Pedal actuating mechanism control method and system for rotating hub verification

Abstract

The invention relates to a pedal actuating mechanism control method and system for rotating hub verification, and the method comprises the steps: carrying out the assignment of automatic driving control strategy parameters according to the value range of the automatic driving control strategy parameters, and the automatic driving control strategy parameters comprise conventional parameters and newly-added parameters established for the verification characteristics of a rotating hub; according to the automatic driving control strategy parameter and the assignment thereof, carrying out pilot run on the vehicle, and obtaining a real-time vehicle condition curve; performing data fitting comparison analysis on the real-time vehicle condition curve and a preset vehicle condition curve on the basis of a redriving control theory, and determining abnormal driving influence parameters in automatic driving control strategy parameters through a preset parameter correction strategy; and according to the value range of the automatic driving control strategy parameter, the abnormal driving influence parameter is re-assigned until the real-time vehicle condition curve is consistent with a preset vehicle condition curve. Hardware transformation is not needed, the car following curve fitting degree is remarkably improved through algorithm and strategy design, and oil consumption is reduced to be close to the manual driving level.

Description

TECHNICAL FIELD

[0001] The application relates to the field of rotating hub experiments, in particular to a pedal actuator control method for rotating hub verification and a system thereof. BACKGROUND

[0002] In the verification of certain user scenarios on an AMT commercial vehicle rotating hub test bench, the matching of the vehicle curve and the fuel consumption are difficult problems in the industry.

[0003] The application of a two-foot pedal actuator in the field of automatic driving of automobiles faces challenges in execution effect, algorithm, cost, function and other aspects. For the verification control scheme of ATM commercial vehicles, the basic index requirements of laboratory verification cannot be met after years of practice; if AI is transformed, the cost of models and AI new technologies is high, and the overall technology is still in the laboratory and high-end application stage, and the application implementation is poor. Therefore, there is currently no solution on the market that can simulate human driving, meet the vehicle following and fuel consumption performance higher than the driver's index in the test process. In the future, with the progress of technology, the defects of the pedal actuator can be gradually overcome, but its application scenarios are still limited in the short term. SUMMARY

[0004] The application provides a pedal actuator control method for rotating hub verification and a system thereof, which can solve the problem in the prior art that the test scheme of the pedal actuator in the test process cannot meet the requirements of vehicle following and fuel consumption performance higher than the driver's index.

[0005] In a first aspect, an embodiment of the application provides a pedal actuator control method for rotating hub verification, which comprises: According to the value range of the automatic driving control strategy parameters, the automatic driving control strategy parameters are valued, the automatic driving control strategy parameters include conventional parameters and newly added parameters established for rotating hub verification characteristics, the conventional parameters include a first throttle parameter and a first pedal control parameter, and the newly added parameters include a second throttle parameter different from the first throttle parameter and a second pedal control parameter different from the first pedal control parameter; According to the automatic driving control strategy parameters and the values thereof, the vehicle is tested, and a real-time vehicle condition curve is obtained, the vehicle condition curve includes a following speed curve and a pedal opening degree curve; Based on the control theory, the real-time vehicle condition curve is compared and analyzed with a preset vehicle condition curve, and an abnormal driving influence parameter in the automatic driving control strategy parameters is determined through a preset parameter correction strategy; The abnormal driving influence parameter is revalued according to the value range of the automatic driving control strategy parameters, until the real-time vehicle condition curve is consistent with the preset vehicle condition curve.

[0006] In combination with the first aspect, in an implementation, before the automatic driving control strategy parameter is assigned a value according to the value range of the automatic driving control strategy parameter, the method further comprises: collecting and optimizing the conventional parameters, and establishing a value range of the optimized conventional parameters; establishing a new vehicle control strategy for the hub verification characteristics, and abstracting a new parameter of the new vehicle control strategy and a value range of the new parameter.

[0007] In combination with the first aspect, in an implementation, the collecting and optimizing the conventional parameters, and establishing a value range of the optimized conventional parameters, specifically comprises: obtaining initial conventional parameters; defining a value range of the initial conventional parameters, and assigning weights to the initial conventional parameters according to multi-objective optimization requirements to form initial constraint conditions for parameter optimization; constructing a UX score function based on the initial conventional parameters and the initial conventional parameter weights; generating a plurality of sets of candidate parameter combinations of the optimized conventional parameters according to a multi-objective optimization algorithm and based on the UX score function and the initial constraint conditions for parameter optimization, and taking the value range of the initial conventional parameters as the value range of the optimized conventional parameters.

[0008] In combination with the first aspect, in an implementation, the new vehicle control strategy comprises: start control strategy, slope control strategy, and sudden acceleration and deceleration control strategy.

[0009] In combination with the first aspect, in an implementation, based on the complex control theory, the real-time vehicle condition curve is compared and analyzed with a preset vehicle condition curve, and an abnormal driving influence parameter in the automatic driving control strategy parameter is determined through a preset parameter correction strategy, specifically comprising: extracting a preset vehicle condition curve from a database, the preset vehicle condition curve comprising a preset following speed curve and a preset pedal opening degree curve; aligning the real-time following speed curve with the preset following speed curve and the real-time pedal opening degree curve with the preset pedal opening degree curve to obtain abnormal point positions in the real-time following speed curve and the real-time pedal opening degree curve; determining the type of the abnormal point positions based on the complex control theory, and determining the abnormal driving influence parameter in the automatic driving control strategy parameter according to the preset parameter correction strategy.

[0010] In combination with the first aspect, in an implementation, the type of the abnormal point positions comprises: the vehicle speed error is greater than a set error threshold range; the vehicle speed amplitude is greater than an amplitude set threshold; The pedal switching frequency is greater than a set switching frequency. The pedal opening degree fluctuation amplitude in the stable speed interval is greater than a fluctuation amplitude threshold. The throttle and brake switching frequency in the long deceleration interval is greater than a first set frequency. When the accelerator is fully opened, the vehicle speed is lower than a target vehicle speed. The opening degree adjustment step is greater than a set adjustment threshold.

[0011] In combination with the first aspect, in an implementation manner, the preset parameter correction strategy comprises: The start-up stage control strategy, the dynamic response optimization strategy, the smoothness control strategy, the long deceleration interval control strategy, the throttle pedal switching strategy, and the full throttle strategy.

