Automobile tail door sound control method and device and computer equipment

Abstract

The invention relates to an automobile tail door sound control method and device and computer equipment. The method comprises the following steps: in response to a control instruction of a target object for an automobile tail door, identifying instruction voiceprint information corresponding to the control instruction; under the condition that the instruction voiceprint information passes the voiceprint verification of the automobile tail door, sensing environment information corresponding to the automobile tail door is obtained; the control instruction and the sensing environment information are input into a tail door opening and closing control model corresponding to the automobile tail door, and opening and closing control parameters of the automobile tail door are obtained; according to the perceived environment information and the opening and closing control parameters, environment adaptation parameters of the automobile tail door are determined; and performing joint optimization on the opening and closing control parameter and the environment adaptation parameter to obtain a target tail gate control parameter corresponding to the control instruction. By adopting the method, the control precision of the automobile tail door can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent automobile technology, and in particular to a method, device and computer equipment for voice control of an automobile tailgate. Background Art

[0002] In traditional technology, the control technology of automobile tailgate mainly includes manual control, remote control key control, in-car button control and intelligent electric tailgate control. In particular, the intelligent electric tailgate system combines motor drive, sensor sensing (such as kick sensing, pressure sensing, position sensing, etc.) and control unit (ECU) to realize the automatic opening, closing and anti-pinch functions of the tailgate, improving the convenience and safety of users. However, the control of automobile tailgate in traditional technology has the disadvantages of high cost, sensor susceptibility to interference, limited reliability and low recognition accuracy in complex environments, resulting in insufficient control accuracy of automobile tailgate. Summary of the invention

[0003] Based on this, it is necessary to provide a car tailgate voice control method, device, computer equipment, computer readable storage medium and computer program product that can effectively improve the control accuracy of the car tailgate in order to solve the above technical problems.

[0004] In a first aspect, the present application provides a vehicle tailgate voice control method, comprising: In response to a control instruction of the target object for the tailgate of the car, identifying the command voiceprint information corresponding to the control instruction; When the command voiceprint information passes the voiceprint verification of the automobile tailgate, acquiring the perception environment information corresponding to the automobile tailgate; Inputting the control instruction and the perceived environment information into a tailgate opening and closing control model corresponding to the tailgate of the vehicle to obtain the opening and closing control parameters of the tailgate of the vehicle, including: According to the control instruction and the perceived environment information, the target object performs control intention recognition on the automobile tailgate to obtain tailgate control intention recognition data; According to the tailgate control intention recognition data and the perceived environment information, the motion parameters of the automobile tailgate are calculated to obtain initial control parameters, including: According to the tailgate control intention recognition data and the perceived environment information, a tailgate scene coupling model of the automobile tailgate is set; Inputting the tailgate control intention recognition data and the perceived environment information into the tailgate scene coupling model to obtain model calculation control parameters; Performing hyperstatic optimization on the model calculation control parameters to obtain the initial control parameters; According to the mechanical control error of the automobile tailgate, the initial control parameter is adjusted to obtain the opening and closing control parameter; Determining environmental adaptation parameters of the automobile tailgate according to the perceived environmental information and the opening and closing control parameters; The opening and closing control parameters and the environmental adaptation parameters are jointly optimized to obtain target tailgate control parameters corresponding to the control instructions.

[0005] In a second aspect, the present application also provides a car tailgate voice control device, comprising: A voiceprint information recognition module, used to respond to a control instruction of a target object for a car tailgate and recognize the command voiceprint information corresponding to the control instruction; An environment information acquisition module, used for acquiring the perception environment information corresponding to the tailgate of the vehicle when the command voiceprint information passes the voiceprint verification of the tailgate of the vehicle; A control parameter obtaining module, used to input the control instruction and the perceived environment information into a tailgate opening and closing control model corresponding to the tailgate of the vehicle to obtain the opening and closing control parameters of the tailgate of the vehicle, including: According to the control instruction and the perceived environment information, the target object performs control intention recognition on the automobile tailgate to obtain tailgate control intention recognition data; According to the tailgate control intention recognition data and the perceived environment information, the motion parameters of the automobile tailgate are calculated to obtain initial control parameters, including: According to the tailgate control intention recognition data and the perceived environment information, a tailgate scene coupling model of the automobile tailgate is set; Inputting the tailgate control intention recognition data and the perceived environment information into the tailgate scene coupling model to obtain model calculation control parameters; Performing hyperstatic optimization on the model calculation control parameters to obtain the initial control parameters; According to the mechanical control error of the automobile tailgate, the initial control parameter is adjusted to obtain the opening and closing control parameter; An adaptation parameter obtaining module, used to determine the environmental adaptation parameters of the automobile tailgate according to the perceived environmental information and the opening and closing control parameters; The control parameter optimization module is also used to jointly optimize the opening and closing control parameters and the environmental adaptation parameters to obtain the target tailgate control parameters corresponding to the control instructions.

[0006] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements any step of a method for voice control of a car tailgate when executing the computer program.

[0007] The above-mentioned method, device and computer equipment for controlling the tailgate of a car ensure that the tailgate control command comes from a legitimate user by introducing the recognition and verification mechanism of the command voiceprint information, thereby improving the security of the system from the source; after the verification is passed, the current perceived environmental information of the vehicle is further obtained, such as the distance of the rear obstacle, the terrain slope, the weather conditions, etc., to provide a comprehensive environmental reference basis for the tailgate control; the control command and the environmental information are input into the trained tailgate opening and closing control model to obtain the preliminary opening and closing control parameters, and the adaptation parameters under the current environment are combined for joint optimization, and finally the target tailgate control parameters that meet the user's intention and are highly matched with the actual environment are output. It can not only effectively improve the control accuracy and intelligence level of the car tailgate, but also effectively reduce the probability of problems such as tailgate collision, object clamping or failure to open normally caused by environmental incompatibility, thereby comprehensively improving the safety, adaptability and user experience of the vehicle tailgate system. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 A diagram showing an application environment of a method for controlling a car tailgate voice in one embodiment; Figure 2 A schematic diagram of a flow chart of a method for voice-controlling a car tailgate in one embodiment; Figure 3 A schematic flow chart of a method for obtaining a switching control parameter in one embodiment; Figure 4 A schematic flow chart of a first method for obtaining initial control parameters in an embodiment; Figure 5 Schematic diagram of a flow chart of a second method for obtaining initial control parameters in one embodiment; Figure 6 A schematic diagram of a flow chart of a method for obtaining a first environment adaptation parameter in an embodiment; Figure 7 It is a flowchart of a method for obtaining a second environment adaptation parameter in an embodiment; Figure 8 is a flow chart of a method for calculating a tailgate avoidance parameter and an object warning parameter in one embodiment; Fig. 9 is a structural block diagram of a car tailgate voice control device in one embodiment; Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0011] The embodiment of the present application provides a vehicle tailgate voice control method, which can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0012] In an exemplary embodiment, Figure 2 As shown, a method for controlling the tailgate of an automobile is provided. Figure 1 The server in the example is used to illustrate, including the following steps 202 to 210. Among them: Step 202: In response to a control instruction of the target object for the tailgate of the automobile, identify the command voiceprint information corresponding to the control instruction.

