204. In this embodiment, the target parameter value 208 can refer to the execution value of the vehicle control parameters determined by comprehensively considering the control direction or control target indicated by the vehicle control intention 202, the user expectation intensity reflected by the intention strength 203, and the physical and environmental constraints provided by the safety boundary information 204. In some examples, various methods can be used to determine the target parameter value 208, for example, see [reference needed]. Figures 3-6 Detailed description of the embodiments.

[0046] The embodiments of this disclosure introduce a quantification mechanism for intent intensity based on vehicle control intent, and further combine safety boundary information to constrain target parameter values. This enables accurate and safe determination of vehicle control parameters without relying entirely on manual operation, thereby improving the efficiency and safety of human-machine interaction during driving.

[0047] It should be understood that all steps and their order in process 200 are exemplary, and embodiments of this disclosure will also cover any modifications to process 200. For example, in some implementations, other operations may continue to be performed after step S240, as detailed below. Figure 8 The corresponding description of the embodiments.

[0048] Figure 3 Another exemplary process 300 for determining vehicle control parameters according to an embodiment is shown. Figure 3 Based on Figure 2 Another exemplary implementation of process 200 is shown. Figure 3 The vehicle control instruction information 301, vehicle control intent 302, intent strength 303, safety boundary information 304, and target parameter value 308 shown can be referenced respectively. Figure 2 The vehicle control instruction information 201, vehicle control intent 202, intent strength 203, safety boundary information 204, and target parameter value 208 shown in the figure. Figure 3 Steps S310, S320, and S330 shown can be referred to respectively. Figure 2 Steps S210, S220, and S230 are shown in the diagram. To avoid repetition, only the differences will be described here.

[0049] In step S340, driver state information 305 can be acquired. In this embodiment, driver state information 305 may include various information characterizing the driver's current physiological state, cognitive state, and / or attention state, to assess the driver's ability to confirm and / or execute the vehicle control intention 202 at the current moment. In some examples, driver state information 305 may be raw perception data directly collected by onboard sensors, or it may be a quantitative indicator obtained after processing the raw perception data using a preset algorithm. In some examples, driver state information 305 may include visual attention state information (such as gaze deviation time, head turning angle), facial fatigue state information (such as blinking frequency, PERCLOS value calculated by the fatigue detection algorithm), physiological signal state information (such as heart rate, heart rate variability), or driving behavior state information (such as steering wheel grip strength, pedal response delay), etc.

[0050] In some examples, driver status information can be directly obtained through an in-vehicle driver monitoring system (DMS). For instance, facial images of the driver can be captured by an in-vehicle camera and processed using visual algorithms to obtain attention or fatigue indicators such as gaze deviation time and blink frequency. In some examples, signals such as heart rate and skin conductance can also be obtained through physiological sensors integrated into the steering wheel / seat. In some examples, a multi-source data fusion model can be used to combine the above visual, physiological, and behavioral data to output a comprehensive driver status index. In one example, the driver status index can range from 0-100%, with lower values ​​indicating less driver concentration. For example, a driver status index of 60% could indicate mild fatigue driving.

[0051] In step S350, the target parameter value 308 of the vehicle control parameters associated with the vehicle control intention 302 can be determined based on the vehicle control intention 302, intention strength 303, safety boundary information 304, and driver state information 305. In this embodiment, the vehicle control parameters associated with the vehicle control intention 302 can refer to specific control variables in the vehicle control system that can be adjusted to execute the vehicle control intention 302. In some examples, a mapping relationship between vehicle control parameters and vehicle control intention 302 can be preset. For example, longitudinal control intentions (such as acceleration, deceleration, cruise speed setting, and following distance adjustment) can correspond to longitudinal control parameters (such as target acceleration, target deceleration, target cruise speed, and target following distance); lateral control intentions (such as steering, lane changing, and overtaking) can correspond to lateral control parameters (such as target steering angle, target yaw rate, target lateral acceleration, and target lane change trajectory); and composite control intentions can correspond to a combination of multiple sets of control parameters. Accordingly, the target parameter value 308 can refer to the target value or target state that the vehicle control parameters should achieve when executing the vehicle control intention 302.

[0052] In this embodiment, the target parameter value 308 of the vehicle control parameters associated with the vehicle control intention 302 can be determined by combining the vehicle control intention 302, the intention intensity 303, the safety boundary information 304, and the driver state information 305. In some examples, the initial reference value of the vehicle control parameter associated with the vehicle control intention 302 can be determined based on the intention intensity 303, then the amplitude can be limited according to the corresponding limit in the safety boundary information 304, and finally, it can be corrected according to the driver state information 305. For example, the vehicle control intention 302 can be determined to be a longitudinal acceleration intention based on the user's voice "faster, faster," and the corresponding intention intensity 303 can be "high." Then, a higher initial target acceleration (e.g., 3 m / s²) can be determined. 2Then, based on the maximum permissible acceleration indicated by safety boundary information 304 (e.g., 2.0 m / s²), 2 The amplitude is limited, and the final correction can be made based on the driver status information 305. For example, if the driver status information confidence index is 0.9, then the final target parameter value 308 is 2.0 m / s. 2 ×0.9=1.8m / s 2 For example, based on the user's gesture of "quickly waving to the left," the vehicle control intention 302 can be determined to be a lane change intention, with a corresponding intention intensity 303 of 0.9. Then, a larger initial target steering angle can be determined (e.g., 0.9 × 8° = 7.2°). Then, based on the safety boundary information 304 indicating the rear vehicle TTC threshold (e.g., 5s), it is checked whether the current rear vehicle TTC meets the requirements. If the rear vehicle TTC does not meet the rear vehicle TTC threshold, the target parameter value 308 of the target steering angle is determined to be 0°, which means lane changing is prohibited.

[0053] By using the above method, user expectations, safety constraints, and driver capability status can be comprehensively considered in the determination of the target parameter value 308, thereby ensuring that the control result meets both user intent and safety requirements.