[0012] In combination with the first aspect, in an implementation manner, the first throttle parameter comprises: an integral coefficient, a differential coefficient, and a proportional coefficient. The first pedal control parameter comprises: a target vehicle speed, a dead band coefficient, and an opening degree adjustment step. The second throttle parameter comprises: a start-up pre-look-ahead time, a long deceleration interval maximum throttle opening degree, a maximum throttle opening degree, a full throttle opening degree detection time, a full throttle opening degree return proportion, and a full throttle opening degree return time. The second pedal control parameter comprises: an integral limit value and a maximum brake opening degree.

[0013] In combination with the first aspect, in an implementation manner, the start-up stage control strategy comprises: if a real-time start-up speed is less than a target start-up speed threshold, correcting the target vehicle speed, the dead band coefficient, and the start-up pre-look-ahead time. The dynamic response optimization strategy comprises: if a vehicle speed amplitude is greater than a target amplitude threshold, reducing a pedal control frequency, and correcting the integral coefficient and the differential coefficient. The smoothness control strategy comprises: if a pedal opening degree fluctuation amplitude in a stable speed interval is greater than a fluctuation amplitude threshold, correcting a proportional coefficient and an opening degree adjustment step. The long deceleration interval control strategy comprises: if a throttle and brake switching frequency in a long deceleration interval is greater than a first set frequency, correcting a long deceleration interval maximum throttle opening degree and an integral limit value. The throttle pedal switching strategy comprises: if a throttle pedal switching frequency is greater than a second set frequency, correcting a maximum brake opening degree and a maximum throttle opening degree. The full throttle strategy comprises: if, when the accelerator is fully opened, a vehicle speed is lower than a target vehicle speed, correcting a full throttle opening degree detection time, a full throttle opening degree return proportion, and a full throttle opening degree return time.

[0014] In a second aspect, the embodiments of the present application provide a pedal actuator control system for hub verification, comprising: a first module, a second module, a third module and a fourth module; the first module is configured to: assign values to automatic driving control strategy parameters according to the value range of the automatic driving control strategy parameters, the automatic driving control strategy parameters comprising conventional parameters and newly added parameters established for hub verification characteristics, the conventional parameters comprising a first throttle parameter and a first pedal control parameter, and the newly added parameters comprising a second throttle parameter different from the first throttle parameter and a second pedal control parameter different from the first pedal control parameter; the second module is configured to: perform a vehicle trial run according to the automatic driving control strategy parameters and the assigned values, and obtain a real-time vehicle condition curve, the vehicle condition curve comprising a following vehicle speed curve and a pedal opening degree curve; the third module is configured to: based on a complex disc control theory, perform data fitting and comparison analysis on the real-time vehicle condition curve and a preset vehicle condition curve, and determine an abnormal driving influence parameter in the automatic driving control strategy parameters through a preset parameter correction strategy; and the fourth module is configured to: reassign values to the abnormal driving influence parameter according to the value range of the automatic driving control strategy parameters, until the real-time vehicle condition curve is consistent with the preset vehicle condition curve.

[0015] The technical scheme provided by the embodiments of the present application has the following beneficial effects: The embodiments of the present application provide a pedal actuator control method and system for hub verification, which do not require hardware modification, but only adjust AI parameters and driver experience parameters to obtain a combination of the first throttle parameter, the first pedal control parameter, the second throttle parameter and the second pedal control parameter, thereby achieving balance optimization of following vehicle precision and fuel consumption; through data fitting and comparison analysis of the real-time vehicle condition curve and the preset vehicle condition curve, the abnormal driving influence parameter is accurately located, and the abnormal driving influence parameter is corrected to make the real-time vehicle condition curve consistent with the preset vehicle condition curve; under the premise of not requiring hardware modification, the method significantly improves the following vehicle curve fitting degree and reduces fuel consumption to approach the level of manual driving, thereby realizing efficient and low-cost landing of automatic driving in hub verification. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of the pedal actuator control method for hub verification of the present application; Figure 2 A schematic diagram of the driver driving curve and fuel consumption on a plain highway of the present application; Figure 3 A schematic diagram of the robot driving curve and fuel consumption before improvement on a plain highway of the present application; Figure 4 A schematic diagram of the robot driving curve and fuel consumption after improvement on a plain highway of the present application; Figure 5Figure 1 is a schematic diagram of a driver's driving curve and fuel consumption in a CHTC-TT road condition according to the present application; Figure 6 Figure 2 is a schematic diagram of a robot's driving curve and fuel consumption in a CHTC-TT road condition before improvement according to the present application; Figure 7 Figure 3 is a schematic diagram of a robot's driving curve and fuel consumption in a CHTC-TT road condition after improvement according to the present application. DETAILED DESCRIPTION

[0017] In order to enable persons skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor fall within the scope of protection of the present application.

[0018] The present application provides a pedal actuator control method and system for hub verification, which can solve the problem that the test scheme of the pedal actuator in the prior art cannot meet the requirements of following the vehicle and the fuel consumption performance being higher than the driver's index in the test process.

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be described in further detail below in combination with the drawings.

[0020] In a first aspect, the embodiments of the present application provide a pedal actuator control method for hub verification, which comprises: 101: assigning values to the automatic driving control strategy parameters according to the value range of the automatic driving control strategy parameters; 102: running the vehicle according to the automatic driving control strategy parameters and their assigned values, and obtaining a real-time vehicle condition curve; 103: based on the control theory of complex disc, comparing and analyzing the real-time vehicle condition curve with a preset vehicle condition curve, and determining an abnormal driving influence parameter in the automatic driving control strategy parameters through a preset parameter correction strategy; 104: reassigning values to the abnormal driving influence parameter according to the value range of the automatic driving control strategy parameters, until the real-time vehicle condition curve is consistent with the preset vehicle condition curve.