[0013] The target object may be a subject that issues a control command to operate the tailgate of a car, usually a legitimate user or an authorized person of the vehicle.

[0014] Among them, the control instruction can be a voice command issued by the target object to the vehicle regarding the tailgate operation, such as "open the tailgate", "close the tailgate" or "stop opening".

[0015] The command voiceprint information may be voiceprint feature data extracted from the voice control command issued by the target object, which contains the audio features of the speaker (such as timbre, frequency, pronunciation habits, etc.) and is used to uniquely identify the speaker.

[0016] Specifically, when the target object (such as the vehicle user) issues a control command for the car tailgate (such as "open the tailgate" or "close the tailgate") by voice, the in-vehicle voice interaction system first accurately collects the voice signal through the built-in microphone array, and performs noise reduction and voice enhancement processing on it to improve the quality of voice recognition. Then, voiceprint recognition technology is used to deeply analyze the speaker's characteristics in the voice signal, extract biometric information such as its vocal tract structure, voice frequency, pronunciation habits, etc., and generate command voiceprint information that uniquely corresponds to the voice command. This command voiceprint information not only contains the content information of the control command, but is also used to compare the authorized user voiceprint template library pre-stored in the vehicle to achieve preliminary confirmation of the identity of the user who issued the command.

[0017] Step 204, when the command voiceprint information passes the voiceprint verification of the car tailgate, the perception environment information corresponding to the car tailgate is obtained.

[0018] Among them, voiceprint verification can be to compare the command voiceprint information collected in real time with the authorized user voiceprint template pre-stored in the vehicle to confirm whether the control command comes from a registered or authorized user.

[0019] Among them, the perceived environmental information can be the tailgate surrounding environmental status data collected in real time by the vehicle through various sensors (such as cameras, radars, slope sensors, weather sensors, etc.), including whether there are obstacles at the rear, whether the vehicle is parked on a slope, whether the current weather is bad, etc., to assist in determining whether the tailgate has safe operating conditions.

[0020] Specifically, when it is confirmed that the command voiceprint information of the target object has passed the voiceprint verification of the car tailgate, that is, after it is confirmed that the voice command comes from an authorized user of the vehicle, the process of collecting perceived environmental information will be automatically started, that is, the vehicle uses a variety of integrated environmental perception devices, such as ultrasonic radar, millimeter-wave radar, reversing image camera, tailgate position sensor, accelerometer, barometer, inclinometer, temperature and humidity sensor, etc., to monitor the rear of the car and its surrounding environment in real time, and then perceives whether there are obstacles (such as walls, people, objects) behind the tailgate, the current ground slope, the vehicle posture, the size of the rear space, weather conditions (such as whether it is raining or snowing), and the temperature difference between the inside and outside of the car based on the environmental collection data, to form the perceived environmental information corresponding to the car tailgate.

[0021] Step 206 , input the control instruction and the perceived environment information into a tailgate opening and closing control model corresponding to the tailgate of the vehicle to obtain the opening and closing control parameters of the tailgate of the vehicle.

[0022] Among them, the tailgate opening and closing control model can be an intelligent decision-making model trained based on a large amount of user operation data and environmental feedback, which can intelligently determine the tailgate opening and closing method, angle, speed and other parameters based on the input control instructions and perceived environmental information. This model is usually constructed through machine learning, neural networks or rule systems to achieve automation and intelligence of tailgate behavior.

[0023] Among them, the opening and closing control parameters can be specific action parameters output by the tailgate opening and closing control model, such as the opening or closing angle of the tailgate, the movement speed, the motor driving force, whether to enable the slow start or slow stop mechanism, etc.

[0024] Specifically, the control command issued by the user (such as "open the tailgate") and the perceived environmental information are input together into the pre-built and trained tailgate opening and closing control model. The tailgate opening and closing control model evaluates whether the current conditions for safe and feasible tailgate opening and closing are met through a comprehensive analysis of the intention recognition of the input command and the environmental characteristics, and calculates the corresponding tailgate opening and closing control parameters accordingly. These control parameters include the opening or closing angle of the tailgate, the opening and closing speed, the motor driving force required during the opening and closing process, the execution delay time, whether the obstacle buffering strategy is enabled, etc., so as to meet the optimal execution effect under specific environments and user needs.

[0025] Step 208, determining the environmental adaptation parameters of the tailgate of the vehicle according to the perceived environmental information and the opening and closing control parameters.

[0026] Among them, the environmental adaptation parameters can be the adjustment amounts introduced when the opening and closing control parameters are corrected according to the perceived environmental information, such as limiting the maximum door opening angle to avoid collision, reducing the opening and closing speed on ramps, and enhancing buffering strategies in rainy and snowy days.

[0027] Specifically, based on the tailgate opening and closing control parameters generated in the previous step and the perceived environmental information of the current vehicle tailgate, it is further identified whether there are special environmental factors such as the obstacle behind the tailgate is too close, the vehicle is in an uphill or downhill position, the space above the tailgate is limited, the wind is strong or it is raining or snowing, and the environmental adaptation parameters that need to be adapted to the tailgate in the current situation are dynamically calculated, such as the maximum allowable opening angle, the lower or upper limit of the opening and closing speed, the slow start or slow closing time setting, the activation of the anti-pinch and anti-collision mechanism, the sensor sensitivity adjustment, etc.

[0028] Step 210 , jointly optimize the opening and closing control parameters and the environmental adaptation parameters to obtain target tailgate control parameters corresponding to the control instructions.

[0029] Among them, joint optimization can be the opening and closing control parameters and environmental adaptation parameters as the overall input, and the two are coordinated and weighed through a multi-objective optimization algorithm to ensure that the final output control strategy reaches the global optimal state among multiple performance indicators (such as safety, efficiency, user experience, etc.), thereby achieving the optimal execution effect of the tailgate operation.

[0030] Among them, the target tailgate control parameters can be a set of final control instruction parameters obtained through joint optimization, which comprehensively considers the user's operation intention, environmental constraints and vehicle operation safety, and is used to guide the tailgate to perform opening and closing actions.

