Calibration method, system and equipment for vehicle door opening and closing control parameters and medium

Abstract

The invention discloses a car door opening and closing control parameter calibration method, system and device and a medium, and the method comprises the steps: training a pre-constructed artificial intelligence model according to a plurality of preset sample working condition data, acquiring corresponding sample vehicle door opening and closing control parameters when the artificial intelligence model simulates vehicle door opening and closing operation under each kind of sample working condition data; according to the sample vehicle door opening and closing control parameters, calibration of the vehicle door opening and closing control parameters of the to-be-calibrated vehicle is completed, so that the calibration efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of automobiles, and in particular to a method, system, equipment and medium for calibrating door switch control parameters. Background Art

[0002] With the rapid development of vehicle automation technology, electric doors are becoming more and more widely used. Therefore, achieving precise and stable control of the opening and closing of doors and providing users with a more comfortable experience have become important research areas in the development of vehicle automation.

[0003] In the prior art, in order to realize the control of the vehicle doors by the vehicle control system, engineers need to calibrate the actual vehicle before the vehicle is put into use, and after adapting the various parameters of the door drive module and the mechanical structure within the system, adapt to different vehicles through a self-learning algorithm to realize automatic control of the doors.

[0004] However, in the existing calibration process of intelligent door control, engineers need to solidify the calibration parameters, and solidifying the calibration parameters takes one month to debug and calibrate the actual vehicle, which not only leads to low calibration efficiency, but also is often affected by factors such as low current sampling accuracy of the circuit, low ripple counting accuracy, complex software algorithms, environmental changes, aging of mechanical components, hinge assembly errors, limit angle deviations, and internal structural size errors of the motor assembly. The influence of the above factors not only makes the door calibration work extremely difficult but also easily causes huge errors. Summary of the invention

[0005] In order to solve the above technical problems, the present invention discloses a method, system, device and medium for calibrating door switch control parameters, so as to improve the efficiency of door switch control parameter calibration.

[0006] In order to achieve the above-mentioned object, in a first aspect, the present invention discloses a method for calibrating door switch control parameters, comprising:

[0007] The pre-built artificial intelligence model is trained according to a plurality of preset sample operating condition data, and the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample operating condition data are obtained;

[0008] The door switch control parameters of the vehicle to be calibrated are calibrated according to the sample door switch control parameters.

[0009] The present invention discloses a method for calibrating door switch control parameters. A pre-constructed artificial intelligence model is trained by using several pre-constructed sample operating condition data to obtain sample door switch control parameters corresponding to the artificial intelligence model when simulating door switch operation under each of the sample operating condition data. The sample door switch control parameters are then used to complete the calibration of the door switch control parameters of the vehicle, thereby realizing automatic collection of sample door switch control parameters, avoiding manual calibration, and improving calibration efficiency.

[0010] As a preferred example, the preset several kinds of sample operating condition data are formed according to preset power supply voltage intervals, ambient temperature intervals and vehicle body posture intervals, and are composed of several kinds of sample operating condition data consisting of different power supply voltages, different ambient temperatures and different vehicle body postures.

[0011] The present invention simulates the environmental conditions corresponding to the vehicle when the door switch operation is actually performed according to different power supply voltages, different ambient temperatures and different vehicle body postures, ensuring that the final calibrated control parameters can adapt to different environments and improving the environmental adaptability and accuracy of the parameters.

[0012] As a preferred example, the pre-built artificial intelligence model is trained according to several preset sample working condition data to obtain the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample working condition data, including:

[0013] Under each of the sample operating condition data, simulating the door switch operation according to a preset number of training times;

[0014] In each simulation of the door switch operation, the door operation state value corresponding to the door switch operation is obtained by adjusting the preset side door motor steering value and the preset PWM duty cycle; wherein the door operation state value includes the Hall pulse count value, the Hall pulse speed value, the side door drive motor speed value, the side door drive motor current pulse count value and the side door drive motor rotation circle count value;

[0015] When it is determined that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation;

[0016] When it is determined that the switch time meets the preset time condition, the adjusted side door motor steering value and the adjusted PWM duty cycle conversion rate are obtained as the sample door switch control parameters.

[0017] The present invention screens the steering value and PWM duty cycle of the side door motor adjusted in each door switch operation by setting the door operation conditions to obtain a sample data set for calibration to perform actual calibration of the vehicle, thereby avoiding manual adjustment and improving marking efficiency.