[0054] Figure 4 An exemplary target parameter value determination process 400 is shown in a method for determining vehicle control parameters according to an embodiment. Figure 4 It can be Figure 3 An exemplary implementation of step S350 in the method. Figure 4 The vehicle control intent 402, intent strength 403, safety boundary information 404, driver state information 405, and target parameter value 408 shown can be referenced respectively. Figure 3 The diagram shows vehicle control intent 302, intent strength 303, safety boundary information 304, driver state information 305, and target parameter value 308. To avoid repetition, only the differences will be described here.

[0055] like Figure 4As shown, in step S410, it can be determined whether a preset intention intensity correction condition is met based on the vehicle control intention 402, intention intensity 403, and driver state information 405. In this embodiment, the intention intensity correction condition can be a set of preset rules used to determine whether to correct the intention intensity 403. In some examples, the intention intensity correction condition can be a single threshold condition (such as triggering correction when the driver state information 405 is below a certain threshold), a combination of multiple conditions (such as multiple state indicators being abnormal simultaneously), or a dynamically weighted condition (such as adjusting the threshold according to different scenarios). In one example, when the vehicle control intention 402 is a longitudinal acceleration intention and the intention intensity 403 is greater than 0.7, if the driver state information 405 indicates that the gaze deviation time exceeds 1.5 seconds, then the intention intensity correction condition is met. In one example, when the vehicle control intention 402 is a lane change intention, if the driver state information 405 indicates that the PERCLOS value exceeds 0.15 or the head deflection angle exceeds 45°, then the intention intensity correction condition is met.

[0056] If step S410 is determined to be yes, steps S420 to S430 can be executed.

[0057] In step S420, the intention intensity 403 can be corrected to obtain a corrected intention intensity 406. In some examples, the correction process may refer to processing the intention intensity 403 according to the driver state information 405 (e.g., weighting, limiting, zeroing, etc.) to reflect the degree to which the driver's current cognitive ability reliably supports their control expectations. In some examples, when the driver state information 405 indicates that the driver is in a state of inattention, fatigue, or stress, the intention intensity 403 can be corrected to a lower value. In some examples, when the driver state information 405 indicates that the driver's state is severely abnormal, the corrected intention intensity 406 can be set to 0, thereby refusing to execute the corresponding vehicle control intention 402.

[0058] In step S430, the target parameter value 408 of the vehicle control parameters associated with the vehicle control intention 402 can be determined based on the vehicle control intention 402, the modified intention strength 406, and the safety boundary information 404. In this embodiment, reference can be made to... Figure 2 In the embodiment, the corresponding operation of step S240 simply involves replacing the intent strength with the modified intent strength.

[0059] If step S410 determines otherwise, in step S440, the target parameter value 408 of the vehicle control parameter associated with the vehicle control intention 402 can be determined based on the vehicle control intention 402, the intention strength 403, and the safety boundary information 404. In this embodiment, reference can be made to... Figure 2 The corresponding operation of step S240 in the embodiment.

[0060] By using the above methods, the execution intensity can be dynamically calibrated according to the driver's real-time status while fully respecting the user's control intentions. This avoids safety risks caused by excessive commands due to driver fatigue, distraction, or other states, and improves the safety of vehicle control.

[0061] Figure 5 Another exemplary target parameter value determination process 500 is shown in the method for determining vehicle control parameters according to an embodiment. Figure 5 It can be Figure 2 An exemplary implementation of step S240 in the method. Figure 5 The intent intensity 503, security boundary information 504, and target parameter value 508 shown can be referenced respectively. Figure 2 The intent intensity 203, security boundary information 204, and target parameter value 208 are shown in the figure. Figure 5 The parameter shown in the figure indicates that the desired adjustment direction 5022 can be... Figure 2 The following is an exemplary implementation of the vehicle control intent 202. To avoid repetition, only the differences will be described here.

[0062] In this embodiment, the vehicle control intent can be used to indicate the vehicle control parameters to be adjusted and the desired adjustment direction 5022. In some examples, the vehicle control intent can be used to indicate the cruise speed and the direction of cruise speed adjustment (e.g., increasing or decreasing). In some examples, the vehicle control intent can be used to indicate the following distance and the direction of following distance adjustment (e.g., increasing or decreasing). In some examples, the lateral vehicle control intent can be used to indicate the steering angle and the direction of steering angle adjustment (e.g., left or right relative to straight driving).

[0063] like Figure 5As shown, in step S510, the parameter adjustment range 511 can be determined at least based on the current parameter value 507 of the vehicle control parameter to be adjusted, the desired parameter adjustment direction 5022, and the safety boundary information 504. In this embodiment, the parameter adjustment range 511 can refer to the allowable adjustment interval determined by using the current parameter value 507 of the vehicle control parameter to be adjusted as the base value, combined with the desired parameter adjustment direction 5022 and the safety boundary information 504. In some examples, the parameter adjustment range 511 can be used to limit the range of values ​​that the vehicle control parameter can vary within a single adjustment cycle. In one example, the parameter adjustment range 511 can use the current parameter value 507 as one endpoint, and then determine the other endpoint based on the corresponding constraint conditions indicated by the safety boundary information 504 and the desired parameter adjustment direction 5022. In one example, the vehicle control parameter to be adjusted could be the cruise speed, the current parameter value 507 could be 100 km / h, the desired adjustment direction 5022 could represent "increase", and the safety boundary information 504 could be 120 km / h. Therefore, the parameter adjustment range 511 can be determined to be [100, 120] km / h. In another example, if the desired adjustment direction 5022 represents "decrease", the parameter adjustment range 511 can be determined to be [v...]. min

[100] km / h. For example, v min It can be determined based on the minimum speed limit of the road. For example, v min It can be determined based on the speed of surrounding vehicles to prevent rear-ending the vehicle in front or being rear-ended by the vehicle behind. For example, v min This can be determined based on the current vehicle speed and the preset deceleration rate for smooth deceleration. In one example, the vehicle control parameters to be adjusted could be following distance, the current parameter value 507 could be 2.4s time interval, the desired adjustment direction 5022 could represent "decrease", and the safety boundary information 504 could be 1.4s time interval, thus determining the parameter adjustment range 511 as [1.4s, 2.4s] time interval. In another example, if the desired adjustment direction 5022 represents "increase", the parameter adjustment range 511 could be determined as [2.4s, t]. max Time interval. For example, t max This can be a value determined based on current vehicle speed and road conditions, allowing sufficient reaction time without affecting road traffic efficiency. For example, t... max It can be determined based on the speed and distance of surrounding vehicles to prevent rear-ending the vehicle in front or being rear-ended by the vehicle behind. In one example, the vehicle control parameter to be adjusted can be the steering angle, the current parameter value 507 can be 0°, the desired adjustment direction of the parameter 5022 can be "left", the safety boundary information 504 can indicate that the maximum left turn angle is 5°, and the parameter adjustment range 511 can be determined to be [0°, 5°].