[0031] Specifically, after the tailgate opening and closing control parameters (including opening and closing angle, opening and closing speed, motor driving torque, execution delay, etc.) and environmental adaptation parameters (such as obstacle safety distance limit, slope road condition correction coefficient for maximum opening angle, speed buffer adjustment in windy and rainy weather, etc.) are input into the joint optimization model as joint optimization variables, the joint optimization model constructs a multi-dimensional parameter space and defines multiple optimization objective functions, such as maximizing operational safety, minimizing energy consumption, optimizing user waiting time, improving tailgate movement stability, etc. At the same time, dynamic constraints are set according to different scene requirements, such as prohibiting full opening operation when the obstacle distance is less than the threshold, or reducing the tailgate movement speed when the slope is large. During the optimization process, the joint optimization model uses a multi-objective weighted method to assign different weights to each optimization objective in order to find the global optimal solution; it can also introduce a reinforcement learning algorithm at the same time to continuously update the control strategy through real-time interaction with the environment; or use a genetic algorithm to iteratively select and evolve among multiple control strategies, gradually approaching the optimal solution, and obtaining the optimal and executable tailgate control parameter combination under the current environment and user intention.

[0032] In the above-mentioned method for controlling the tailgate of a car, the recognition and verification mechanism of the command voiceprint information is introduced to ensure that the tailgate control command comes from a legitimate user, thereby improving the safety of the system from the source; after the verification is passed, the current perceived environmental information of the vehicle is further obtained, such as the distance of the rear obstacle, the terrain slope, the weather conditions, etc., to provide a comprehensive environmental reference basis for the tailgate control; the control command and the environmental information are input into the trained tailgate opening and closing control model to obtain the preliminary opening and closing control parameters, and the adaptation parameters under the current environment are combined for joint optimization, and finally the target tailgate control parameters that meet the user's intention and are highly matched with the actual environment are output. It can not only effectively improve the control accuracy and intelligence level of the car tailgate, but also effectively reduce the probability of problems such as tailgate collision, object clamping or failure to open normally caused by environmental incompatibility, thereby comprehensively improving the safety, adaptability and user experience of the vehicle tailgate system.

[0033] In an exemplary embodiment, Figure 3As shown, the control instruction and the perceived environment information are input into the tailgate opening and closing control model corresponding to the tailgate of the vehicle to obtain the opening and closing control parameters of the tailgate of the vehicle, including steps 302 to 306. Among them: Step 302 , based on the control instruction and the perceived environment information, the control intention of the target object for the tailgate of the car is identified to obtain the tailgate control intention identification data.

[0034] Among them, control intention recognition can be to determine the specific type and degree of operation that the user wants the tailgate to perform through semantic understanding, context analysis and multimodal information fusion after receiving the user's voice control command and the vehicle's current perceived environment information. For example, the system can distinguish whether the user wants to open the tailgate completely, only partially, or close it slowly, and further infer the user's real needs based on environmental factors (such as whether the space is limited).

[0035] Among them, the tailgate control intention recognition data can refer to the structured data results output by the system after the control intention recognition is completed, which represents the specific goal and behavioral intention of the user to control the tailgate. This data usually contains parameter information such as intention type (such as full opening, half opening, closing), action mode (such as slow opening, fast opening and closing), target angle range, operation priority, etc., which serves as the key input for the tailgate control model to generate motion parameters.

[0036] Specifically, after receiving the control command issued by the target object (such as "open the tailgate" or "open the trunk a little") and the corresponding perceived environmental information, the voice command is parsed using speech recognition and semantic understanding technology to extract the key elements of the user's operating intention; combined with the vehicle's current perceived environmental information (such as whether the tailgate is closed, whether the vehicle is on a ramp, whether there are obstacles at the rear, etc.), the specific control intention of the target object is judged through a multimodal intention recognition model. For example, based on the "open a little" in the command and the perceived data of insufficient rear space, the system can infer that the user's intention is to partially open the tailgate to place small items, rather than fully open it to carry large items. In the recognition process, semantic features, user historical behavior, scene context and environmental constraints are also integrated to ultimately output the tailgate control intention recognition data.

[0037] Step 304 , calculating the motion parameters of the tailgate of the vehicle according to the tailgate control intention recognition data and the perceived environment information to obtain initial control parameters.

[0038] Among them, the initial control parameters can be the first set of specific control instructions calculated by the tailgate opening and closing control model after analyzing the user's control intention and current environmental information, including a preset tailgate opening and closing angle, motor output torque, opening and closing speed, acceleration curve, etc.

[0039] Specifically, the required tailgate action type (such as full opening, partial opening, slow closing, etc.) is determined based on the tailgate control intention recognition data, and then combined with the current perceived environmental information (such as the distance of obstacles behind the tailgate, the body posture, the current position of the tailgate and weather conditions), the physical modeling algorithm or the data-driven machine learning module in the tailgate opening and closing control model is called to perform fine modeling and simulation prediction on the control behavior to ensure that the tailgate action has good executableness and environmental adaptability within the scope of the user's intention, and to obtain the initial control parameters.

[0040] Step 306, adjusting the initial control parameters according to the mechanical control error of the automobile tailgate to obtain the opening and closing control parameters.

[0041] Among them, the mechanical control error can be the deviation between the actual motion trajectory and the control model expectation caused by factors such as response delay, structural wear, component tolerance, temperature change, etc. in the vehicle hardware system during the actual opening and closing of the tailgate. In order to ensure the accuracy and reliability of the tailgate action, the system needs to perceive these errors in real time and make dynamic compensation adjustments to make the tailgate action more consistent with the initial control target and adapt to the actual mechanical state.

[0042] Specifically, due to the mechanical control errors that may occur during the actual execution of the tailgate, these errors may come from motor response lag, mechanical transmission clearance, tailgate hinge wear, vehicle posture changes, the influence of ambient temperature on material expansion, etc. Therefore, by real-time monitoring of the feedback information during the movement of the tailgate, such as the data of the tailgate position sensor, motor current and voltage changes, speed feedback signals, etc., the deviation between the actual execution effect and the initial preset parameters is dynamically evaluated. Then, according to the established error compensation model or adaptive control algorithm, the key parameters such as opening and closing angle, speed, torque, etc. are fine-tuned to ensure that the tailgate movement process is more accurate and smooth, avoid overshoot, stagnation or insufficient force, and obtain the opening and closing control parameters after eliminating mechanical errors.

[0043] In this embodiment, by introducing a tailgate control intention recognition mechanism based on control instructions and perceived environmental information, the user's real needs for tailgate operation can be accurately analyzed, and the tailgate motion parameters can be intelligently calculated in combination with environmental factors to generate initial control parameters that meet the current usage scenario; at the same time, dynamic corrections are further made to the mechanical control errors existing in the actual operation of the car tailgate, thereby outputting accurate and stable opening and closing control parameters. This not only improves the response accuracy and intelligence level of the tailgate action, but also enhances the execution reliability of the tailgate in complex environments, effectively reduces the risk of misoperation and equipment wear, and comprehensively improves the safety, adaptability and user experience of the vehicle tailgate control system.