[0018] As a preferred example, the step of adjusting a preset side door motor steering value and a preset PWM duty cycle to obtain a door operation state value corresponding to the door switch operation includes:

[0019] During each simulated door opening operation, the preset side door motor steering value is adjusted to the first side door motor steering value, and the preset PWM duty cycle is adjusted to the first PWM duty cycle initial value 1;

[0020] Acquire a first PWM duty cycle change rate according to the first PWM duty cycle initial value 1 and the preset PWM duty cycle;

[0021] Acquiring a first door operation state value based on the first side door motor steering value and the first PWM duty cycle change rate;

[0022] After determining that the first door operation state value reaches the preset first stage threshold, entering the door opening action acceleration or deceleration stage, adjusting the preset side door motor steering value to the second side door motor steering value, and adjusting the preset PWM duty cycle to the second PWM duty cycle initial value 2;

[0023] According to the second PWM duty cycle initial value 2 and the preset PWM duty cycle, obtaining a second PWM duty cycle change rate;

[0024] A second door operation state value is acquired based on the second side door motor steering value and the second PWM duty cycle change rate.

[0025] The present invention sets different stage thresholds to simulate the different stages corresponding to the door opening, and then completes the different control parameters corresponding to the door during the entire opening operation period through the set stage thresholds, thereby ensuring the smooth opening operation.

[0026] As a preferred example, the step of adjusting a preset side door motor steering value and a preset PWM duty cycle to obtain a door operation state value corresponding to the door switch operation includes:

[0027] During each simulated door closing operation, the preset side door motor steering value is adjusted to the third side door motor steering value, and the preset PWM duty cycle is adjusted to the third PWM duty cycle initial value 3;

[0028] According to the third PWM duty cycle initial value 3 and the preset PWM duty cycle, obtaining a third PWM duty cycle change rate;

[0029] Based on the third side door motor steering value and the third PWM duty cycle change rate, obtaining a third door operation state value;

[0030] After determining that the third door operation state value reaches the preset third stage threshold, entering the door closing action acceleration or deceleration stage, adjusting the preset side door motor steering value to the fourth side door motor steering value, and adjusting the preset PWM duty cycle to the fourth PWM duty cycle initial value 4;

[0031] According to the fourth PWM duty cycle initial value 4 and the preset PWM duty cycle, obtaining a fourth PWM duty cycle change rate;

[0032] A fourth door operation state value is acquired based on the fourth side door motor steering value and the fourth PWM duty cycle change rate.

[0033] The present invention sets different stage thresholds to simulate different stages corresponding to the door when closing the door, and then completes different control parameters corresponding to the door during the entire closing operation period through the set stage thresholds, thereby ensuring smooth closing operation.

[0034] As a preferred example, when determining that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation includes:

[0035] Determining that the second door operation state value reaches a preset second stage threshold;

[0036] Get the door opening time corresponding to the opening operation.

[0037] The present invention determines that the opening operation has been completed according to the second stage threshold, and then obtains the door opening time, which is used to subsequently judge whether the corresponding control parameters in the current opening operation meet the requirements according to the door opening time, thereby completing the parameter screening.

[0038] As a preferred example, when determining that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation includes:

[0039] Determining that the fourth door operation state value reaches a preset fourth stage threshold;

[0040] Get the door closing duration corresponding to the closing operation.

[0041] The present invention determines that the closing operation has been completed according to the fourth stage threshold, and then obtains the closing time, which is used to subsequently judge whether the corresponding control parameters in the current closing operation meet the requirements according to the closing time, thereby completing the parameter screening.

[0042] As a preferred example, the obtaining of sample door switch control parameters corresponding to the door switch operation simulated by the artificial intelligence model under each of the sample operating condition data includes:

[0043] Obtaining the door opening duration corresponding to the opening operation, and when determining that the door opening duration is within a preset duration threshold range, outputting the first side door motor steering value, the first PWM duty cycle change rate, the second side door motor steering value and the second PWM duty cycle change rate as the door opening control parameters corresponding to the opening operation;

[0044] Obtain the door closing duration corresponding to the closing operation, and when it is determined that the door closing duration is within the duration threshold range, output the third side door motor steering value, the third PWM duty cycle change rate, the fourth side door motor steering value and the fourth PWM duty cycle change rate as the door closing control parameters corresponding to the closing operation.

[0045] The present invention utilizes the duration threshold range to determine that when a vehicle performs a door opening or closing operation according to the collected control parameters, the corresponding door opening and closing operation can be completed within a duration range that meets user requirements, thereby improving the user experience.

[0046] As a preferred example, the step of completing the calibration of the door switch control parameters of the vehicle to be calibrated according to the sample door switch control parameters includes:

[0047] According to the sample door switch control parameters corresponding to each door switch operation, fitting the structural component coefficient change curve of the vehicle to be calibrated to obtain a first fitting door switch control parameter;

[0048] According to the first fitting door switch control parameter, fitting the motor assembly change curve of the vehicle to be calibrated to obtain the second fitting door switch control parameter;

[0049] When the difference between the second fitting door switch control parameter and the sample door switch control parameter is within a preset tolerance threshold range, the sample door switch control parameter is written into the vehicle to be calibrated to complete the calibration of the door control parameters of the vehicle to be calibrated.