[0064] In step S520, a target parameter value 508 for the vehicle control parameter associated with the vehicle control intention can be determined from the parameter adjustment range 511, at least based on the intention intensity 503. In this embodiment, the target parameter value 508 can be a specific value that the vehicle control parameter should achieve, mapped from the intention intensity 503 within the parameter adjustment range 511. In some examples, the intention intensity 503 can be mapped to a relative position within the parameter adjustment range 511. For example, the higher the intention intensity 503, the closer the target parameter value 508 is to the far boundary of the parameter adjustment range 511 along the desired parameter adjustment direction 5022. In some examples, the target parameter value 508 can be determined by a pre-trained first mapping model. For example, the intention intensity 503, the desired parameter adjustment direction 5022, and the parameter adjustment range 511 can be input into a trained first neural network model to obtain the model output target parameter value 508.

[0065] In some implementations, the target parameter value 508 of the vehicle control parameters associated with the vehicle control intention can be determined using a pre-built linear mapping model, at least based on the intention intensity 503. In some examples, the linear mapping model can be used to indicate at least the mapping relationship between the vehicle control parameters associated with the vehicle control intention, the intention intensity, and the parameter adjustment range. In one example, when the desired adjustment direction 5022 of the parameter indicates "increase," the linear mapping model can be represented as target=T. min +R·(T max -T min When the expected adjustment direction of the parameter 5022 represents "decreasing", the linear mapping model can be expressed as target=T. min +(1-R)·(T max -T min ), where target can be used to represent the target parameter value 508, T max and T min These can be used to represent the upper and lower boundaries of the parameter adjustment range 511, respectively, and R can be used to represent the intent intensity 503.

[0066] In some implementations, Figure 5 It could also be Figure 3An exemplary implementation of step S350 in the method. In some implementations, step S520 may also involve determining a target parameter value 508 of the vehicle control parameter associated with the vehicle control intention from the parameter adjustment range 511 based on the intention intensity 503 and driver state information. In some examples, the target parameter value 508 can be determined by a pre-trained second mapping model. For example, the intention intensity 503, driver state information, desired parameter adjustment direction 5022, and parameter adjustment range 511 can be input into a trained second neural network model to obtain the model output target parameter value 508. In some examples, reference can also be made to... Figure 4 The intent intensity correction process described in the embodiment obtains the corrected intent intensity based on the driver's state information, and then uses the linear mapping model as described above to determine the target parameter value 508. The meaning of R is simply replaced with the corrected intent intensity, which will not be elaborated here.

[0067] By using the above method, the intensity of user intent can be mapped to specific target values ​​of vehicle control parameters under the constraints of safety boundaries, thereby achieving a gradient response to user expectations. This not only avoids control parameters from exceeding the safety range but also ensures differentiated execution intensity corresponding to different intensities of intent, thus improving control accuracy and safety.

[0068] Figure 6 Another exemplary target parameter value determination process 600 is shown in the method for determining vehicle control parameters according to an embodiment. Figure 6 It can be Figure 2 Another exemplary implementation of step S240 in the method. Figure 6 The intent intensity 603, security boundary information 604, and target parameter value 608 shown can be referenced respectively. Figure 2 The intent intensity 203, security boundary information 204, and target parameter value 208 are shown in the figure. Figure 6 The parameter shown in the figure has a target value of 6023 that can be... Figure 2 The following is an exemplary implementation of the vehicle control intent 202. To avoid repetition, only the differences will be described here.

[0069] In this embodiment, the vehicle control intention can be used to indicate the vehicle control parameters to be adjusted and the expected target value 6023 of the parameters. The expected target value 6023 of the parameters can be a specific value that the user expects to achieve for the vehicle control parameters, directly expressed by the user through vehicle control instruction information. In one example, the expected target value 6023 corresponding to the user's voice "accelerate to 120" is a cruising speed of 120 km / h. In another example, the expected target value 6023 corresponding to the user's voice "maintain a following distance of 100 meters" is a following distance of 100 meters.

[0070] like Figure 6 As shown, in step S610, it can be determined whether the expected target value 6023 of the parameter meets the preset safety conditions. In this embodiment, the safety conditions are determined at least based on the safety boundary information 604. In some examples, the safety conditions are used to check whether the expected target value 6023 of the parameter exceeds the safe range allowed by the current traffic environment and vehicle dynamics. For example, it may include a numerical upper limit condition, a numerical lower limit condition, or an interval inclusion condition. For example, the vehicle control intention may be a cruise speed setting intention, the expected target value 6023 of the parameter may be 130 km / h, and the safety boundary information 604 may indicate that the current road speed limit is 120 km / h. Therefore, it can be determined that the expected target value 6023 of the parameter exceeds the upper limit of the vehicle speed in the safety boundary information 604 and does not meet the safety conditions. As another example, the vehicle control intention may be a lane change intention, the expected target value 6023 of the parameter may be a steering angle of 8°, and the safety boundary information 604 may indicate that the maximum allowable steering angle at the current vehicle speed is 6°. Therefore, it can be determined that the expected target value 6023 of the parameter exceeds the upper limit of the steering angle in the safety boundary information 604 and does not meet the safety conditions.

[0071] If step S610 determines that it is yes, in step S620, the expected target value 6023 of the parameter can be determined as the target parameter value 608 of the vehicle control parameter associated with the vehicle control intention.