[0021] In the present application, without hardware modification, only by adjusting the AI parameters and the driver experience parameters, the combination of the first throttle parameter, the first pedal control parameter, the second throttle parameter and the second pedal control parameter is obtained, the balance optimization of following vehicle precision and fuel consumption is realized; through the fitting comparison analysis of the real-time vehicle condition curve and the preset vehicle condition curve, the abnormal driving influence parameter is accurately positioned, and the abnormal driving influence parameter is corrected to make the real-time vehicle condition curve consistent with the preset vehicle condition curve; without hardware modification, the method significantly improves the fitting degree of the following vehicle curve, reduces the fuel consumption to approach the artificial driving level, and realizes the efficient and low-cost landing of automatic driving in the hub verification.

[0022] Among them, the automatic driving control strategy parameters include conventional parameters and newly added parameters established for the hub verification characteristics, the conventional parameters include the first throttle parameter and the first pedal control parameter, and the newly added parameters include the second throttle parameter different from the first throttle parameter and the second pedal control parameter different from the first pedal control parameter; the vehicle condition curve includes the following vehicle speed curve and the pedal opening degree curve.

[0023] The core of the control theory is to analyze the difference between the real-time vehicle condition curve and the preset vehicle condition curve, combine the experience data of the expert system and the AI algorithm, identify and correct the key parameters affecting the control performance, and form a closed loop optimization. Through the combination of data driving (real-time curve fitting) and experience driving (expert system analysis), the complex working condition adaptation that the traditional PID control cannot cover is realized, and finally the following vehicle curve fitting degree is improved to 98.2%, the fuel consumption is reduced, and the pedal action frequency is reduced, which significantly improves the efficiency and economy of the hub verification.

[0024] Before detailing the present scheme, first introduce the design basis conditions of the present scheme: The investment in hub verification is large, the hardware equipment is basically fixed, and the hardware also has certain application in other fields with low precision requirements, so the hardware modification is basically impossible; if the hardware is modified, the investment will be increased.

[0025] The existing software design scheme is: according to the following vehicle speed, calculate the pedal opening degree at 0.1 second frequency, when the actual vehicle speed ≤ following vehicle speed, increase the throttle pedal until the actual vehicle speed ≥ following vehicle speed; when the actual vehicle speed ≥ following vehicle speed, reduce the throttle pedal and increase the brake pedal opening degree. This scheme has two key index problems that have existed after years of optimization and modification by industry assistance and algorithm: the robot simulation user driving following vehicle curve cannot be matched, the following vehicle deviation exceeds the national standard, and the verification demand cannot be met; the robot driving fuel consumption is much higher than the driver, and the national standard requirement cannot be met.

[0026] Through the analysis of the existing design scheme, it is found that if the national standard is to be met and the following vehicle deviation and fuel consumption are to fully meet the requirements, the scene with high fuel consumption needs to be analyzed and a solution is to be given, and the problem of following the trajectory needs to be precisely controlled. If the existing equipment is completely replaced by big data and AI model, the computing power and communication cannot meet the requirements, and the cost of upgrading the hardware will be geometric, and the system will be completely abandoned, which will lose the meaning of the project. Therefore, the design can only be made from the perspective of the combination of software and hardware.

[0027] In addition, since the speed accuracy is directly related to the sensitivity of the PID control, if the traditional scheme strictly retains 4-5 decimal places, the PID calculation will be too sensitive. For example: a small fluctuation in vehicle speed (such as 0.001 km / h) may be misjudged as an error by the PID, triggering frequent pedal adjustments (such as throttle opening from 50% to 55% to 50%), and further exacerbating the speed amplitude and error accumulation. In the present application, after data acquisition, the vehicle speed data is preprocessed to retain 1-2 decimal places, the PID calculation frequency is reduced, and the pedal action is reduced.

[0028] The traditional PID control only relies on the current speed, which is easy to cause the following vehicle to lag or overshoot. In the present application, by analyzing the current speed and the speed to be driven, the pedal action is optimized by using the principle of vehicle inertia.

[0029] The present application is designed by considering the above factors. The expected goals of the present application include: the synchronization rate of the robot driving behavior and the human driver is greater than or equal to 95%; the key indicators such as the energy consumption per 100 kilometers, the following vehicle overshoot times and duration are close to or exceed the driving performance of an experienced driver.

[0030] In the present application, an AI online driving parameter optimizer is used to obtain the automatic driving control strategy parameters; during the trial operation of the vehicle, real-time vehicle condition data is collected through the control box system, and a real-time vehicle condition curve is obtained. In the implementation process of the scheme, a driving expert digital system is used, new parameters are added, possible problems are analyzed, and adjustment strategies are supplemented. Through the combination of the AI parameter optimizer and the expert experience database, the expert system is improved, and the precise control of the automatic driving pedal actuator in the hub verification is realized.

[0031] On the basis of the above embodiment, in the present embodiment, before the automatic driving control strategy parameters are valued according to the value range of the automatic driving control strategy parameters, the method further includes: Collect and optimize the conventional parameters, and establish the value range of the optimized conventional parameters: read the vehicle condition data through the pedal actuator control box system, optimize it in combination with the experienced driver experience database, and form the conventional parameters and the value range of the conventional parameters.

[0032] A new vehicle control strategy is established for the hub verification characteristics, and the new parameters of the new vehicle control strategy and the value range of the new parameters are abstracted: various abnormal conditions of data after years of hub verification are analyzed, including a large number of vehicle speeds that cannot follow the following speed, large offset, severe pedal control oscillation, pedal frequency uniformity, abnormal increase, and abnormal high and low pedal PID control value. Solutions are given for each abnormal state, and new control strategy parameters are abstracted, and value ranges are given for each parameter.

[0033] Among them, the conventional parameters are collected and optimized, and the value range of the optimized conventional parameters is established, based on the AI online driving parameter optimizer and the multi-objective optimization algorithm, the parameter range and weight are dynamically adjusted, and the candidate parameter combination is generated by combining the weighted scoring model, which specifically includes: First, the initial conventional parameters are obtained; then the value range of the initial conventional parameters is defined, and the initial conventional parameter weight is allocated according to the multi-objective optimization requirement to form the initial constraint condition of parameter optimization; then based on the initial conventional parameters and the initial conventional parameter weight, the UX score function is constructed; finally, according to the multi-objective optimization algorithm, and based on the UX score function and the initial constraint condition of parameter optimization, a plurality of groups of candidate parameter combinations of optimized conventional parameters are generated, and the value range of the initial conventional parameters is taken as the value range of the optimized conventional parameters.