[0044] In an exemplary embodiment, Figure 4As shown, according to the tailgate control intention recognition data and the perceived environment information, the motion parameters of the car tailgate are calculated to obtain the initial control parameters, including steps 402 to 406. Among them: Step 402 , setting a tailgate scene coupling model of the car tailgate according to the tailgate control intention recognition data and the perceived environment information.

[0045] Among them, the tailgate scene coupling model can be an intelligent decision-making model for jointly modeling the user's control intention and the current environmental state of the vehicle. It analyzes the complex relationship between the tailgate control intention recognition data and the perceived environmental information, establishes a coupling mapping between multi-dimensional variables, and is used to determine the reasonable opening and closing strategy of the tailgate in specific scenarios (such as obstacles approaching, ramp parking, rainy and snowy weather, etc.).

[0046] Specifically, since the tailgate scene coupling model is used to integrate the complex correlation between the user's control intention and the vehicle's current environment, it is necessary not only to consider the specific actions that the user wants the tailgate to perform (such as full opening, half opening, slow opening, fast closing, etc.), but also to couple and analyze multiple variable factors in the environment, such as the size of the space behind the tailgate, the location of obstacles, the slope of the terrain, the wind speed and direction, the temperature and humidity, the vehicle posture, etc. In order to achieve this multi-dimensional factor coupling modeling, state space modeling, fuzzy logic rule base, neural network model or reinforcement learning framework are used to transform the tailgate control intention recognition data and the perceived environment information into the input feature space of the tailgate control parameters, and the weight relationship is set through historical data training or expert experience rules to construct a dynamic decision model that can reflect the tailgate response behavior under various scene conditions as the tailgate scene coupling model.

[0047] Step 404 , input the tailgate control intention recognition data and the perceived environment information into the tailgate scene coupling model to obtain the model calculation control parameters.

[0048] Among them, the model calculation control parameters can be a set of tailgate motion control variables deduced and output by the model after the tailgate control intention recognition data and the perceived environment information are input into the tailgate scene coupling model. These parameters usually include the angle range of the tailgate expected to open or close, the opening and closing speed, the acceleration curve, the motor output torque, the action delay, etc., which are idealized control schemes calculated by the system under the matching relationship between specific intentions and the environment, and are used to guide the subsequent control optimization and tailgate execution actions.

[0049] Specifically, the tailgate scene coupling model is used as the calculation core, and the tailgate control intention recognition data and perceived environmental information are injected into the tailgate scene coupling model as input variables for joint processing. The tailgate scene coupling model extracts features, recognizes patterns, and maps behaviors of the input data inside the model. According to the scene logic relationship, parameter constraints, and coupling rules set in the model, the ideal motion state of the tailgate is calculated and deduced, thereby outputting a set of model calculation control parameters. These parameters usually include the estimated opening and closing angle, motion speed, acceleration and deceleration curve, maximum output torque, motor response strategy, etc., which are used to describe how the tailgate should move under the current user intention and environmental scenario to achieve a control effect that meets the operation goal and avoids potential risks.

[0050] Step 406: Perform hyperstatic optimization on the model calculation control parameters to obtain initial control parameters.

[0051] Among them, parameter hyperstatic optimization can be the process of further mathematical optimization of these parameters after the model preliminarily calculates the tailgate control parameters to solve problems such as environmental changes, redundant constraints, multi-objective conflicts or mechanical structure response uncertainties. The optimization process usually introduces multiple constraints and optimization goals (such as safety, response time, energy consumption control, etc.), and uses technologies such as linear programming, genetic algorithms, and particle swarm optimization to eliminate redundant variables, balance various performance indicators, and output a set of optimal parameters that meet physical feasibility, control stability and execution efficiency.

[0052] Specifically, in order to convert the model calculation control parameters obtained by the model calculation into tailgate action instructions that can be actually executed and have stability, it is necessary to perform parameter hyperstatic optimization on these model calculation control parameters. In the case where the model calculation control parameters output by the model may not adapt to environmental changes, have redundancy, physical infeasibility, or not fully match the actual mechanical structure of the vehicle, by establishing an optimization framework containing multiple constraints, introducing methods such as redundant degree of freedom elimination, multi-objective weighted balance, and dynamic response stability analysis, the control parameters are finely adjusted and screened. During the specific implementation process, the detection model calculates whether there is excessive coupling or conflict between control instructions in the control parameters (such as requiring high speed and low energy consumption at the same time), or detects whether the nearby environment has changed from the perceived environmental information. When the detection result indicates that there is excessive coupling or conflict, or when changes in the nearby environment are detected, mathematical optimization algorithms (such as linear programming, constraint solver, genetic algorithm, etc.) are used to reconstruct and optimize the parameters so that they meet multiple goals such as safety, execution efficiency, and response stability while having high executableness, mechanical coordination, and risk resistance. The final initial control parameters are obtained, which can directly drive the optimized instruction set executed by the tailgate, providing a stable control basis for subsequent error compensation and precise motion control of the tailgate.

[0053] In one embodiment, the expression of the tailgate scene coupling model is: in, is the angular acceleration The moment of inertia of the tailgate around the axis of rotation is calculated by integration or CAD software based on the tailgate structure size and mass distribution. is the angular acceleration that changes with time t, obtained by differentiating the increment of the angle sensor, is the angular velocity that changes with time t, which is also obtained by differentiating the increment of the angle sensor, is the angle of the tailgate relative to the initial closed position, measured in real time by a sensor (such as a rotary encoder or angle sensor). is the motor torque constant, provided by the motor manufacturer or calibrated experimentally. is the motor current varying with time t, measured by the sensor, With angular velocity The damping torque that changes with time t is obtained through statistics of a large number of tailgate rotation experiments. is the gravity moment, obtained through CAD / CAE analysis, is the external environmental torque that changes with time t, which is calculated based on the wind prediction model, sensor detection of obstacles / external forces, or digital twins. is the temperature of the motor and its transmission components at time t, measured by the temperature sensor, is the equivalent heat capacity of the motor and related mechanical components, calculated from the component mass, material specific heat capacity and structural combination. is the temperature-dependent resistance of the motor winding, which is determined by the motor manufacturer's data sheet and experimental calibration. The thermal power generated by mechanical energy loss is obtained by converting the friction torque calculated in the mechanical dynamics model into thermal power. is the heat dissipation power, obtained through heat conduction / convection calculation or empirical formula, is the convective heat transfer coefficient, determined by the manufacturer through static or dynamic testing, is the ambient temperature, measured by the sensor, is the motor output torque that changes with time t, obtained by multiplying the torque constant by the current, is the motor input voltage that changes with time t, which is obtained by monitoring the vehicle power system or inverter. k is the motor back electromotive force constant, which is determined by the motor structure / coil turns and is marked on the motor. L is the motor armature inductance, which is determined by the motor structure / coil turns and is marked on the motor.