[0050] The present invention fits the sample door switch control parameters with the structural component coefficient change curve and the motor assembly change curve corresponding to the vehicle to obtain the difference between the parameters before and after fitting, and compares the difference with a preset tolerance threshold range to provide feedback on the compatibility of the vehicle with the parameters and ensure the calibration effect.

[0051] In a second aspect, the present invention discloses a calibration system for door switch control parameters, the system comprising a training module and a calibration module;

[0052] The training module is used to train the pre-built artificial intelligence model according to a plurality of preset sample operating condition data, and obtain the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample operating condition data;

[0053] The calibration module is used to complete the calibration of the door switch control parameters of the vehicle to be calibrated according to the sample door switch control parameters.

[0054] The present invention discloses a door switch control parameter calibration system, which trains a pre-constructed artificial intelligence model through several pre-constructed sample operating condition data to obtain sample door switch control parameters corresponding to the artificial intelligence model when simulating door switch operation under each of the sample operating condition data, and then uses the sample door switch control parameters to complete the calibration of the vehicle's door switch control parameters, thereby realizing automatic collection of sample door switch control parameters, avoiding manual calibration, and improving calibration efficiency.

[0055] In a third aspect, the present invention discloses an electronic device, comprising: at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute a method for calibrating door switch control parameters as described in the first aspect.

[0056] In a fourth aspect, the present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute a method for calibrating a door switch control parameter as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 : A schematic flow chart of a method for calibrating door switch control parameters disclosed in an embodiment of the present invention;

[0058] Figure 2 : is a structural schematic diagram of a door switch control parameter calibration system disclosed in an embodiment of the present invention;

[0059] Figure 3 : A schematic flow chart of a method for calibrating door switch control parameters disclosed in another embodiment of the present invention;

[0060] Figure 4 : A schematic diagram of the opening / closing action flow of a vehicle side door disclosed in another embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Embodiment 1

[0063] This embodiment discloses a method for calibrating a door switch control parameter. For details, please refer to Figure 1 , mainly including step 101 to step 102, the steps are mainly:

[0064] Step 101: Train a pre-built artificial intelligence model according to several preset sample operating condition data, and obtain sample door switch control parameters corresponding to the artificial intelligence model when simulating door switch operation under each of the sample operating condition data.

[0065] Step 102: completing the calibration of the door switch control parameters of the vehicle to be calibrated according to the sample door switch control parameters.

[0066] In certain implementations of the present embodiment, before training the artificial intelligence model in step 101 to obtain the sample door switch control parameters, the artificial intelligence model for simulating the door switch can be constructed by dynamics software such as ADAMS software, MapleSim software, etc., or by using a deep learning model, etc., and then the actual door opening and closing time, the actual side door motor steering value, and the actual PWM duty cycle are used as input values ​​of the model, and the door operation status value is used as the output value to train the constructed artificial intelligence model, so that when the difference between the door operation status value output by the artificial intelligence model and the actual side door motor steering value and the actual PWM duty cycle is small enough, the model construction of the artificial intelligence model is completed.

[0067] In certain implementations of this embodiment, the preset several types of sample operating condition data in step 101 are formed based on preset power supply voltage intervals, ambient temperature intervals and vehicle body posture intervals, and are composed of several types of sample operating condition data consisting of different power supply voltages, different ambient temperatures and different vehicle body postures.

[0068] Based on the several kinds of sample operating condition data constructed above, in some implementations of this embodiment, step 101 includes the following steps:

[0069] S1011: Under each of the sample operating condition data, simulating the door switch operation according to a preset number of training times;

[0070] S1012: In each simulation of the door switch operation, the door operation status value corresponding to the door switch operation is obtained by adjusting the preset side door motor steering value and the preset PWM duty cycle; wherein the door operation status value includes the Hall pulse count value, the Hall pulse speed value, the side door drive motor speed value, the side door drive motor current pulse count value and the side door drive motor rotation circle count value.

[0071] S1013: When it is determined that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation;

[0072] S1014: When it is determined that the switch time meets a preset time condition, the adjusted side door motor steering value and the adjusted PWM duty cycle change rate are obtained as the sample door switch control parameters.

[0073] Specifically, the door switch operation in step S1012 is divided into an opening operation and a closing operation. The parameter acquisition process corresponding to the opening operation is as follows: each time the door opening operation is simulated, the preset side door motor steering value is adjusted to the first side door motor steering value, and the preset PWM duty cycle is adjusted to the first PWM duty cycle initial value 1; according to the first PWM duty cycle initial value 1 and the preset PWM duty cycle, the first PWM duty cycle change rate is obtained; based on the first side door motor steering value and the first PWM duty cycle change rate, the first door operation state value is obtained; after determining that the first door operation state value reaches the preset first stage threshold, the door opening action acceleration or deceleration stage is entered, the preset side door motor steering value is adjusted to the second side door motor steering value, and the preset PWM duty cycle is adjusted to the second PWM duty cycle initial value 2;

[0074] According to the second PWM duty cycle initial value 2 and the preset PWM duty cycle, a second PWM duty cycle change rate is obtained; based on the second side door motor steering value and the second PWM duty cycle change rate, a second door operation state value is obtained.