[0072] If step S610 determines otherwise, in step S630, the target parameter value 608 of the vehicle control parameter associated with the vehicle control intention can be determined at least based on the intention intensity 603 and the safety boundary information 604. In some examples, the target parameter value 608 of the vehicle control parameter associated with the vehicle control intention can be determined as the boundary value indicated by the safety boundary information 604 or kept at its current value based on the intention intensity 603. In one example, if the intention intensity 603 is greater than a preset intention threshold, the target parameter value 608 of the vehicle control parameter associated with the vehicle control intention can be determined as the boundary value indicated by the safety boundary information 604; if the intention intensity 603 is not greater than the preset intention threshold, the target parameter value 608 of the vehicle control parameter associated with the vehicle control intention can be kept at its current value. For example, in the example where the expected target value 6023 of the parameter exceeds the upper limit of vehicle speed in the safety boundary information 604, if the intention intensity 603 is greater than the preset intention threshold (e.g., 0.7), the expected target value 6023 of the parameter can be corrected to 120 km / h; otherwise, the expected target value 6023 of the parameter can be kept at its current value. For example, in the example mentioned above where the expected target value 6023 of the parameter exceeds the upper limit of the steering angle in the safety boundary information 604, if the intent intensity 603 is greater than the preset intent threshold (e.g., 0.7), the expected target value 6023 of the parameter can be limited and corrected to 6°; otherwise, the expected target value 6023 of the parameter can be kept at its current value.

[0073] In some implementations, Figure 6 It could also be Figure 3 An exemplary implementation of step S350 in the method. In some implementations, step S620 may also involve determining a target parameter value 608 for the vehicle control parameters associated with the vehicle control intention based on the expected target value 6023 and driver state information. In some examples, the vehicle control intention may be a cruise speed setting intention, the expected target value 6023 may be 120 km / h, and the safety boundary information 604 may indicate an upper limit of 120 km / h. If the driver state information indicates a line-of-sight deviation time of 1.8 seconds (e.g., corresponding to a confidence index of 0.6), the target speed in the target parameter value 608 can be determined to be 120 km / h. Optionally, the target acceleration can also be corrected to 0.6 times the original value to reach the set value at a smoother rate. If the driver state information indicates a PERCLOS value of 0.2 (e.g., corresponding to a confidence index of 0.3, below the threshold), the setting can be rejected, and the target parameter value 608 remains unchanged at the current speed.

[0074] In some implementations, step S630 may also involve determining the target parameter value 508 of the vehicle control parameters associated with the vehicle control intention based on the intention intensity 603, safety boundary information 604, and driver state information. In some examples, the vehicle control intention may be a cruise speed setting intention, the expected target value 6023 may be 130 km / h, and the safety boundary information 604 may indicate an upper limit of 120 km / h. If the intention intensity 603 is 0.85 (higher than the preset threshold of 0.7) and the confidence index corresponding to the driver state information is 0.9, then the target speed in the target parameter value 508 can be determined to be 120 km / h. Optionally, the target acceleration can also be determined to be 2.0 m / s² based on the intention intensity 603 and the confidence index. 2 ×0.85×0.9=1.53m / s 2 This allows the vehicle to approach the upper limit of the safety boundary at a higher execution rate. If the intent strength 603 is 0.4 (lower than the preset threshold of 0.7), it can be determined that the user's confirmation intention is insufficient, and the target parameter value 508 can be kept unchanged at the current cruising speed.

[0075] In the above manner, when users directly set the target value of control parameters, the safety boundary can be used as a hard constraint to limit the set value for verification. Furthermore, by combining the intent strength and driver status information, it can be further determined whether to limit execution or refuse execution, and the execution rate after limiting can also be determined. This ensures that the set value does not exceed the safety boundary, and can also realize differentiated execution strategies based on the user's confirmed intention and the driver's ability status, thereby improving the safety of control and user experience.

[0076] Figure 7 An exemplary intent and intent intensity determination process 700 in a method for determining vehicle control parameters according to an embodiment is shown. Figure 7 It can be Figure 2 An exemplary implementation of step S220 in the method. Figure 7 The vehicle control intent 702 and intent strength 703 shown can be referred to respectively. Figure 2 The vehicle control intent 202 and intent intensity 203 are shown in the figure. Figure 7 The voice signal 7011 shown in the figure can be Figure 2 The following is an exemplary implementation of the vehicle control instruction information 701 shown. To avoid repetition, only the differences will be described here.

[0077] In this embodiment, the vehicle control instruction information may include a voice signal 7011 indicating the vehicle control intention. In this embodiment, the voice signal 7011 may refer to the original acoustic signal containing the vehicle control intention input by the user via voice or digital audio data after analog-to-digital conversion.

[0078] like Figure 7 As shown, in step S710, speech signal 7011 can be processed for speech recognition to obtain speech-recognized text 711. In some examples, technologies such as end-to-end speech recognition models based on deep learning, acoustic models based on a hybrid of hidden Markov models and deep neural networks, or speech recognition systems based on the Transformer architecture can be used to convert speech signal 7011 into a text sequence, which serves as speech-recognized text 711.

[0079] In step S720, vehicle control intent 702 can be extracted from the speech-recognized text 711. In some examples, intent recognition methods based on rule template matching, intent recognition methods based on word vectors and classification models, or semantic understanding methods based on large language models can be used to extract control intent categories from the speech-recognized text 711, such as acceleration intent, deceleration intent, cruise speed setting intent, or lane change intent, as vehicle control intent 702.

[0080] In step S730, the intention intensity indication information 731 can be extracted from the speech recognition text 711 and / or the speech signal 7011. In this embodiment, the intention intensity indication information 731 can refer to the characteristic information that can reflect the strength of the user's vehicle control intention expression from the semantic level or the acoustic level. For example, it can include at least one of the following: modal particles, adjectives, adverbs of degree, mood emotions, and acoustic emotion features. In some examples, modal particles can include mitigating modal particles such as "ne, ba, a", and urging modal particles such as "wei, kuai dian, gan jin"; adjectives and adverbs of degree can include weak degree words such as "wei wei, shao wei, dian", and strong degree words such as "hen hen, tai, fei chang"; mood emotions can refer to the information extracted from the speech recognition text 711 that reflects the overall emotional tendency and emotional attitude of the user when expressing the vehicle control intention, which can be obtained through various text emotion analysis methods (such as rule matching based on an emotion dictionary, a text emotion classification model based on deep learning, or semantic understanding based on a large language model), and can include categories such as eagerness, impatience, calmness, polite request, command, and blame; acoustic emotion features can include the fundamental frequency dynamic range, short-time energy, speech rate change rate, and formant parameters of the speech signal.