[0054] In this embodiment, by constructing a tailgate scene coupling model, the tailgate control intention recognition data and the perceived environmental information are deeply integrated and modeled to achieve dynamic adaptation of vehicle tailgate operations in various complex scenarios; the above data is further input into the model, and the model calculation control parameters matching the current scene are intelligently output to effectively improve the accuracy of tailgate action decisions; then the parameter hyperstatic optimization mechanism is introduced to perform multi-objective constraint adjustment and redundant parameter correction on the model output to generate more stable and reasonable initial control parameters. It can significantly improve the adaptive ability, decision-making accuracy and execution stability of the tailgate control system in practical applications, enhance the system's fault tolerance to environmental changes, and thus comprehensively optimize the safety and intelligence level of tailgate operations.

[0055] In an exemplary embodiment, Figure 5 As shown, the model calculation control parameters are subjected to hyperstatic optimization to obtain initial control parameters, including steps 502 to 506. Among them: Step 502, predicting the environmental change of the tailgate of the vehicle based on the perceived environmental information, and obtaining environmental mutation prediction data.

[0056] Among them, the environmental mutation prediction data can be a set of result data obtained by analyzing the drastic changes that may occur in the environment around the tailgate in a short period of time through a prediction algorithm based on the current perceived environmental information. These prediction data reflect the adverse factors that may occur in the future, such as the sudden approach of obstacles, sudden changes in vehicle posture (for example, from flat ground to slope), sudden weather changes (such as heavy rain or strong wind), etc., aiming to provide an early warning mechanism for the tailgate control strategy so that the system can adjust the control parameters in advance to avoid risks.

[0057] Specifically, based on the current perceived environmental information, the real-time data around the tailgate of the car is analyzed, such as changes in obstacle distance, vehicle posture, terrain slope, air pressure and humidity, wind speed and direction, temperature fluctuations, etc., combined with historical environmental data and the tailgate usage scenario model, using time series prediction, deep learning (such as LSTM neural network) or Bayesian inference algorithms, to predict whether the environmental state around the tailgate will change significantly in a short period of time. For example, the system can predict whether there are pedestrians or objects approaching behind the tailgate, or evaluate whether the tailgate opening angle needs to be restricted when it is about to rain. The prediction results are output in the form of structured "environmental mutation prediction data".

[0058] Step 504 , redundancy identification is performed on the model calculation control parameters and the environmental mutation prediction data to obtain the hyperstatic adjustment data of the vehicle tailgate.

[0059] Redundancy identification can be a process of comparing and analyzing the environmental mutation prediction data with the model-calculated control parameters, with the goal of identifying those parts of the control parameters that become invalid, redundant, or no longer safe in the upcoming environmental changes. For example, when a large opening angle is set in the control parameters, but the environmental prediction shows that an obstacle will soon appear behind the tailgate, the angle setting is judged to be redundant or high risk.

[0060] Among them, the hyperstatic adjustment data can be a set of intermediate optimization data generated by quantitatively correcting the parameters to be optimized according to the analysis results after completing the redundancy identification. These data usually include the specific control parameters to be adjusted (such as angle, speed, torque), the recommended adjustment range, priority level and corresponding constraints, which are used to guide the system to accurately reconstruct the model calculation control parameters to ensure that the final generated control strategy still has a high degree of stability and execution reliability in a complex and changing environment.

[0061] Specifically, a redundant identification algorithm is used to analyze the coupling relationship between the degrees of freedom in the model-calculated control parameters and the environmental mutation prediction data to determine whether there is an under-response or over-response to the sudden change environment. For example, if it is predicted that an obstacle is approaching behind the tailgate, and the door opening angle calculated by the model remains fully open, the control parameter is judged to be redundant or high-risk. If the judgment result indicates that there is an under-response or over-response to the sudden change environment, the parameters that need to be adjusted (such as angle, speed, torque) will be extracted based on the judgment result, and the adjustment range and priority of these parameters will be quantified in combination with the severity and time sensitivity of the environmental changes, and hyperstatic adjustment data for optimizing control instructions will be generated.

[0062] Step 506: Adjust the model calculation control parameters according to the hyperstatic adjustment data to obtain initial control parameters.

[0063] Specifically, based on the hyperstatic adjustment data, a multi-dimensional optimization algorithm (such as nonlinear optimization with constraints, multi-objective weighted adjustment, or adaptive parameter adjustment mechanism based on control feedback) is used to dynamically correct the key control variables such as the tailgate opening and closing angle, speed, torque, and execution delay, so that the control instructions can better adapt to possible environmental mutations. For example, the system may adjust the originally planned door opening angle from full opening to partial opening, or reduce the opening speed to avoid collision or entrapment risks based on sudden obstacle approach information. The entire adjustment process ensures the safety of the tailgate execution while retaining the user's original operating intention and ease of use to the maximum extent, and obtains the initial control parameters.

[0064] In this embodiment, by dynamically analyzing the perceived environmental information, the possible environmental mutations around the tailgate are predicted, and the environmental mutation prediction data is obtained in advance, so that the system has the forward-looking risk prediction capability; the control parameters are calculated in combination with the model, and redundancy identification is further performed to extract the redundant control items that are not adapted to future environmental changes, and generate hyperstatic adjustment data for optimization and adjustment; finally, through fine adjustment of the control parameters, more stable, controllable, and initial control parameters that conform to the environmental evolution trend are obtained. It can effectively improve the adaptive and anti-interference capabilities of the tailgate control system in a dynamic and complex environment, enhance the safety and robustness of the tailgate operation, avoid malfunctions or abnormal operation caused by environmental changes, and significantly optimize the intelligent level of the tailgate control and the user experience.

[0065] In an exemplary embodiment, Figure 6 As shown, according to the perceived environmental information and the opening and closing control parameters, the environmental adaptation parameters of the tailgate of the automobile are determined, including steps 602 to 606. Among them: Step 602: Identify pedestrian movement information and other vehicle movement information from the perceived environment information.

[0066] Among them, pedestrian movement information can be real-time dynamic behavior data of pedestrians near the tailgate identified and tracked by the vehicle-mounted sensors, including the pedestrian's current position, moving speed, movement direction, acceleration, stay time, and whether it is approaching the tailgate.

[0067] Among them, the other vehicle motion information may be the motion status data of other vehicles detected in the area around the tailgate except the vehicle itself, including the driving speed, relative position, direction of travel, acceleration, whether in reversing or approaching mode, etc. of other vehicles.

[0068] Specifically, image recognition and target detection algorithms (such as YOLOv5, Faster R-CNN) are used to identify pedestrians and vehicles in the perceived environment information, and then target tracking algorithms (such as Kalman filtering, SORT or DeepSORT) are used to track the position, speed, acceleration and movement direction of these targets in real time. At the same time, key behavioral features such as the relative distance between the target and the tailgate, approach trend and existence time are extracted to determine whether these moving objects are in a potential interference area, and obtain the movement information of people and other vehicles.