[0075] The parameter acquisition process corresponding to the closing operation is as follows: each time the closing operation of the simulated vehicle door is performed, the preset side door motor steering value is adjusted to the third side door motor steering value, and the preset PWM duty cycle is adjusted to the third PWM duty cycle initial value 3; according to the third PWM duty cycle initial value 3 and the preset PWM duty cycle, the third PWM duty cycle change rate is obtained; based on the third side door motor steering value and the third PWM duty cycle change rate, the third vehicle door operation state value is obtained; after determining that the third vehicle door operation state value reaches the preset third stage threshold, entering the door closing action acceleration or deceleration stage, adjusting the preset side door motor steering value to the fourth side door motor steering value, and adjusting the preset PWM duty cycle to the fourth PWM duty cycle initial value 4; according to the fourth PWM duty cycle initial value 4 and the preset PWM duty cycle, the fourth PWM duty cycle change rate is obtained; based on the fourth side door motor steering value and the fourth PWM duty cycle change rate, the fourth vehicle door operation state value is obtained.

[0076] In certain implementations of this embodiment, after simulating the closing operation and the opening operation respectively, the S1013 specifically includes: determining that the second door operating status value reaches a preset second stage threshold; obtaining the door opening time corresponding to the opening operation; determining that the fourth door operating status value reaches a preset fourth stage threshold; obtaining the door closing time corresponding to the closing operation.

[0077] Further, after obtaining the door opening duration and the door closing duration, the step S1014 includes: obtaining the door opening duration corresponding to the opening operation, and when it is determined that the door opening duration is within a preset duration threshold range, outputting the first side door motor steering value, the first PWM duty cycle change rate, the second side door motor steering value and the second PWM duty cycle change rate as the door opening control parameters corresponding to the opening operation; obtaining the door opening duration corresponding to the closing operation, and when it is determined that the door closing duration is within the duration threshold range, outputting the third side door motor steering value, the third PWM duty cycle change rate, the fourth side door motor steering value and the fourth PWM duty cycle change rate as the door closing control parameters corresponding to the closing operation.

[0078] In some implementations of this embodiment, when calibrating the door control parameters of the vehicle based on the sample door switch control parameters obtained in step 101, step 102 includes the following steps:

[0079] S1021: fitting a structural component coefficient variation curve of the vehicle to be calibrated according to the sample door switch control parameters corresponding to each door switch operation to obtain a first fitting door switch control parameter;

[0080] S1022: fitting a motor assembly variation curve of the vehicle to be calibrated according to the first fitting door switch control parameter to obtain a second fitting door switch control parameter;

[0081] S1023: When the difference between the second fitting door switch control parameter and the sample door switch control parameter is within a preset tolerance threshold range, the sample door switch control parameter is written into the vehicle to be calibrated, thereby completing the calibration of the door control parameters of the vehicle to be calibrated.

[0082] Furthermore, during the calibration of the door control parameters in step S1023, the real-time operating status value of the door of the vehicle to be calibrated when performing an opening operation or a closing operation according to the sample door switch control parameters can also be obtained; the difference between the real-time operating status value and the door operating status value corresponding to the door switch operation simulated according to the sample door switch control parameters is obtained; and then, based on the difference, a preset algorithm strategy such as a PID control algorithm (a control algorithm that combines proportional, integral and differential links into one) is called to optimize the sample door switch control parameters.

[0083] On the other hand, this embodiment also discloses a door switch control parameter calibration system. The specific structure of the calibration system is shown in FIG. Figure 2 The system includes a training module 201 and a calibration module 202.

[0084] The training module 201 is used to train the pre-built artificial intelligence model according to several preset sample operating condition data, and obtain the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample operating condition data.

[0085] The calibration module 202 is used to complete the calibration of the door switch control parameters of the vehicle to be calibrated according to the sample door switch control parameters.

[0086] In addition to the above-mentioned method and system, this embodiment also discloses an electronic device and a computer-readable storage medium; wherein, the electronic device includes: at least one processor; and a memory communicatively connected to at least one of the processors; wherein, the memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute a method for calibrating a door switch control parameter described in this embodiment; and the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute a method for calibrating a door switch control parameter described in this embodiment.