[0081] In step S740, the intention intensity 703 corresponding to the vehicle control intention 702 can be determined according to the intention intensity indication information 731. In some examples, according to a preset intention intensity mapping table or a pre-trained intention intensity evaluation model, the extracted intention intensity indication information 731 can be mapped to the intention intensity 703. For example, if the speech recognition text 711 is "kuai dian kuai dian", the corresponding intention intensity 703 can be determined to be 0.8 according to the repeated urging modal particles and the adverb of degree "kuai" contained therein. Another example is that if the speech recognition text 711 is "shao wei jia dian su", the corresponding intention intensity 703 can be determined to be 0.3 according to the adverb of degree "shao wei" contained therein.

[0082] In some examples, the intention intensity 703 can be determined by synthesizing the intention intensity indication information 731 in multiple dimensions. For example, if the speech recognition text 711 is "neng bu neng kuai dian a", the corresponding semantic sub-intensity can be determined to be 0.7 according to the mood emotion of "eagerness", the corresponding adverb sub-intensity can be determined to be 0.6 according to the adverb of degree "kuai dian", and the corresponding acoustic sub-intensity can be determined to be 0.8 according to the acoustic emotion features that the fundamental frequency of the speech signal is 40% higher than the mean value and the short-time energy exceeds the threshold. Then, the determined semantic sub-intensity, adverb sub-intensity, and acoustic sub-intensity can be fused according to a preset weight (such as 0.4:0.3:0.3), and the intention intensity 703 = 0.7×0.4 + 0.6×0.3 + 0.8×0.3 = 0.70 can be obtained.

[0083] By using the above methods, we can extract intent intensity indication information from user voice from multiple dimensions, and combine semantic tone and emotion, degree adverbs and acoustic emotional features to achieve accurate quantification of the intensity of user control expectations. This avoids misjudgment caused by a single feature, improves the accuracy and robustness of intent intensity, and provides a basis for determining subsequent target values.

[0084] Figure 8 Another exemplary process 800 for determining vehicle control parameters according to an embodiment is shown. Figure 8 Based on Figure 2 Another exemplary implementation of process 200 is shown. Figure 8 The vehicle control instruction information 801, vehicle control intent 802, intent strength 803, safety boundary information 804, and target parameter value 808 shown can be referenced respectively. Figure 2 The vehicle control instruction information 201, vehicle control intent 202, intent strength 203, safety boundary information 204, and target parameter value 208 shown in the figure. Figure 8 Steps S810, S820, S830, and S840 shown can be referred to respectively. Figure 2 Steps S210, S220, S230, and S240 are shown in the diagram. To avoid repetition, only the differences will be described here.

[0085] like Figure 8 As shown, in step S850, it can be determined whether the target parameter value 808 is consistent with the vehicle control intention 802. In some examples, the determination result of the target parameter value 808 relative to the vehicle control intention 802 can include three situations: direct execution, modified execution, and refusal to execute. When the target parameter value 808 directly responds to the control direction and / or control quantity indicated by the vehicle control intention 802, it can be determined to be consistent; when the target parameter value 808, due to the constraints of the safety boundary information 804 and / or driver state information, performs a limit correction, direction adjustment, or directly refuses to execute the control quantity indicated by the vehicle control intention 802, it can be determined to be inconsistent.

[0086] If step S850 determines otherwise, in step S860, explanation information 861 can be generated at least based on the safety boundary information 804. In this embodiment, explanation information 861 can refer to information output to the user explaining why the target parameter value 808 deviates from the user's original expectation when the target parameter value 808 is inconsistent with the vehicle control intention 802. In some examples, explanation information 861 can include at least one of text information, voice information, visual information, or tactile information. In some examples, explanation information 861 can be natural language text or voice broadcast, such as "The distance to the vehicle ahead is too close, acceleration is not possible at this time", "The vehicle behind is approaching, it is not suitable to change lanes at this time", "The current road surface is slippery, you have adjusted to a safe distance". In some examples, the text or voice can be generated by filling in a preset template with specific triggering conditions in the safety boundary information 804, or it can be automatically generated by a generative model based on the safety boundary information 804. In some examples, the explanatory information 861 can be visual information, such as projecting a red warning box onto the rear of the vehicle ahead via an augmented reality head-up display, or displaying a current safety boundary diagram (such as an icon of the vehicle ahead and distance markings) on a central control display screen. In some examples, the content and location of the visual information can be determined based on the specific constraints triggered in the safety boundary information 804. In some examples, the explanatory information 861 can be tactile information, such as conveying tactile codes for refusal to comply or safety warnings to the user through steering wheel vibration or seat vibration patterns. In some examples, the vibration frequency and amplitude of the tactile information can be determined based on the risk level in the safety boundary information 804.

[0087] If step S850 determines the condition is yes, the process can return to step S810 to begin a new round of vehicle control parameter determination. In some examples, user input can be re-acquired, environmental conditions can be perceived, and safety boundary information can be updated to achieve periodic dynamic optimization of vehicle control parameters.

[0088] By using the above methods, when the user's intentions are corrected by the system's safety constraints, the system can proactively provide feedback to the user on the reasons for the correction and the basis for the decision, avoiding confusion or a crisis of trust for the user due to the control results not meeting expectations. Furthermore, continuous monitoring and dynamic optimization can be achieved through periodic cycles, thereby improving driving safety and user experience.

[0089] Figure 9 An exemplary flowchart of a method 900 for determining vehicle control parameters according to an embodiment is shown.

[0090] like Figure 9 As shown, in step S910, the vehicle control instruction information input by the user can be obtained.