[0034] Specifically, specifically, the initial conventional parameters include proportional coefficient Kp, integral coefficient Ki, differential coefficient Kd, throttle opening, etc., and their value ranges need to be defined according to the hub verification scene (such as plain high speed, long deceleration interval), and the weight is allocated to balance fuel consumption, following accuracy and dynamic response. The UX score function quantifies the influence of parameters on user experience by linearly superimposing the normalized scores of each parameter and the multi-objective compensation term. The multi-objective optimization algorithm (such as NSGA-III) searches for Pareto optimal solutions in a 14-dimensional parameter space, generates a plurality of groups of candidate parameter combinations of optimized conventional parameters, and needs to meet the constraint conditions (such as pedal frequency ≤100Hz, rapid throttle change rate ≤30% / 100ms).

[0035] Among them, the new vehicle control strategy is established for the hub verification characteristics, and the new parameters of the new vehicle control strategy and the value range of the new parameters are abstracted, which specifically includes: For the verification characteristics of the hub, by combining the fixed slope, no real traffic interference and other scene characteristics, the vehicle adds control strategy and multiple parameters are designed to solve the shortcomings of traditional PID control in following precision, fuel consumption and pedal stability. In this embodiment, the vehicle adds control strategy includes: start control strategy, slope control strategy and sudden acceleration and deceleration control strategy. The added parameters include start preview time (1-2 seconds), maximum throttle opening in long deceleration interval (40-60%), sudden acceleration throttle threshold (70-90%), sudden acceleration throttle detection time (3-5 seconds), full throttle opening return ratio (10-30%) and return time (3-6 seconds), etc. The value range of these parameters is dynamically adjusted based on the experienced driver experience database and actual verification data, for example, increasing the start preview time from 1 second to 2 seconds can improve the start response speed, and limiting the long deceleration interval throttle to 60% to reduce frequent switching. Through the NSGA-III multi-objective optimization algorithm of the AI online driving parameter optimizer, the added parameters are combined with the UX score function, and the compensation points are calculated by the Pareto level and the constraint satisfaction rate to ensure that the parameter combination meets the following error ≤±2km / h, fuel consumption close to the artificial driving level and pedal action smoothness requirements. For example, the sudden acceleration throttle strategy suppresses overshoot through the threshold value of 70% and the detection time of 5 seconds, and the full throttle return ratio of 30% and the time of 6 seconds reduces energy consumption. Finally, the added parameters and the conventional parameters form a cooperative control, and the dynamic weight distribution further optimizes the scene adaptability, so that the following curve fitting degree reaches 98.2%, the overage times are reduced, and the hardware modification cost is significantly reduced.

[0036] Among them, the start control strategy is:

[0037] In the start control strategy, the start is determined according to the following speed. When the following speed is low, the acceleration preparation is made in advance, considering the engine delay reaction problem, then the speed reaches the following speed ±2, and the pedal is-10%. When the following speed is too large, considering the ATM commercial vehicle national standard speed limit and other factors, the pedal is to 90%, the pedal is-30%; the speed reaches the following speed ±2, and the pedal is-10% to improve energy efficiency.

[0038] Slope control strategy:

[0039] The uphill strategy in the slope control strategy: according to the slope size and the current engine speed data, the throttle pedal can be set to 30%, when the speed is too high, the throttle is released or adjusted to 10% opening, instead of the brake pedal strategy; in order to accelerate, the throttle is not increased to 90% or 100%, etc.

[0040] In the downhill strategy, to accelerate, a little throttle can be taken, such as 5% of the accelerator pedal, or even no throttle, instead of the old strategy of pressing the accelerator to 70-80% or more... In the sudden acceleration and deceleration control strategy: the pedal opening degree is increased by 30%, the pedal reaches 90%, the pedal is -30%, and then the pedal opening degree is increased by 20%; the speed reaches the following vehicle speed ± 2, and the pedal is -10%.

[0041] According to the practice, it is proved that after the implementation of the scheme, the following curve and the fuel consumption per 100 kilometers are obviously close to the driver or even better than the driver, the fitting degree of the following curve is improved to 98.2%, and the fuel consumption is greatly reduced by about 10% compared with manual driving; it will have a profound impact on future hub verification, save a lot of manpower and material resources, and produce relatively far-reaching economic benefits.

[0042] In the embodiment, the first throttle parameter and its value range, the first pedal control parameter and its value range, the second throttle parameter and its value range, and the second pedal control parameter and its value range are: The first throttle parameter and its value range include: integral coefficient Ki (%) (1, 2, 3), differential coefficient Kd (%) (0, 1), proportional coefficient Kp (0.1, 0.2, 0.3), integral limit value (1, 3, 5), throttle opening degree (%) (10-100), and return throttle opening degree (%) (10-100); The first pedal control parameter and its value range include: target vehicle speed (km / h) (30-95), idle stroke coefficient (5%-30%), opening degree adjustment step (10, 20, 30), tolerance band range (km / h) (1, 2), pedal switching time interval (s) (0.1, 0.2), proportional coefficient p in the pid calculation formula (0.1, 0.2, 0.3, 0.4, 0.5); The second throttle parameter and its value range include: start-up preview advance time (s) (1, 2), maximum throttle opening degree (%) (40, 50, 60) in the long deceleration interval, maximum throttle opening degree (%) (70, 80, 90, 100), full throttle opening degree detection time (s) (5, 10), full throttle opening degree return proportion (%) (10, 20, 30), and full throttle opening degree return time (s) (3, 6), sudden throttle opening degree threshold value (%) (70, 80, 90), sudden throttle opening degree detection time (s) (3, 5), sudden throttle opening degree return proportion (%) (10, 20, 30), sudden throttle opening degree strategy (0, 1), and full throttle opening degree strategy (0, 1); The second pedal control parameter and its value range include: control strategy (whether with slope) (yes, no), integral limit value (3, 5), vehicle speed decimal point retention bit number (0, 1, 2), and maximum brake opening degree (%) (60, 70, 80).