[0087] The present embodiment discloses a door switch control parameter calibration method, system, device and medium, which trains a pre-constructed artificial intelligence model through several pre-constructed sample operating condition data to obtain sample door switch control parameters corresponding to the artificial intelligence model when simulating door switch operation under each of the sample operating condition data, and then uses the sample door switch control parameters to complete the calibration of the door switch control parameters of the vehicle to be calibrated, thereby realizing the automatic collection of sample door switch control parameters, avoiding manual calibration, and improving calibration efficiency.

[0088] Embodiment 2

[0089] Based on the calibration method provided in the first embodiment, in the possible implementation example, the vehicle battery voltage range, the operating temperature range, and the vehicle body posture, i.e., the slope of the vehicle position, set in the stored sample operating condition data are used as input variables of the pre-built AI model, so as to obtain the corresponding door control parameters each time the vehicle door switch operation is performed by training the AI ​​model. Specifically, in this embodiment, the AI ​​model can be automatically trained through the training platform built by the Linux system, and then the control parameter set of the door is generated by the AI ​​model according to different sample operating condition data and the preset number of training times under each sample operating condition data, and then the door control of the vehicle is calibrated according to the control parameter set. Among them, the AI ​​model is a mathematical model that uses machine learning, data mining and other methods to analyze, process, predict and optimize data with certain regularity and predictability. In this embodiment, the AI ​​model can use any one of the models such as Claude, Gemini, etc. to calibrate the door switch control parameters.

[0090] In the application example, the specific implementation of calibrating the vehicle door control parameters based on the above-set input variables and training closed loop can be found in Figure 3 , mainly including steps 301 to 303, wherein the steps are mainly:

[0091] Step 301: Construct a plurality of calibration conditions according to different power supply voltages, calibration temperatures and calibration slopes of the vehicle.

[0092] Specifically, in the preset operating condition sample, the battery voltage V range is set to 9 to 16V, the temperature T range is -30℃ to +85℃, and the body posture P range is a slope of ±20%. Then, any voltage value is selected from the battery voltage V range, any temperature value is selected from the temperature T range, and any slope value is selected from the body posture P range. Then, the calibration operating condition is composed according to the voltage value, the temperature value and the slope value, that is, the calibration operating condition is a numerical combination of P / T / V.

[0093] In one embodiment, if the calibration temperature is set to only -30°C, 25°C, and 85°C, it can be seen that there are 3 types of temperature values. If the calibration voltage is set to only 9V, 13.5V, and 16V, it can be seen that there are 3 types of voltage values. The slope requirement of the vehicle body posture is generally 20%. If only flat ground, 20% upward inclination, 20% downward inclination, 20% left inclination, and 20% right inclination are used, then according to the calibration temperature, the calibration voltage and the slope requirement, it can be seen that a total of 3*3*5 calibration conditions can be formed, such as the first calibration condition corresponding to the temperature of -30°C, the calibration voltage of 9V, and the vehicle body posture of 20% upward inclination, or the second calibration condition corresponding to the temperature of -30°C, the calibration voltage of 9V, and the vehicle body posture of 20% downward inclination, and so on.

[0094] Step 302: Under each of the calibration conditions, control the training platform to perform a number of door opening / closing actions and store control parameter data corresponding to each door opening / closing action.

[0095] Specifically, the battery voltage is set to 13.5V, the temperature is set to 25°C, and the flat ground (slope is 0%) is used as the initial training condition, which is defined as V0TOP0 (wherein the V0 represents the battery voltage of 13.5V, the T0 represents the temperature of 25°C, and the P0 represents the slope of 0%), and the count Counter=0 is set to count the number of times the door opening / closing action corresponds to.

[0096] Under the V0TOP0 working condition, each time a round of door opening / closing work is completed and the car door returns to the zero state, the Counter is accumulated by 1, and the control parameter data corresponding to this round of door opening / closing work is stored until the number of door opening / closing works reaches a preset number threshold and then the working condition combination is updated. For example, when the temperature working condition in the V0TOP0 working condition combination is changed, that is, the temperature is modified to 30°C, and the battery voltage and the slope are not changed, the V0TOP0 working condition combination is updated to V0T1 P0, and the door opening / closing work continues to be performed under the V0T1 P0 working condition and the control parameter data corresponding to each round of door opening / closing work is stored.

[0097] Further, please refer to the corresponding process when the training platform performs each door opening / closing operation. Figure 4 ,like Figure 4 As shown, first, the training platform enters the door opening mode and reads the currently set working condition combination, that is, reads the currently input vehicle power V, vehicle body posture P, and vehicle internal and external temperature T.

[0098] Next, the training platform is provided with an obstacle detection mode, that is, the training platform imitates the surrounding environment information of the vehicle door to detect whether there is an obstacle according to the surrounding environment information, and determines whether the vehicle enters the vehicle door obstacle avoidance mode according to the obstacle detection result.