[0091] In step S920, the vehicle control intention and the corresponding intention intensity can be determined based on the vehicle control instruction information. The intention intensity is used to characterize the expected strength of the vehicle control intention.

[0092] In step S930, safety boundary information corresponding to the vehicle control intention can be obtained. The safety boundary information is used to indicate the boundaries of the safety constraints that the vehicle should comply with.

[0093] In step S940, the target parameter value of the vehicle control parameter associated with the vehicle control intention can be determined at least based on the vehicle control intention, the intention strength, and the safety boundary information.

[0094] In one implementation, the method may further include: acquiring driver state information, wherein determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based at least on the vehicle control intention, the intention strength, and the safety boundary information includes: determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based on the vehicle control intention, the intention strength, the safety boundary information, and the driver state information.

[0095] In one implementation, determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based on the vehicle control intention, the intention intensity, the safety boundary information, and the driver state information includes: determining whether a preset intention intensity correction condition is met based on the driver state information, the vehicle control intention, and the intention intensity; correcting the intention intensity in response to meeting the intention intensity correction condition to obtain a corrected intention intensity; and determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based on the vehicle control intention, the corrected intention intensity, and the safety boundary information.

[0096] In one implementation, the vehicle control intention is used to indicate the vehicle control parameter to be adjusted and the desired adjustment direction of the parameter. The step of determining the target parameter value of the vehicle control parameter associated with the vehicle control intention, at least based on the vehicle control intention, the intention strength, and the safety boundary information, includes: determining the parameter adjustment range, at least based on the current parameter value of the vehicle control parameter to be adjusted, the desired adjustment direction of the parameter, and the safety boundary information; and determining the target parameter value of the vehicle control parameter to be adjusted from the parameter adjustment range, at least based on the intention strength.

[0097] In one implementation, the target parameter value of the vehicle control parameter to be adjusted is determined using a pre-built linear mapping model, at least based on the intent intensity, wherein the linear mapping model is used to indicate the mapping relationship between the vehicle control parameter to be adjusted, the intent intensity, and the parameter adjustment range.

[0098] In one implementation, determining the target parameter value of the vehicle control parameter to be adjusted from the parameter adjustment range, at least based on the intent intensity, includes: determining the target parameter value of the vehicle control parameter to be adjusted from the parameter adjustment range based on the intent intensity and the driver state information.

[0099] In one implementation, the vehicle control intention is used to indicate vehicle control parameters to be adjusted and expected target values ​​for the parameters. Determining the target parameter value of the vehicle control parameter associated with the vehicle control intention, at least based on the vehicle control intention, the intention strength, and the safety boundary information, includes: determining whether the expected target value of the parameter satisfies a preset safety condition, the safety condition being determined at least based on the safety boundary information; in response to satisfying the safety condition, determining the expected target value of the parameter as the target parameter value of the vehicle control parameter associated with the vehicle control intention; and in response to not satisfying the safety condition, determining the target parameter value of the vehicle control parameter associated with the vehicle control intention, at least based on the intention strength and the safety boundary information.

[0100] In one implementation, the vehicle control instruction information includes a voice signal indicating a vehicle control intention. Determining the vehicle control intention and its corresponding intensity based on the vehicle control instruction information includes: performing speech recognition processing on the voice signal to obtain speech recognition text; extracting the vehicle control intention from the speech recognition text; extracting intent intensity indication information from the speech recognition text and / or the voice signal, wherein the intent intensity indication information includes at least one of the following: modal particles, adjectives, degree adverbs, emotional tone, and acoustic emotion features; and determining the intent intensity corresponding to the vehicle control intention based on the intent intensity indication information.

[0101] In one implementation, the method further includes: in response to the inconsistency between the target parameter value and the vehicle control intention, generating explanatory information based at least on the safety boundary information.

[0102] Figure 10 An exemplary flowchart 100 of a method for controlling a vehicle according to an embodiment is shown.

[0103] like Figure 10As shown, in step S1010, the target parameter value of the vehicle control parameter associated with the vehicle control intention can be obtained. In some examples, the target parameter value of the vehicle control parameter associated with the vehicle control intention can be based on the above... Figures 1 to 9 The method described is determined by the vehicle control method.

[0104] In step S1020, vehicle control prompt information corresponding to the vehicle control intention can be output based on the target parameter value. In this embodiment, the vehicle control prompt information can refer to the information output by the system to the user before the target parameter value is officially executed, which is used to prompt the vehicle control action to be executed and the corresponding parameters. In some examples, the vehicle control prompt information may include at least one of voice broadcast information, visual prompt information, and tactile prompt information. In one example, the target parameter value may indicate that the following distance is shortened from 3 seconds to 2.4 seconds. This can be indicated by broadcasting "The following distance is about to decrease" through the in-vehicle voice system, or by projecting a green virtual light carpet in front of the vehicle through an augmented reality head-up display device to visually prompt the upcoming longitudinal control action. Optionally, the length, brightness, and color of the virtual light carpet can be determined according to the vehicle control parameters associated with the vehicle control intention, for example, it can be adjusted according to the cruise speed increment and the amount of decrease in following distance. In some examples, the target parameter value can indicate a left lane change with a target steering angle of 3°, which can be indicated by steering wheel vibration or by displaying "About to change lanes to the left, please confirm" on the central control display screen, thus providing a combination of tactile and visual prompts for the upcoming lateral control action.

[0105] In step S1030, in response to the absence of a cancellation operation message corresponding to the vehicle control prompt within a preset time window, or the receipt of a confirmation operation message corresponding to the vehicle control prompt within the preset time window, the vehicle is controlled according to the target parameter value. In this embodiment, the preset time window may refer to the set duration after the system outputs the vehicle control prompt message and waits for the user to confirm again, thus providing the user with a control confirmation window or an opportunity to cancel the operation. In some examples, the cancellation operation message or confirmation operation message may refer to information input by the user through voice, touch, gesture, gaze focus (such as gaze confirmation, eye movement confirmation), or key presses, used to cancel or confirm the vehicle control action to be executed. In some examples, if no cancellation operation message is detected within the preset time window, or if the user's implicit or explicit confirmation is detected, the vehicle can be driven to perform the corresponding control according to the target parameter value. In one example, after outputting a vehicle control prompt, a 3-second countdown begins. If the user does not utter a cancellation statement such as "cancel" or "never mind," nor click the cancel button on the touchscreen, the vehicle will be controlled according to the target parameter value when the 3-second countdown expires. This could include actions such as accelerating to 120 km / h, shortening the following distance, or changing lanes. In other examples, after outputting a lane change prompt, a 5-second countdown begins. If the user utters "cancel" or a cancellation operation is detected during this time, the lane change operation will not be performed, the target parameter value will be invalidated, and the vehicle will continue driving in the current lane and at the current speed.