[0043] On the basis of the above-mentioned embodiments, in this embodiment, according to the automatic driving control strategy parameters and their assignments, the vehicle is tested and run, and real-time vehicle condition curves are obtained, including a following speed curve and a pedal opening degree curve, specifically: Based on the test run process of the automatic driving control strategy parameters and their assignments, combined with the AI online driving parameter optimizer and the experienced driver experience database, the collection and analysis of the real-time vehicle condition curves (including the following speed curve and the pedal opening degree curve) are the core links of the optimization of the control strategy.

[0044] During the test run stage, specific scenarios (such as plain highway and long deceleration interval) are simulated on the hub bench, and vehicle dynamic data are collected in real time through the control box system to generate the following speed curve and the pedal opening degree curve.

[0045] On the basis of the above-mentioned embodiments, in this embodiment, the real-time vehicle condition curve is compared and analyzed with the preset vehicle condition curve, and the abnormal driving influence parameters in the automatic driving control strategy parameters are determined through the preset parameter correction strategy. When the real-time vehicle condition curve is compared and analyzed with the preset vehicle condition curve, the abnormality is determined by quantifying the numerical difference (such as speed error and pedal opening degree mutation rate) and the overall linear feature (such as amplitude and frequency) of the key points of the real-time curve and the preset curve. Specifically, the comparison and analysis need to be combined with the scoring model of the experienced driver experience database and the AI online driving parameter optimizer.

[0046] Specifically, steps 1031 to 1033 are included: Step 1031: Extracting the preset vehicle condition curve from the database, the preset vehicle condition curve including a preset following speed curve and a preset pedal opening degree curve.

[0047] Specifically, the preset vehicle condition curve (preset following speed curve and preset pedal opening degree curve) extracted from the database is a benchmark based on the experienced driver experience data, which is used to locate the abnormal points by comparing with the real-time vehicle condition curve. The preset following speed curve is generated by recording the speed during manual driving, for example, in the plain highway scenario, the preset curve needs to ensure that the average error is ≤±1.5 km / h, and the number of out-of-tolerance times is ≤2 (national standard requirement), and in the long deceleration interval scenario, the speed fluctuation needs to be avoided (such as amplitude ≤3 km / h / s); the preset pedal opening degree curve is generated by recording the throttle / brake opening degree change during manual driving, for example, in the starting stage, the preset curve requires the throttle opening degree to gradually increase from 10% to 60% (air travel coefficient ≤15%), avoiding overshoot caused by sudden acceleration, and in the long deceleration interval, the throttle opening degree upper limit needs to be limited (such as ≤60%) to reduce frequent switching.

[0048] Step 1032: Align the real-time following speed curve with the preset following speed curve, and align the real-time pedal opening degree curve with the preset pedal opening degree curve, to obtain abnormal point positions in the real-time following speed curve and the real-time pedal opening degree curve.

[0049] Specifically, when aligning the real-time following speed curve with the preset following speed curve, time stamp matching or linear interpolation algorithm is used to ensure time axis consistency, then error value (such as real-time vehicle speed-preset vehicle speed) is calculated point by point, and error accumulation (such as sliding window analysis of error mean value within 5 seconds), over-limit times (such as error > ±2km / h is counted as an abnormal point position) and amplitude (such as vehicle speed change rate > 3km / h / s is marked as an oscillation point position) are counted.

[0050] The alignment of the real-time pedal opening degree curve with the preset pedal opening degree curve also needs to pay attention to the opening degree mutation rate (such as Δ opening degree / Δ time > 30% / 100ms is marked as abnormal) and switching frequency (such as switching interval < 0.1s is marked as high-frequency switching point position), for example, when the start preview time is insufficient, the real-time pedal opening degree curve may show that the throttle opening degree suddenly increases to 90%, while the preset pedal opening degree curve requires the opening degree to increase to 60% (idle stroke coefficient = 15%), at this time, by adjusting the start preview advance time (such as from 1s to 2s), the overshoot can be reduced.

[0051] Step 1033: Determine the type of abnormal point position, and determine the abnormal driving influence parameter in the automatic driving control strategy parameter according to the preset parameter correction strategy.

[0052] Among them, the type of abnormal point position includes: the vehicle speed error is greater than the set error threshold range; the vehicle speed amplitude is greater than the amplitude set threshold; the pedal switching switching frequency is greater than the set switching frequency; the pedal opening degree fluctuation amplitude in the stable speed interval is greater than the fluctuation amplitude threshold; the throttle and brake switching frequency in the long deceleration interval is greater than the first set frequency; when the throttle is full, the vehicle speed is lower than the target vehicle speed; the opening degree adjustment step is greater than the set adjustment threshold.

[0053] In this embodiment, the preset parameter correction strategy includes: start-up stage control strategy, dynamic response optimization strategy, smoothness control strategy, long deceleration interval control strategy, throttle pedal switching strategy and full throttle strategy.

[0054] Specifically, the fusion logic of the type of abnormal point position and the preset parameter correction strategy is realized by AI online driving parameter optimizer and experienced driver experience database to realize closed-loop control.

[0055] When the real-time vehicle condition curve (following speed curve and pedal opening curve) is compared with the preset vehicle condition curve, if the vehicle speed error exceeds ±2 km / h (national standard requirement), the starting stage control strategy is triggered. For example, when the vehicle speed exceeds the error too many times, the integral coefficient Ki needs to be reduced (such as from 3 to 2) to suppress error accumulation, or the differential coefficient Kd needs to be increased (such as from 1 to 1.5) to smooth the speed change; when the pedal opening oscillates, the opening adjustment step needs to be reduced (such as from 30% to 10%) and the switching time interval needs to be extended (such as from 0.1s to 0.15s), for example, in a long deceleration interval, the real-time vehicle condition curve shows that the throttle opening frequently switches between 50% and 60%, while the preset vehicle condition curve requires the opening to be stable at 55%, at this time the maximum throttle opening in the long deceleration interval needs to be limited (such as from 80% to 60%) and the integral limit value needs to be adjusted (such as from 5 to 3) to reduce the action frequency. Finally, through the NSGA-III multi-objective optimization algorithm of the AI online driving parameter optimizer, combined with the UX score function to iteratively generate candidate parameter combinations, the compensation term is dynamically adjusted through the Pareto level and the constraint satisfaction rate to ensure that the optimized parameters achieve a following accuracy of ≥98.2% and fuel consumption close to the level of manual driving under the national standard requirements, while adapting to different verification requirements through dynamic weight allocation to form a closed-loop control logic.