[0099] When it is determined that there is no obstacle, the training platform executes the door opening action. It should be noted that at this time, in order to realize the automation of the door switch, the training platform sets the door to be in a fully locked state, that is, the door lock is in a completely locked state, and starts to enter the door opening acceleration or deceleration stage by adjusting the side door motor steering value and the PWM (pulse width modulation) duty cycle, and obtains the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value corresponding to the current side door motor steering value and the current PWM duty cycle. Whether they reach the preset first stage threshold, when the first stage threshold is not reached, the side door motor steering value and the PWM duty cycle are repeatedly adjusted until the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value reach the first stage threshold.

[0100] After reaching the first stage threshold, the second stage of door opening deceleration or acceleration is entered according to the preset side door motor steering value and PWM duty cycle, and the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value corresponding to the current side door motor steering value and PWM duty cycle are obtained to see whether they reach the preset second stage threshold. When the second stage threshold is not reached, the side door motor steering value and the PWM duty cycle are repeatedly adjusted until the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value reach the second stage threshold. Among them, since the gravity analysis of the car door at different slopes is divided into two stages, that is, when the car door is opened, the tension changes to pressure or from pressure to tension, the corresponding motor driving capacity requirement is from large to small and then large, or from small to large and then small, that is, in the same door opening process, the motor will appear to be always forward, or from forward to reverse. In addition, the door opening action also needs to include the side door motor to open the door lock in the semi-locked state. Therefore, two stage thresholds, namely the first stage threshold and the second stage threshold, need to be set in the door opening process, so as to adjust the smoothness of the door opening action in combination with the force condition of the support rod. Furthermore, the first stage threshold and the second stage threshold are theoretical initial values ​​calculated by analyzing the force change under ideal conditions. Specifically, in a specific application scenario, assuming that under normal temperature, 20% downhill slope, and battery voltage of 13.5V, the opening angle of the left front door of a certain vehicle model is 60°, and the door opening action is required to be completed in 3 seconds, then the corresponding second-stage threshold is that the motor needs to rotate 1800 times, the number of Hall pulses is 43200, and the number of motor current pulses is 18000. The average motor speed is 6000rpm, and the average Hall pulse speed is 14400 / s. Among them, when the left front door is opened to 18°, the support rod changes from being under pressure to being under tension. At this time, the motor changes from forward to reverse, and the driving current drops from the large current of 2.95A corresponding to the initial moment to the minimum of 1.27A (corresponding to the duty cycle required by the A-size motor at this moment is 12%). When the corresponding stroke is 1 / 3, the corresponding first-stage threshold is 14400 Hall pulses, 6000 motor current pulses, and the average speed of Hall pulses is 14400 / s. The door opening action is continued according to this ideal data, and the door opening action is finally completed in 3s±0.1s. If the door opening action cannot be completed within 3s±0.1s, the above threshold is adjusted according to the PID algorithm until it is met.

[0101] After reaching the threshold of the second stage, the time corresponding to the door opening action performed by the training platform is counted by a timer preset in the training platform, and the time is compared with the preset door opening time range, and the side door motor steering value and PWM duty cycle, Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value corresponding to each stage of this door opening action are stored in the corresponding case. Among them, when the time is within the door opening time range, this case is classified as a door opening OK case. When the time is outside the door opening time range, this case is classified as a door opening NG case.

[0102] After the door opening action is completed, the training platform enters the door closing mode, and detects whether there is an obstacle according to the set obstacle detection mode, and determines whether the vehicle enters the door anti-pinch mode according to the obstacle detection result.

[0103] When it is determined that there is no obstacle, the training platform executes the door closing action. At this time, the training platform enters the door closing deceleration / acceleration stage by adjusting the side door motor steering value and the PWM (pulse width modulation) duty cycle, and obtains the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value corresponding to the current side door motor steering value and the current PWM duty cycle to determine whether they reach the preset third stage threshold. When the third stage threshold is not reached, the side door motor steering value and the PWM duty cycle are repeatedly adjusted until the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value reach the third stage threshold.

[0104] After reaching the third stage threshold, the fourth stage of door closing acceleration / deceleration phase is entered according to the preset side door motor steering value and PWM duty cycle, and the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value corresponding to the current side door motor steering value and PWM duty cycle are obtained to determine whether they have reached the preset fourth stage threshold. When the fourth stage threshold is not reached, the side door motor steering value and the PWM duty cycle are repeatedly adjusted until the Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value reach the fourth stage threshold. Among them, since the gravity analysis of the car door at different slopes is divided into two stages, that is, when the car door is closed, the tension changes to pressure or from pressure to tension, the corresponding motor driving capacity requirement is from large to small and then large, or from small to large and then small, that is, in the same door closing process, the motor will appear to be reversed all the time, or from forward to reverse. In addition, the door closing action also needs to include the side door motor accelerating into the door lock so that the door lock can be attracted. Therefore, two stage thresholds, namely the third stage threshold and the fourth stage threshold, need to be set in the door closing process, so as to adjust the smoothness of the door closing action in combination with the force of the support rod. Furthermore, the third stage threshold and the fourth stage threshold are theoretical initial values ​​calculated by analyzing the force changes under ideal conditions.