[0069] Step 604: predict the object motion envelope corresponding to the moving object based on the motion information of the person and the motion information of the other vehicle.

[0070] Among them, the moving object can be any entity with relative motion capability, including pedestrians, other vehicles, bicycles, pets and even small robots. Due to their dynamic properties, these objects may interfere with or threaten the normal opening and closing action of the tailgate.

[0071] Among them, the object motion envelope can be a prediction of the spatial range in which the object may move in a short period of time in the future based on the motion state of the moving object at the current moment (such as position, speed, acceleration, direction, etc.).

[0072] Specifically, after completing the recognition of the motion state of pedestrians and other vehicles, the spatial behavior of these moving objects at future moments is further predicted, that is, the spatial range that they may pass through or occupy within a period of time. In the prediction process, the current target's speed, acceleration, direction of movement, and historical trajectory data are used in combination with the spatial structure of the tailgate and the vehicle's own motion state, and the trajectory prediction algorithm (such as LSTM-based time series model, Bayesian filtering, or social force model) is used to fit and extrapolate the future movement paths of these moving objects; the prediction result is not just a single trajectory point, but an uncertain area with a spatial extension boundary, indicating the entire area where the target may move in the next few seconds. When predicting, factors such as the prediction error range and the possibility of sudden behavior (such as pedestrians turning suddenly) are also considered to expand or adjust the envelope range, thereby obtaining the object's motion envelope.

[0073] Step 606: Generate environmental adaptation parameters for the tailgate of the automobile according to the opening and closing control parameters and the object motion envelope.

[0074] Specifically, the tailgate opening and closing control parameters (such as opening angle, movement speed, torque output, etc.) and the object motion envelope are cross-analyzed in space and time to evaluate whether the tailgate opening and closing process may overlap or interfere with the path of the moving object. If it is detected that the opening and closing trajectory of the tailgate may invade the motion envelope area at a specific moment, the vehicle with a faster speed and approaching direction directly facing the tailgate is considered to have a higher risk level based on the specific type of the moving object (such as pedestrians or vehicles), the current speed, the direction of approaching the tailgate, and the risk level assessed by the system for the target. Therefore, the generated environmental adaptation parameters will limit the maximum opening angle of the tailgate, delay the opening time, or directly suspend the opening action; while for pedestrians who are approaching but moving slowly or passing laterally, the door opening speed can be appropriately reduced, a slow start action can be added, or a higher frequency environmental monitoring mechanism can be enabled. In the case of multiple targets, a multi-target dynamic avoidance strategy is formulated by integrating the behavior patterns of multiple targets, so that the tailgate can achieve safe and flexible response control in a complex and dynamic environment. Finally, the adjustment results are output in the form of structured environmental adaptation parameters to obtain environmental adaptation parameters.

[0075] In this embodiment, by accurately identifying the movement information of pedestrians and other vehicles from the perceived environmental information, the dynamic risk sources around the tailgate can be perceived in real time; further, based on these movement information, the movement envelope of the moving object is predicted, and its possible movement range in the future is grasped in advance, so as to realize the dynamic prediction of potential collision risks; finally, the tailgate behavior is intelligently adjusted in combination with the current opening and closing control parameters of the tailgate, and the environmental adaptation parameters are generated, so as to dynamically optimize the opening and closing strategy of the tailgate. It can significantly improve the active perception and obstacle avoidance capabilities of the tailgate system in complex environments, enhance the safety, adaptability and accuracy of the tailgate operation, avoid interference with people or vehicles, and comprehensively improve the user experience and vehicle intelligence level.

[0076] In an exemplary embodiment, Figure 7 As shown, according to the opening and closing control parameters and the object motion envelope, the environmental adaptation parameters of the car tailgate are generated, including steps 702 to 706. Among them: Step 702, calculating the envelope overlap information of the tailgate of the automobile and the moving object according to the opening and closing control parameters and the object motion envelope.

[0077] Among them, the envelope overlap information can be the specific data of the intersection of the two within a certain time period obtained by superimposing the predicted opening and closing trajectory of the tailgate and the future motion envelope of the moving object in space and time. This information includes indicators such as the position, volume, duration, contact point, and relative speed of the overlapping area, which are used to determine whether there is a collision risk for the tailgate.

[0078] Specifically, the opening and closing trajectory predicted by the current opening and closing control parameters of the car tailgate is modeled in three dimensions to form a spatial representation of the dynamic opening and closing range; at the same time, the motion envelope of the moving object in the next few seconds is mapped into a spatial region with time-series attributes. Using geometric calculations and time-space intersection algorithms (such as AABB bounding box detection, time-series frame comparison, etc.), it is analyzed whether the two have an intersection in a specific time period, and the key data such as the volume, duration, contact position, relative speed, etc. of the intersection are extracted to form envelope overlap information.

[0079] Step 704 , when the envelope overlap information indicates that the tailgate of the vehicle is about to collide with the moving object, the tailgate avoidance parameter and the object warning parameter are calculated according to the envelope overlap information.

[0080] Among them, the tailgate avoidance parameters can be a set of control adjustment data generated to avoid actual contact when a potential collision risk between the tailgate and a moving object is detected. These parameters include reducing the tailgate opening angle, reducing the opening and closing speed, delaying the tailgate action, adjusting the opening path or switching to a slow start mode, etc.

[0081] Among them, the object warning parameters can be a set of control adjustment data generated to avoid actual contact when a potential collision risk is detected between the tailgate and a moving object. These parameters include reducing the tailgate opening angle, reducing the opening and closing speed, delaying the tailgate action, adjusting the door opening path or switching to a slow start mode.

[0082] Specifically, when the envelope overlap information indicates that the opening and closing path of the tailgate will overlap with the motion envelope of the moving object in a certain period of time, and there is a risk of collision or interference, the risk response mechanism will be immediately activated to calculate the "tailgate avoidance parameters" and "object warning parameters" respectively. The calculation of the tailgate avoidance parameters is based on the overlap characteristics of the current tailgate opening and closing plan and the envelope in the envelope overlap information, combined with the object approach speed, overlap duration and contact position, and the control plan of the tailgate is dynamically adjusted through the control strategy library or optimization algorithm, such as reducing the maximum opening angle, reducing the door opening speed, increasing the opening and closing delay, enabling the slow start mechanism or suspending the operation execution, so as to avoid the impending collision from the action level. At the same time, the object warning parameters are calculated according to the object type (such as pedestrians or vehicles), risk level and approach direction, such as starting the sound and light alarm, flashing lights in the tailgate area, vehicle external horn reminder or vehicle voice prompt, so as to actively remind the outside world that the tailgate is about to open or close or there is a potential danger.