[0056] Further, the starting stage control strategy includes: if the real-time starting speed is less than the target starting speed threshold, then correcting the target vehicle speed, the idle stroke coefficient and the start preview advance time; The dynamic response optimization strategy includes: if the vehicle speed amplitude is greater than the target amplitude threshold, then reducing the pedal control frequency, and correcting the integral coefficient and the differential coefficient; The smoothness control strategy includes: if the pedal opening fluctuation amplitude is greater than the fluctuation amplitude threshold in the stable speed interval, then correcting the proportional coefficient and the opening adjustment step; The long deceleration interval control strategy includes: if the throttle and brake switching frequency is greater than the first set frequency in the long deceleration interval, then correcting the maximum throttle opening in the long deceleration interval and the integral limit value; The throttle pedal switching strategy includes: if the throttle pedal switching frequency is greater than the second set frequency, then correcting the maximum brake opening and the maximum throttle opening; The full throttle strategy includes: if the vehicle speed is lower than the target vehicle speed when the full throttle is applied, then correcting the full throttle detection time, the full throttle return proportion and the full throttle return time.

[0057] Wherein, the starting speed does not follow, that is, the real-time starting speed is less than the target starting speed threshold, which means that the throttle (brake) is given but the initial acceleration (deceleration) is insufficient, then the initial speed parameter can be increased, the idle stroke throttle parameter can be increased, and the start preview advance time can be corrected.

[0058] Specifically, during the start-up phase, if insufficient vehicle acceleration leads to a lag in vehicle speed, the response speed needs to be optimized by increasing the initial speed parameter (e.g., increasing the target speed from 30 km / h to 35 km / h), increasing the idle travel throttle parameter (e.g., increasing the idle travel coefficient from 15% to 20%), and correcting the start-up aiming advance time (e.g., increasing it from 1s to 2s).

[0059] Severe fluctuations in speed amplitude indicate that the accumulated speed error is too large and the road conditions have deteriorated. Therefore, the PID controller should be continuously adjusted. This can be done by reducing the pedal control frequency parameter or adjusting the initial speed parameter.

[0060] Specifically, when the speed amplitude fluctuates violently, the pedal control frequency needs to be reduced (e.g., from 100Hz to 50Hz) or the initial speed parameters need to be adjusted, consistent with the strategy of suppressing overshoot through the integral coefficient Ki and the derivative coefficient Kd.

[0061] Excessive error: ±2 deviation is within the national standard allowable range. If the accumulated error is too large, it will cause abnormal PID adjustment. In this case, the integral coefficient and derivative coefficient can be adjusted.

[0062] Specifically, when the accumulated vehicle speed error exceeds the national standard allowable range (±2 km / h), adjusting the integral coefficient Ki and the derivative coefficient Kd (e.g., Ki=2, Kd=1.5) can reduce over-adjustment of the PID controller. Frequent accelerator pedal switching: If the frequency is too high, it indicates that the PID accuracy is too high, and it is necessary to set the maximum brake opening and the maximum throttle amplitude.

[0063] Specifically, when the accelerator pedal is switched frequently, if the switching frequency of the accelerator pedal is greater than the second set frequency, the maximum throttle amplitude (e.g., reduced from 100% to 80%) and the maximum brake opening (e.g., reduced from 70% to 50%) need to be set.

[0064] If the pedal opening oscillates within a relatively stable speed range, it indicates that the PID precision is too high, affecting the proportional coefficient p in the PID calculation formula.

[0065] Specifically, if the pedal opening fluctuation exceeds the fluctuation threshold within the stable speed range, the proportional coefficient P (e.g., from 0.5 to 0.3) and the opening adjustment step size (e.g., from 30% to 10%) need to be reduced.

[0066] Starting aiming advance time: Increase this parameter when the vehicle starts slowly, and decrease it when the vehicle starts too quickly.

[0067] If the vehicle speed fluctuates more easily on and off the curve, it indicates that the error range is too large, and the integral coefficient and derivative coefficient need to be lowered. If frequent switching between throttle and brake occurs during a long deceleration range, it indicates that the throttle is too large and the throttle opening needs to be limited. The maximum throttle opening during the long deceleration range needs to be reduced.

[0068] Pedal switching time interval, avoid fueling after braking, then adjust the time interval between pedal control; When the full throttle, the car still can not keep up with the speed, then need to back to the throttle after fueling strategy, generally see more high-speed section, continuous full throttle time long; The opening adjustment step, the greater the opening, the greater the return within a certain time, the parameter is associated with the pedal switching time interval, the parameter is too large, on the slope of the acceleration will be displayed speed can not keep up, but the fuel consumption is greatly increased, and the speed is too large, and the brake response is frequent.

[0069] In summary, through the design scheme of the application, the following curve and fuel consumption per 100 kilometers are close to or even better than the driver, the fitting degree of the following curve is improved to 98.2%, especially the fuel consumption is greatly reduced by about 10% compared with manual driving; It will have a profound impact on future hub verification, save a lot of manpower and material resources, and produce relatively far-reaching economic benefits.

[0070] For example, for plain high-speed road conditions: The robot automatically sets the following parameters before the conventional parameter optimization: The proportional coefficient p in the pid calculation formula: 0.5; integral limit value: 5; pedal control frequency: 0.2 seconds; maximum throttle for long deceleration: 90%; maximum brake: 70%; throttle and brake step: 5%; idle stroke: throttle 19%, brake 23%; other parameters take the default values obtained by the vehicle.

[0071] After the key parameter optimization by the system: the proportional coefficient p in the pid calculation formula: 0.2; integral limit value: 3; pedal control frequency: 0.1 seconds; maximum throttle for long deceleration: 50%; maximum brake: 50%; throttle and brake step: 2%; when the throttle opening is greater than the threshold value 70%, reduce back 30%, time interval 3 seconds; idle stroke: throttle 21%, brake 21%; other parameters take the default values obtained by the vehicle.