[0105] After reaching the fourth stage threshold, the time corresponding to the door closing action performed by the training platform is counted by a timer preset in the training platform, and the time is compared with the preset door closing time range, and the side door motor steering value and PWM duty cycle, Hall pulse count value, Hall pulse speed value, side door drive motor speed value, side door drive motor current pulse count value and side door drive motor circle count value corresponding to each stage of this door closing action are stored in the corresponding case. Among them, when the time is within the door closing time range, this case is classified as a door closing OK case. When the time is outside the door closing time range, this case is classified as a door closing NG case.

[0106] Specifically, the self-learning calibration data collection corresponding to the training platform is formed according to the combination counts of the temperature, voltage and slope, the number of door opening / closing actions performed under each combination, and the door opening OK case or door opening NG case, door closing OK case or door closing NG case stored corresponding to each door opening / closing action, that is, the training platform outputs the corresponding control parameter set.

[0107] Step 303: Fit the control parameter data to the structural component coefficient variation curve and the motor assembly variation curve, so as to calibrate the vehicle according to the fitting result and the preset algorithm strategy.

[0108] Specifically, after the training platform completes the self-learning calibration data collection, the control parameter data formed during the training process is fit to the mechanical structure coefficient change curve of the vehicle to be calibrated to obtain the first fitting parameter data, and the first fitting parameter data is fit to the motor assembly change curve of the vehicle to be calibrated to generate the final fitting control parameter data.

[0109] In order to ensure the compatibility of the control parameters with the vehicle's door control system, the difference between each fitted control parameter data and the control parameter data formed during the training process is obtained. When the difference is within a preset tolerance range, it indicates that the parameter is compatible with the vehicle's door control system. The tolerance range can be manually set according to the type of vehicle to be calibrated, and is not specifically limited.

[0110] Furthermore, after determining that the door control system of the target vehicle is compatible with the control parameter data, in the process of calibrating the door control system of the target vehicle according to the control parameter data, the difference between the value of the real-time operating status and the database output by the training platform is compared to identify the state change of the door switching process, and call a preset algorithm strategy, such as a PID control algorithm (a control algorithm that combines proportional, integral and differential links into one) to optimize the control parameter data so that the door control smoothness reaches and maintains the best state, thereby completing the calibration of the target vehicle.

[0111] A door switch control parameter calibration method disclosed in this embodiment can build a training platform through a Linux system, and then perform automated training through the training platform to obtain an AI database of the door with different motors (torque coefficients) and different structures (structure screening is classified as structural variables), thereby avoiding the technical problem of low calibration efficiency caused by manual testing and improving the efficiency of calibration.

[0112] Furthermore, when calibrating the vehicle according to the data generated by the training platform, the door operation status data collected in real time can be compared with the operation status values ​​in the data set generated by the training platform to identify the state changes of the door opening and closing process, and then the preset algorithm strategy can be called to optimize the control parameters used during calibration, so that the door control smoothness reaches and maintains the best state, thereby improving the accuracy of the calibration.

[0113] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for calibrating door switch control parameters, characterized in that: include: The pre-built artificial intelligence model is trained according to a plurality of preset sample operating condition data, and the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample operating condition data are obtained; The door switch control parameters of the vehicle to be calibrated are calibrated according to the sample door switch control parameters.

2. A method for calibrating door switch control parameters according to claim 1, characterized in that: The preset several kinds of sample operating condition data are formed according to preset power supply voltage interval, ambient temperature interval and vehicle body posture interval, and are composed of several kinds of sample operating condition data consisting of different power supply voltages, different ambient temperatures and different vehicle body postures.

3. A method for calibrating door switch control parameters according to claim 2, characterized in that: The pre-built artificial intelligence model is trained according to the preset several kinds of sample working condition data to obtain the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample working condition data, including: Under each of the sample operating condition data, simulating the door switch operation according to a preset number of training times; In each simulation of the door switch operation, the door operation state value corresponding to the door switch operation is obtained by adjusting the preset side door motor steering value and the preset PWM duty cycle; wherein the door operation state value includes the Hall pulse count value, the Hall pulse speed value, the side door drive motor speed value, the side door drive motor current pulse count value and the side door drive motor rotation circle count value; When it is determined that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation; When it is determined that the switch time meets the preset time condition, the adjusted side door motor steering value and the adjusted PWM duty cycle conversion rate are obtained as the sample door switch control parameters.