[0106] By employing the above methods, a pre-execution verification mechanism can be introduced before changing vehicle control parameters, avoiding unexpected control actions caused by intent recognition errors, user slips of the tongue, or temporary changes in intent. Simultaneously, interactive devices such as augmented reality head-up displays can be used to present the upcoming control results to the user through multiple dimensions, including visual, tactile, and auditory senses, thus balancing interactive convenience with control safety and enhancing user trust in natural interactive driving control functions.

[0107] Figure 11 A schematic block diagram of a device 1100 for determining vehicle control parameters according to an embodiment is shown.

[0108] like Figure 11As shown, the device 1100 for determining vehicle control parameters may include: a control information acquisition module 1110, configured to acquire vehicle control instruction information input by a user; an intent determination module 1120, configured to determine a vehicle control intent and a corresponding intent intensity based on the vehicle control instruction information, wherein the intent intensity is used to characterize the expected strength of the vehicle control intent; a safety information acquisition module 1130, configured to acquire safety boundary information corresponding to the vehicle control intent, wherein the safety boundary information is used to indicate the boundary of safety constraints that the vehicle should comply with; and a parameter determination module 1140, configured to determine target parameter values ​​of vehicle control parameters associated with the vehicle control intent based at least on the vehicle control intent, the intent intensity, and the safety boundary information.

[0109] Furthermore, the apparatus 1100 for determining vehicle control parameters may also include any other modules configured to perform any operation of the method for determining vehicle control parameters according to the above embodiments of the present disclosure.

[0110] Figure 12 A schematic block diagram of a device 1200 for controlling a vehicle according to an embodiment is shown.

[0111] like Figure 12 As shown, the device 1200 for controlling a vehicle may include: a device 1210 for determining vehicle control parameters; an information output module 1220 configured to output vehicle control prompt information corresponding to the vehicle control intention based on the target parameter value; and a vehicle control module 1230 configured to control the vehicle according to the target parameter value in response to not receiving a cancellation operation information corresponding to the vehicle control prompt information within a preset time window or receiving a confirmation operation information corresponding to the vehicle control prompt information within a preset time window. The device 1210 for determining vehicle control parameters can be referred to the foregoing description. Figure 11 The device 1100 in the embodiment for determining vehicle control parameters.

[0112] Figure 13 A schematic block diagram of a processing apparatus 1300 according to an embodiment is shown.

[0113] The processing apparatus or processing system 1300 may include one or more control units or processors 1310 that execute one or more machine-readable instructions stored in a machine-readable storage medium (i.e., memory 1320). In one embodiment, the processor 1310 is configured, when executing program instructions, to execute the instructions in conjunction with the above. Figures 1 to 10 The various operations and functions described herein. Those skilled in the art will understand that the apparatus described in the embodiments of this disclosure may also include various other components, such as various communication modules, bus modules, and possibly user interface modules.

[0114] According to one embodiment, the processing device 1300 may be an on-board electronic control unit.

[0115] Figure 14 A schematic block diagram of a driving assistance system 1400 according to an embodiment is shown.

[0116] like Figure 14 As shown, the driving assistance system 1400 may include: a sensor 1410 configured to acquire environmental perception information; and a processing device 1420, wherein the safety boundary information is determined at least based on the environmental perception information. The processing device 1420 may refer to the foregoing. Figure 13 The processing device 1300 in the embodiment. In one implementation, the sensor 1410 may include at least one of a camera, millimeter-wave radar, and lidar.

[0117] According to one embodiment, a machine-readable storage medium is provided. This readable medium may store executable instructions that, when executed by a processor, can perform the above-described combinations of various embodiments of this disclosure. Figures 1 to 10 The various operations and functions described.

[0118] According to one embodiment, a computer program product is provided. The computer program product includes machine-executable instructions that, when executed by a processor, are capable of performing the above-described combinations in various embodiments of this disclosure. Figures 1 to 10 The various operations and functions described.

[0119] The specific embodiments described above with reference to the accompanying drawings are exemplary embodiments, but do not represent all embodiments that can be implemented or fall within the scope of the claims. The term "example" as used throughout this disclosure means "serving as an example, instance, or illustration" and does not imply that it is "preferred" or "advantageous" compared to other embodiments. Specific details are included to provide an understanding of the described techniques. However, these techniques can be practiced without these specific details. In some instances, well-known structures and apparatuses are shown in block diagram form to avoid obscuring the concepts of the described embodiments.

[0120] In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0121] Not all steps and units in the above process and system structure diagrams are mandatory; some steps or units can be omitted as needed. The execution order of each step is not fixed and can be determined as required. The device structure described in the above embodiments can be a physical structure or a logical structure. That is, some units may be implemented by the same physical entity, or some units may be implemented by multiple physical entities, or they may be jointly implemented by certain components in multiple independent devices.

[0122] The foregoing description of this application is provided to enable any person skilled in the art to implement or use the application. Various modifications to the application will be apparent to those skilled in the art, and the general principles defined herein can be applied to other variations without departing from the scope of protection of this application. Therefore, this application is not limited to the examples and designs described herein, but is consistent with the widest scope of the principles and novel features disclosed herein.

Claims

1. A method for determining vehicle control parameters, comprising: Obtain vehicle control instructions input by the user; Based on the vehicle control instruction information, the vehicle control intention and the corresponding intention intensity are determined, wherein the intention intensity is used to characterize the expected strength of the vehicle control intention; Obtain safety boundary information corresponding to the vehicle control intention, wherein the safety boundary information is used to indicate the boundaries of the safety constraints that the vehicle should comply with; as well as The target parameter values ​​of the vehicle control parameters associated with the vehicle control intention are determined based at least on the vehicle control intention, the intensity of the intention, and the safety boundary information.