[0072] After the hub real vehicle verification, the following and fuel consumption are greatly optimized, which is better than the national standard requirement:

[0073] For CHTC-TT road conditions: The robot automatically sets the following parameters before the conventional parameter optimization: The proportional coefficient p in the pid calculation formula: 0.5; integral limit value: 5; pedal control frequency: 0.3 seconds; maximum throttle for long deceleration: 100%; maximum brake: 90%; throttle and brake step: 4%; idle stroke: throttle 19%, brake 18%; other parameters take the default values obtained by the vehicle.

[0074] After the key parameter optimization by the system: the proportional coefficient p in the pid calculation formula: 0.3; integral limit value: 3; pedal control frequency: 0.1 seconds; long deceleration maximum throttle: 50%; maximum brake: 50%; throttle and brake step size: 2%; when the throttle opening is greater than the threshold value 90%, reduce by 30% in 3 seconds; idle stroke: throttle 23%, brake 32%; other parameters take the default values obtained by the vehicle.

[0075] After the hub real vehicle verification, the follow-up and fuel consumption are greatly optimized, which is better than the national standard requirements.

[0076] In a second aspect, the embodiments of the present application provide a pedal actuator control system for hub verification, comprising: a first module, a second module, a third module and a fourth module; the first module is configured to: according to the value range of the automatic driving control strategy parameters, assign values to the automatic driving control strategy parameters, the automatic driving control strategy parameters including regular parameters and newly added parameters established for hub verification characteristics, the regular parameters including a first throttle parameter and a first pedal control parameter, and the newly added parameters including a second throttle parameter different from the first throttle parameter and a second pedal control parameter different from the first pedal control parameter; the second module is configured to: according to the automatic driving control strategy parameters and their assigned values, run the vehicle for trial, and obtain real-time vehicle condition curves, the vehicle condition curves including follow-up speed curves and pedal opening degree curves; the third module is configured to: based on the complex disc control theory, perform data fitting and comparative analysis on the real-time vehicle condition curves and preset vehicle condition curves, and determine abnormal driving influence parameters in the automatic driving control strategy parameters through a preset parameter correction strategy; and the fourth module is configured to: according to the value range of the automatic driving control strategy parameters, reassign values to the abnormal driving influence parameters until the real-time vehicle condition curves and the preset vehicle condition curves are consistent.

[0077] In the present application, without hardware modification, only by adjusting the AI parameters and the driver experience parameters, the combination of the first throttle parameter, the first pedal control parameter, the second throttle parameter and the second pedal control parameter is obtained, the balance optimization of follow-up accuracy and fuel consumption is realized; through the comparative analysis of the real-time vehicle condition curves and the preset vehicle condition curves, the abnormal driving influence parameters are accurately positioned, the abnormal driving influence parameters are corrected to make the real-time vehicle condition curves consistent with the preset vehicle condition curves; without hardware modification, through algorithm and strategy design, the follow-up curve fitting degree is significantly improved, the fuel consumption is reduced to approach the artificial driving level, and the efficient and low-cost landing of automatic driving in hub verification is realized. The functions of each module in the pedal actuator control device for hub verification correspond to the steps in the pedal actuator control method for hub verification, and the functions and implementation processes will not be repeated here.

[0078] In a third aspect, the embodiments of the present application provide a pedal actuator control device for hub verification. The pedal actuator control device for hub verification can be a personal computer (PC), a notebook computer, a server, or the like device having a data processing function.

[0079] In the embodiments of the present application, the pedal actuator control device for hub verification can include a processor, a memory, a communication interface, and a communication bus.

[0080] The communication bus can be of any type for interconnecting the processor, the memory, and the communication interface.

[0081] The communication interface includes an input / output (I / O) interface, a physical interface, and a logical interface, and the like interface for interconnecting devices inside the pedal actuator control device for hub verification, and an interface for interconnecting the pedal actuator control device for hub verification with other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, a fiber interface, an ATM interface, and the like; the user device can be a display (Display), a keyboard (Keyboard), and the like.

[0082] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), and the like.

[0083] The processor can be a general-purpose processor that can invoke a pedal actuator control program for hub verification stored in the memory and execute the pedal actuator control method for hub verification provided by the embodiments of the present application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the pedal actuator control program for hub verification is invoked can refer to each embodiment of the pedal actuator control method for hub verification of the present application, which will not be described here.

[0084] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium.

[0085] The computer readable storage medium stores a pedal actuator control program for hub verification, and when the pedal actuator control program for hub verification is executed by a processor, the steps of the pedal actuator control method for hub verification are implemented.

[0086] The method implemented when the pedal actuator control program for hub verification is executed can refer to each embodiment of the pedal actuator control method for hub verification of the present application, and will not be described here.

[0087] It should be noted that the above-mentioned sequence numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0088] The terms "comprising" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device. The terms "first", "second" and "third" and the like descriptions are used to distinguish different objects, and do not represent the order or limit the types of "first", "second" and "third".

[0089] In the description of the embodiments of the present application, "exemplary", "for example" or "for instance" is used to represent an example, illustration or description. Any embodiment or design scheme described as "exemplary", "for example" or "for instance" in the embodiments of the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the words "exemplary", "for example" or "for instance" are intended to present the relevant concept in a specific way.

[0090] In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in the text only describes the relationship between the associated objects, which means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0091] In some of the processes described in this specification, the order of operations or steps can be modified. Specifically, the serial order of any two consecutive steps carried out according to the processes described in this specification can be changed so that these two steps can be carried out in parallel or simultaneously, or the order of these two steps can be reversed.

[0092] Those skilled in the art can clearly understand the above-mentioned embodiment method from the description of the above embodiments, which can be realized by software and a necessary general hardware platform, of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disc) as described above, and includes a plurality of instructions for causing a terminal device to execute the methods described in the various embodiments of the present application.