4. A method for calibrating door switch control parameters according to claim 3, characterized in that: The step of adjusting a preset side door motor steering value and a preset PWM duty cycle to obtain a door operation state value corresponding to the door switch operation includes: During each simulated door opening operation, the preset side door motor steering value is adjusted to the first side door motor steering value, and the preset PWM duty cycle is adjusted to the first PWM duty cycle initial value 1; Acquire a first PWM duty cycle change rate according to the first PWM duty cycle initial value 1 and the preset PWM duty cycle; Acquiring a first door operation state value based on the first side door motor steering value and the first PWM duty cycle change rate; After determining that the first door operation state value reaches the preset first stage threshold, entering the door opening action acceleration or deceleration stage, adjusting the preset side door motor steering value to the second side door motor steering value, and adjusting the preset PWM duty cycle to the second PWM duty cycle initial value 2; According to the second PWM duty cycle initial value 2 and the preset PWM duty cycle, obtaining a second PWM duty cycle change rate; A second door operation state value is acquired based on the second side door motor steering value and the second PWM duty cycle change rate.

5. The method for calibrating door switch control parameters according to claim 3, characterized in that: The step of adjusting a preset side door motor steering value and a preset PWM duty cycle to obtain a door operation state value corresponding to the door switch operation includes: During each simulated door closing operation, the preset side door motor steering value is adjusted to the third side door motor steering value, and the preset PWM duty cycle is adjusted to the third PWM duty cycle initial value 3; According to the third PWM duty cycle initial value 3 and the preset PWM duty cycle, obtaining a third PWM duty cycle change rate; Based on the third side door motor steering value and the third PWM duty cycle change rate, obtaining a third door operation state value; After determining that the third door operation state value reaches the preset third stage threshold, entering the door closing action acceleration or deceleration stage, adjusting the preset side door motor steering value to the fourth side door motor steering value, and adjusting the preset PWM duty cycle to the fourth PWM duty cycle initial value 4; According to the fourth PWM duty cycle initial value 4 and the preset PWM duty cycle, obtaining a fourth PWM duty cycle change rate; A fourth door operation state value is acquired based on the fourth side door motor steering value and the fourth PWM duty cycle change rate.

6. A method for calibrating door switch control parameters according to claim 4, characterized in that: When determining that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation includes: Determining that the second door operation state value reaches a preset second stage threshold; Get the door opening time corresponding to the opening operation.

7. The method for calibrating door switch control parameters according to claim 5, characterized in that: When determining that the door operation state value satisfies a preset door operation condition, obtaining a switch time corresponding to each door switch operation includes: Determining that the fourth door operation state value reaches a preset fourth stage threshold; Get the door closing duration corresponding to the closing operation.

8. A method for calibrating door switch control parameters according to any one of claims 4 to 5, characterized in that: The obtaining of the sample door switch control parameters corresponding to the door switch operation simulated by the artificial intelligence model under each of the sample operating condition data includes: Obtaining the door opening duration corresponding to the opening operation, and when determining that the door opening duration is within a preset duration threshold range, outputting the first side door motor steering value, the first PWM duty cycle change rate, the second side door motor steering value and the second PWM duty cycle change rate as the door opening control parameters corresponding to the opening operation; Obtain the door closing duration corresponding to the closing operation, and when it is determined that the door closing duration is within the duration threshold range, output the third side door motor steering value, the third PWM duty cycle change rate, the fourth side door motor steering value and the fourth PWM duty cycle change rate as the door closing control parameters corresponding to the closing operation.

9. The method for calibrating door switch control parameters according to claim 8, characterized in that: The step of completing the calibration of the door switch control parameters of the vehicle to be calibrated according to the sample door switch control parameters includes: According to the sample door switch control parameters corresponding to each door switch operation, fitting the structural component coefficient change curve of the vehicle to be calibrated to obtain a first fitting door switch control parameter; According to the first fitting door switch control parameter, fitting the motor assembly change curve of the vehicle to be calibrated to obtain the second fitting door switch control parameter; When the difference between the second fitting door switch control parameter and the sample door switch control parameter is within a preset tolerance threshold range, the sample door switch control parameter is written into the vehicle to be calibrated to complete the calibration of the door control parameters of the vehicle to be calibrated.

10. A door switch control parameter calibration system, characterized in that: The system includes a training module and a calibration module; The training module is used to train the pre-built artificial intelligence model according to a plurality of preset sample operating condition data, and obtain the sample door switch control parameters corresponding to the artificial intelligence model when simulating the door switch operation under each of the sample operating condition data; The calibration module is used to complete the calibration of the door switch control parameters of the vehicle to be calibrated according to the sample door switch control parameters.

11. An electronic device, characterized in that: include: at least one processor; And a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute a method for calibrating a door switch control parameter as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute a method for calibrating a door switch control parameter according to any one of claims 1 to 9.

Images