2. The method as described in claim 1, wherein, The method further includes: Obtain driver status information. The step of determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based at least on the vehicle control intention, the intention strength, and the safety boundary information includes: Based on the vehicle control intention, the intensity of the intention, the safety boundary information, and the driver state information, the target parameter values ​​of the vehicle control parameters associated with the vehicle control intention are determined.

3. The method as described in claim 2, wherein, The step of determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based on the vehicle control intention, the intention strength, the safety boundary information, and the driver state information includes: Based on the driver status information, the vehicle control intention, and the intention intensity, it is determined whether the preset intention intensity correction condition is met; In response to satisfying the intent strength correction condition, the intent strength is corrected to obtain a corrected intent strength; and Based on the vehicle control intention, the modified intention strength, and the safety boundary information, target parameter values ​​for vehicle control parameters associated with the vehicle control intention are determined.

4. The method as described in any one of claims 1 to 3, wherein, The vehicle control intent is used to indicate the vehicle control parameters to be adjusted and the desired direction of parameter adjustment. The step of determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based at least on the vehicle control intention, the intention strength, and the safety boundary information includes: The parameter adjustment range is determined at least based on the current parameter values ​​of the vehicle control parameters to be adjusted, the desired adjustment direction of the parameters, and the safety boundary information; and The target parameter value of the vehicle control parameter to be adjusted is determined from the parameter adjustment range, at least based on the intensity of the intent.

5. The method of claim 4, wherein, The target parameter value of the vehicle control parameter to be adjusted is determined using a pre-built linear mapping model, at least based on the intent intensity. The linear mapping model is used to indicate the mapping relationship between the vehicle control parameter to be adjusted, the intent intensity, and the parameter adjustment range.

6. The method of claim 4, wherein, The step of determining the target parameter value of the vehicle control parameter to be adjusted from the parameter adjustment range, at least based on the intensity of the intent, includes: Based on the intent intensity and driver state information, the target parameter value of the vehicle control parameter to be adjusted is determined from the parameter adjustment range.

7. The method as described in any one of claims 1 to 3, wherein, The vehicle control intent is used to indicate the vehicle control parameters to be adjusted and the expected target values ​​for those parameters. The step of determining the target parameter value of the vehicle control parameter associated with the vehicle control intention based at least on the vehicle control intention, the intention strength, and the safety boundary information includes: Determine whether the expected target value of the parameter meets the preset safety conditions, wherein the safety conditions are determined at least based on the safety boundary information; In response to satisfying the safety condition, the expected target value of the parameter is determined as a target parameter value of the vehicle control parameter associated with the vehicle control intention; and In response to the failure to meet the safety conditions, target parameter values ​​for vehicle control parameters associated with the vehicle control intention are determined, at least based on the intent strength and the safety boundary information.

8. The method of claim 1, wherein, The vehicle control instruction information includes voice signals indicating the vehicle control intent. The step of determining the vehicle control intention and the corresponding intention strength based on the vehicle control instruction information includes: The speech signal is processed by speech recognition to obtain the speech-recognized text; Extract vehicle control intent from the speech-recognized text; Intent intensity indication information is extracted from the speech-recognized text and / or the speech signal, wherein the intent intensity indication information includes at least one of the following: modal particles, adjectives, degree adverbs, emotional tone, and acoustic emotion features; and The intent intensity corresponding to the vehicle control intent is determined based on the intent intensity indication information.

9. The method of claim 1, further comprising: In response to the discrepancy between the target parameter value and the vehicle control intent, explanatory information is generated at least based on the safety boundary information.

10. A method for controlling a vehicle, comprising: Obtain the target parameter value of the vehicle control parameter associated with the vehicle control intention determined by the method for determining vehicle control parameters according to any one of claims 1 to 9; Based on the target parameter value, output vehicle control prompt information corresponding to the vehicle control intention; as well as In response to the failure to receive a cancellation operation message corresponding to the vehicle control prompt within a preset time window, or the receipt of a confirmation operation message corresponding to the vehicle control prompt within a preset time window, the vehicle is controlled according to the target parameter value.

11. An apparatus for determining vehicle control parameters, comprising: The control information acquisition module is configured to acquire vehicle control instruction information input by the user; The intent determination module is configured to determine the vehicle control intent and the corresponding intent intensity based on the vehicle control instruction information, wherein the intent intensity is used to characterize the expected strength of the vehicle control intent. The safety information acquisition module is configured to acquire safety boundary information corresponding to the vehicle control intention, wherein the safety boundary information is used to indicate the boundaries of the safety constraints that the vehicle should comply with. as well as The parameter determination module is configured to determine target parameter values ​​for vehicle control parameters associated with the vehicle control intention, based at least on the vehicle control intention, the intention strength, and the safety boundary information.

12. A device for controlling a vehicle, comprising: The apparatus for determining vehicle control parameters as described in claim 11; The information output module is configured to output vehicle control prompt information corresponding to the vehicle control intention based on the target parameter value; as well as The vehicle control module is configured to control the vehicle according to the target parameter value in response to either not receiving a cancellation operation message corresponding to the vehicle control prompt message within a preset time window or receiving a confirmation operation message corresponding to the vehicle control prompt message within a preset time window.

13. A processing apparatus, comprising: processor; as well as The memory stores instructions that, when executed by the processor, cause the processor to perform operations in the method for determining vehicle control parameters as described in any one of claims 1 to 9 and / or the method for controlling a vehicle as described in claim 10.

14. A driving assistance system, comprising: Sensors are configured to acquire environmental perception information; as well as The processing apparatus of claim 13, wherein the security boundary information is determined at least based on the environmental perception information.

15. A computer program product comprising executable instructions that, when executed by a processor, cause the processor to perform operations in the method for determining vehicle control parameters as described in any one of claims 1 to 9 and / or the method for controlling a vehicle as described in claim 10.

Images