[0093] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A control method for a pedal actuator used in hub verification, characterized in that, It includes: The autonomous driving control strategy parameters are assigned values ​​according to the value range of the autonomous driving control strategy parameters. The autonomous driving control strategy parameters include regular parameters and newly added parameters established for the hub verification characteristics. The regular parameters include a first throttle parameter and a first pedal control parameter. The newly added parameters include a second throttle parameter that is different from the first throttle parameter and a second pedal control parameter that is different from the first pedal control parameter. Based on the autonomous driving control strategy parameters and their assigned values, the vehicle is tested and real-time vehicle condition curves are obtained, including the following speed curve and the pedal opening / closing curve. Based on the retrospective control theory, the real-time vehicle condition curve is compared and analyzed with the preset vehicle condition curve, and the abnormal driving impact parameters in the autonomous driving control strategy parameters are determined through the preset parameter correction strategy. Based on the value range of the autonomous driving control strategy parameters, the parameters affecting abnormal driving are reassigned until the real-time vehicle condition curve matches the preset vehicle condition curve.

2. The pedal actuator control method for hub verification as described in claim 1, characterized in that, Before assigning values ​​to the autonomous driving control strategy parameters based on their value range, the method further includes: Collect and optimize common parameters, and establish the value range of the optimized common parameters; A new control strategy for the vehicle is established based on the characteristics of the rotating drum verification, and the new parameters of the new control strategy and the value range of the new parameters are abstracted.

3. The pedal actuator control method for hub verification as described in claim 2, characterized in that, Collect and optimize common parameters, and establish the value range of the optimized common parameters, specifically including: Obtain initial general parameters; Define the range of values ​​for the initial conventional parameters, and assign weights to the initial conventional parameters according to the multi-objective optimization requirements to form the initial constraints for parameter optimization; Construct a UX scoring function based on initial conventional parameters and their weights; Based on the multi-objective optimization algorithm and the initial constraints of the UX scoring function and parameter optimization, multiple sets of candidate parameter combinations are generated to optimize the regular parameters, and the value range of the initial regular parameters is used as the value range of the optimized regular parameters.

4. The pedal actuator control method for hub verification as described in claim 2, characterized in that, The newly added vehicle control strategies include: Starting control strategy, gradient control strategy, and rapid acceleration / deceleration control strategy.

5. The pedal actuator control method for hub verification as described in claim 1, characterized in that, Based on retrospective control theory, the real-time vehicle condition curve is compared and analyzed with the preset vehicle condition curve. An abnormal driving impact parameter in the autonomous driving control strategy is determined through a preset parameter correction strategy, specifically including: Extract preset vehicle condition curves from the database. The preset vehicle condition curves include preset following speed curves and preset pedal opening / closing curves. Align the real-time following speed curve with the preset following speed curve, and align the real-time pedal opening curve with the preset pedal opening curve to obtain abnormal points in the real-time following speed curve and the real-time pedal opening curve. Based on retrospective control theory, the types of abnormal points are determined, and the abnormal driving impact parameters in the autonomous driving control strategy parameters are determined according to the preset parameter correction strategy.

6. The pedal actuator control method for hub verification as described in claim 5, characterized in that, The types of abnormal locations include: The vehicle speed error exceeds the set error threshold range; The vehicle speed amplitude exceeds the set amplitude threshold. The pedal switching frequency is greater than the set switching frequency; The pedal opening fluctuation range is greater than the fluctuation range threshold within the stable speed range; The frequency of throttle and brake switching is greater than the first set frequency during the long deceleration range; When the accelerator is fully depressed, the vehicle speed is lower than the target speed; The opening adjustment step is greater than the set adjustment threshold.

7. The pedal actuator control method for hub verification as described in claim 1, characterized in that, The preset parameter correction strategy includes: The strategies include: start-up phase control strategy, dynamic response optimization strategy, smoothness control strategy, long deceleration range control strategy, accelerator pedal switching strategy, and full throttle strategy.

8. The pedal actuator control method for hub verification as described in claim 7, characterized in that, The first throttle parameters include: integral coefficient, derivative coefficient, and proportional coefficient; The first pedal control parameters include: target vehicle speed, idle travel coefficient, and opening adjustment step size; The second throttle parameters include: start-up aiming advance time, maximum throttle opening in long deceleration range, maximum throttle opening, full throttle opening detection time, full throttle opening return ratio, and full throttle opening return time; The second pedal control parameters include: integral limit value and maximum brake opening.

9. The pedal actuator control method for hub verification as described in claim 8, characterized in that: The start-up phase control strategy includes: if the real-time start-up speed is less than the target start-up speed threshold, then the target vehicle speed, empty travel coefficient, and start-up aiming advance time are corrected; The dynamic response optimization strategy includes: if the vehicle speed amplitude is greater than the target amplitude threshold, then reduce the pedal control frequency and correct the integral coefficient and derivative coefficient; The smoothness control strategy includes: if the pedal opening fluctuation amplitude is greater than the fluctuation amplitude threshold within the stable speed range, then the proportional coefficient and the opening adjustment step size are corrected. The long deceleration range control strategy includes: if the throttle and brake switching frequency is greater than the first set frequency in the long deceleration range, then the maximum throttle opening and integral limit value in the long deceleration range are corrected. The accelerator pedal switching strategy includes: if the accelerator pedal switching frequency is greater than the second set frequency, then the maximum brake opening and the maximum accelerator opening are corrected. The full throttle strategy includes: if the vehicle speed is lower than the target speed when the throttle is fully depressed, then the full throttle opening detection time, the full throttle opening return ratio, and the full throttle opening return time are adjusted.

10. A control system for a pedal actuator used in hub verification, characterized in that, It includes: The first module is used to: assign values ​​to the autonomous driving control strategy parameters according to the value range of the autonomous driving control strategy parameters. The autonomous driving control strategy parameters include regular parameters and newly added parameters established for the hub verification characteristics. The regular parameters include a first throttle parameter and a first pedal control parameter. The newly added parameters include a second throttle parameter different from the first throttle parameter and a second pedal control parameter different from the first pedal control parameter. The second module is used to: test run the vehicle according to the autonomous driving control strategy parameters and their values, and obtain real-time vehicle condition curves, including following speed curves and pedal opening / closing curves. The third module is used to: perform data fitting and comparison analysis between the real-time vehicle condition curve and the preset vehicle condition curve based on the retrospective control theory, and determine the abnormal driving impact parameters in the autonomous driving control strategy parameters through the preset parameter correction strategy. The fourth module is used to reassign values ​​to the abnormal driving impact parameters according to the value range of the autonomous driving control strategy parameters, until the real-time vehicle condition curve is consistent with the preset vehicle condition curve.

Images