[0083] Step 706: The tailgate avoidance parameters and the object warning parameters are integrated to obtain environment adaptation parameters.

[0084] Specifically, the tailgate avoidance parameters and object warning parameters are fused through a pre-set fusion strategy. For example, the avoidance strategy is applied first in high-risk situations, and the warning response is appropriately enhanced in medium- and low-risk scenarios, while avoiding control logic conflicts and repeated feedback. The time synchronization of the parameters is also processed during the fusion process to ensure that delayed door opening and warning prompts can be triggered in coordination to form a coherent control action. The final output of the environmental adaptation parameters is a set of structured instructions, including dynamically adjusted opening and closing angles, speeds, start timings, and warning methods, which can be directly used to control the tailgate action in real time, so that it can achieve the comprehensive goals of intelligent obstacle avoidance, safety reminders, and smooth opening and closing in a complex and dynamic environment.

[0085] In this embodiment, by comparing and analyzing the tailgate opening and closing control parameters with the motion envelope of the moving object, the envelope overlap information of the possible spatial overlap between the tailgate and the moving object in the future period is calculated, thereby realizing accurate identification of potential collision risks; after identifying the collision risk, the tailgate avoidance parameters and object warning parameters are further dynamically generated based on the overlap information, which are used to adjust the tailgate behavior and actively issue warnings to the outside respectively; finally, the two are integrated to generate environmental adaptation parameters for real-time correction of the tailgate control strategy. It not only improves the active obstacle avoidance and risk response capabilities of the tailgate system in dynamic scenes, but also realizes human-vehicle collaborative warning and intelligent control, effectively prevents collision accidents caused by tailgate movement, and significantly enhances the safety, intelligence and user experience of tailgate operation.

[0086] In an exemplary embodiment, Figure 8 As shown, according to the envelope overlap information, the tailgate avoidance parameter and the object warning parameter are calculated, including steps 802 to 806. Among them: Step 802: Perform interpolation analysis on the opening and closing time of the tailgate of the automobile according to the envelope overlap information to obtain tailgate safety opening and closing information.

[0087] The tailgate safe opening and closing information can be a dynamic data set obtained by combining the tailgate opening and closing control parameters with the moving object motion envelope through time interpolation and spatial overlap analysis, and is used to evaluate whether the tailgate is safe to open and close in various time periods. It describes in detail the time period, degree of intersection, relative movement trend and potential risk level of the tailgate during the entire opening and closing process, when the tailgate may intersect with the surrounding moving objects, so as to help the system accurately identify when it is safe to open and close and when it needs to be avoided or delayed.

[0088] Specifically, the predicted opening and closing trajectory of the tailgate is divided into multiple discrete interpolation periods according to the time dimension based on the envelope overlap information, and in each time slice, the current position of the tailgate is calculated with the predicted moving object motion envelope for spatial intersection, identifying the specific time periods in which the tailgate trajectory is completely separated from the object envelope (indicating absolute safety), the time periods in which there is partial overlap (indicating potential interference risk), and the time periods in which the overlap is high (indicating possible collision). On this basis, the dynamic factors such as the acceleration change of the tailgate action, the object approach speed and direction are integrated to generate a set of tailgate safety opening and closing information including time, safety level and spatial risk distribution.

[0089] Step 804, based on the tailgate safety opening and closing information, emergency adjustment is made to the motor power and battery power of the car tailgate to determine the tailgate avoidance parameters.

[0090] Specifically, the system identifies the time periods with relatively safe spatial windows during the tailgate opening and closing process from the tailgate safety opening and closing information, and then adjusts the motor output power according to these time periods. For example, the power is increased within the safe time window to speed up the tailgate opening and closing action and improve efficiency. In the time period with a higher potential collision risk, active avoidance is performed by reducing the motor power, increasing the opening and closing buffer, slowing down the movement rhythm, etc. At the same time, the instantaneous distribution strategy of the battery power is optimized according to the operation of the motor to avoid energy waste or system instability caused by high load output, and the battery output is dynamically matched with the motor power demand through intelligent distribution, so as to obtain the tailgate avoidance parameters.

[0091] Step 806, calculating the brightness and angle of the warning information for the moving object according to the tailgate safety opening and closing information.

[0092] Specifically, based on the tailgate safe opening and closing information, combined with the spatial relationship between the tailgate and the moving object, the overlapping period and the risk level, the "warning information measurement" and "warning information angle" used to issue an effective warning are calculated. The warning information measurement is a quantitative assessment of the intensity and frequency of the warning, including the volume of the alarm sound, the flashing frequency of the buzzer or light, the number of repetitions of the voice prompt, etc. Its value will be dynamically set according to the type of moving object (such as pedestrians, vehicles), approach speed and degree of danger. For example, a fast approaching vehicle may trigger a higher intensity warning. The warning information angle refers to the direction area of ​​the warning signal. The relative orientation of the moving object is determined by the spatial positioning algorithm, and the optimal angle range that the warning device (such as a directional speaker or a focused light) should face is calculated to achieve a directional and efficient reminder effect.

[0093] Step 808: Generate object warning parameters according to the warning information brightness and the warning information angle.

[0094] Specifically, the brightness and angle of the warning information are optimized based on the type of moving object, the state of motion, the approach direction, and the current opening and closing stage of the tailgate, and the most appropriate warning method and execution strategy are selected as the object warning parameters. For example, when the system identifies a high-risk pedestrian approaching on the left rear side of the tailgate, the object warning parameters may include a high-frequency beep from the left directional speaker, a high-frequency flashing of the left rear light, and a voice broadcast of "The tailgate is open, please pay attention to safety" and other information. These parameters will be output in a structured form to control the corresponding hardware devices (such as sound and light prompt modules, external LEDs, and speakers) to execute accurately, while ensuring that each warning action is highly synchronized in time with the opening and closing behavior of the tailgate, forming a continuous and clear warning feedback mechanism.

[0095] In this embodiment, by performing time interpolation analysis on the coincidence information of the tailgate and the envelope of the moving object, the safe period of the tailgate during the opening and closing process is accurately identified, thereby generating the tailgate safe opening and closing information, and improving the system's time resolution ability for potential risks; on this basis, the tailgate motor and battery power are adjusted dynamically in an emergency, and avoidance parameters are generated to ensure that the tailgate action is smoothly executed within the safety window to avoid misoperation during high-risk periods; at the same time, the brightness and direction of the warning information are calculated based on the safe opening and closing information, so that the warning signal is both visible and directional, and the reminder effect on surrounding moving objects (such as pedestrians and vehicles) is enhanced; the object warning parameters finally generated can realize a directional, high-brightness, high-frequency response active safety prompt mechanism. It can significantly improve the safety control, energy consumption optimization and warning response capabilities of the tailgate system in complex dynamic scenes, ensure that the tailgate operation is accurate, safe and intelligent, and comprehensively improve the active safety performance of the vehicle and user confidence.

[0096] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0097] Based on the same inventive concept, the embodiment of the present application also provides a vehicle tailgate voice control device for implementing the above-mentioned vehicle tailgate voice control method. Fig. 9 As shown, it includes: a voiceprint information recognition module 902, an environmental information acquisition module 904, a control parameter acquisition module 906, an adaptation parameter acquisition module 908 and a control parameter optimization module 910. The implementation solution for solving the problem provided by the device is similar to the implementation solution recorded in the above method. Therefore, the specific limitations in one or more embodiments of a car tailgate voice control device provided below can refer to the above limitations on a car tailgate voice control method, which will not be repeated here.

[0098] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.10 The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface. Those skilled in the art will understand that Fig.10The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0099] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0100] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0101] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0102] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0103] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A car tailgate voice control method, characterized in that: The method comprises: In response to a control instruction of the target object for the tailgate of the car, identifying the command voiceprint information corresponding to the control instruction; When the command voiceprint information passes the voiceprint verification of the automobile tailgate, acquiring the perception environment information corresponding to the automobile tailgate; Inputting the control instruction and the perceived environment information into a tailgate opening and closing control model corresponding to the tailgate of the vehicle to obtain the opening and closing control parameters of the tailgate of the vehicle, including: According to the control instruction and the perceived environment information, the control intention of the target object for the automobile tailgate is identified to obtain tailgate control intention identification data; According to the tailgate control intention recognition data and the perceived environment information, the motion parameters of the automobile tailgate are calculated to obtain initial control parameters, including: According to the tailgate control intention recognition data and the perceived environment information, a tailgate scene coupling model of the automobile tailgate is set; Inputting the tailgate control intention recognition data and the perceived environment information into the tailgate scene coupling model to obtain model calculation control parameters; Performing hyperstatic optimization on the model calculation control parameters to obtain the initial control parameters; According to the mechanical control error of the automobile tailgate, the initial control parameter is adjusted to obtain the opening and closing control parameter; Determining environmental adaptation parameters of the automobile tailgate according to the perceived environmental information and the opening and closing control parameters; The opening and closing control parameters and the environmental adaptation parameters are jointly optimized to obtain target tailgate control parameters corresponding to the control instructions.

2. The method according to claim 1, characterized in that The step of performing hyperstatic optimization on the model calculation control parameters to obtain the initial control parameters includes: Predicting environmental changes of the tailgate of the vehicle based on the perceived environmental information to obtain environmental mutation prediction data; Redundancy identification is performed on the model calculation control parameters and the environmental mutation prediction data to obtain hyperstatic adjustment data of the automobile tailgate; The model calculation control parameters are adjusted according to the hyperstatic adjustment data to obtain the initial control parameters.

3. The method according to claim 2, characterized in that The expression of the tailgate scene coupling model is: in, is the angular acceleration The changing moment of inertia of the tailgate around the axis of rotation, is the angular acceleration varying with time t, is the angular velocity varying with time t, is the angle of the tailgate relative to the initial closed position, is the motor torque constant, is the motor current varying with time t, With angular velocity and the damping torque varying with time t, is the gravitational torque, is the external environmental torque that changes with time t, is the temperature of the motor and its transmission components at time t, is the equivalent thermal capacity of the motor and related mechanical components, is the temperature-dependent resistance of the motor winding, is the heat power generated by mechanical energy loss, is the heat dissipation power, is the convective heat transfer coefficient, is the ambient temperature, is the motor output torque that changes with time t, is the motor input voltage that changes with time t, k is the motor back electromotive force constant, and L is the motor armature inductance.

4. The method according to claim 1, characterized in that: The step of determining the environmental adaptation parameters of the automobile tailgate according to the perceived environmental information and the opening and closing control parameters includes: Identifying pedestrian movement information and other vehicle movement information from the perceived environment information; Predicting an object motion envelope corresponding to the moving object according to the human motion information and the other vehicle motion information; The environmental adaptation parameters of the automobile tailgate are generated according to the opening and closing control parameters and the object motion envelope.

5. The method according to claim 4, characterized in that The step of generating the environmental adaptation parameters of the automobile tailgate according to the opening and closing control parameters and the object motion envelope includes: Calculating the overlap information of the envelope of the automobile tailgate and the moving object according to the opening and closing control parameters and the motion envelope of the object; When the envelope overlap information indicates that the tailgate of the vehicle is about to collide with the moving object, calculating a tailgate avoidance parameter and an object warning parameter according to the envelope overlap information; The tailgate avoidance parameter and the object warning parameter are integrated to obtain the environment adaptation parameter.

6. The method according to claim 5, characterized in that The step of calculating the tailgate avoidance parameter and the object warning parameter according to the envelope overlap information includes: According to the envelope overlap information, an interpolation analysis is performed on the opening and closing time of the tailgate of the vehicle to obtain the tailgate safety opening and closing information; According to the tailgate safety opening and closing information, emergency adjustment is performed on the motor power and battery power of the automobile tailgate to determine the tailgate avoidance parameter; Calculating the brightness and angle of the warning information for the moving object according to the tailgate safety opening and closing information; The object warning parameter is generated according to the warning information brightness and the warning information angle.

7. A car tailgate voice control device, characterized in that: The device comprises: A voiceprint information recognition module, used to respond to a control instruction of a target object for a car tailgate and recognize the command voiceprint information corresponding to the control instruction; An environment information acquisition module, used for acquiring the perception environment information corresponding to the tailgate of the vehicle when the command voiceprint information passes the voiceprint verification of the tailgate of the vehicle; A control parameter obtaining module, used to input the control instruction and the perceived environment information into a tailgate opening and closing control model corresponding to the tailgate of the vehicle to obtain the opening and closing control parameters of the tailgate of the vehicle, including: According to the control instruction and the perceived environment information, the control intention of the target object for the automobile tailgate is identified to obtain tailgate control intention identification data; According to the tailgate control intention recognition data and the perceived environment information, the motion parameters of the automobile tailgate are calculated to obtain initial control parameters, including: According to the tailgate control intention recognition data and the perceived environment information, a tailgate scene coupling model of the automobile tailgate is set; Inputting the tailgate control intention recognition data and the perceived environment information into the tailgate scene coupling model to obtain model calculation control parameters; Performing hyperstatic optimization on the model calculation control parameters to obtain the initial control parameters; According to the mechanical control error of the automobile tailgate, the initial control parameter is adjusted to obtain the opening and closing control parameter; An adaptation parameter obtaining module, used to determine the environmental adaptation parameters of the automobile tailgate according to the perceived environmental information and the opening and closing control parameters; The control parameter optimization module is also used to jointly optimize the opening and closing control parameters and the environmental adaptation parameters to obtain the target tailgate control parameters corresponding to the control instructions.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

Images