New energy vehicle braking system and method based on machine learning

Through the new energy vehicle braking system based on machine learning, the vehicle's usage habits and pedal frequency are analyzed, the pedal amplitude range is adjusted, and early intervention and precise braking are achieved, and the problems of braking system delay and control delay in the existing technology are solved. Improves the response speed and effect of the brake system.

CN118977685APending Publication Date: 2024-11-19GUANGXI YUCHAI NEW ENERGY AUTOMOBILE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411368627.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-29
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing new energy vehicle braking system cannot intervene in advance, resulting in delayed braking regulation, affecting the braking effect, and there is a control delay in the alternation between the power system and the braking system.

Method used

The new energy vehicle braking system based on machine learning, analyzes the pedaling frequency and vehicle speed change rate in the vehicle usage habits, adjusts the pedaling range, intervenes in the vehicle speed changes in advance, assists the braking power system to intervene accurately, and reduces the response of the braking system. time.

Benefits of technology

Effectively reduce the vehicle's speed change rate in emergencies, improve the intervention ability of the brake system, ensure the braking effect, and accurately control the vehicle speed through auxiliary braking force, enhancing the timeliness and accuracy of the brake system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118977685A_ABST
    Figure CN118977685A_ABST
Patent Text Reader

Abstract

The invention discloses a new energy vehicle braking system and method based on machine learning, and relates to the technical field of vehicle braking control, and the braking system comprises a power regulation and control module, a braking performance detection module, a braking analysis module and a braking estimation module. According to the new energy vehicle braking system and method based on machine learning, the operation times, the operation duration and the vehicle speed change rate of a vehicle at different vehicle speeds serve as basic data of vehicle use habits, and the high-frequency treading amplitude of a current vehicle brake pedal and the corresponding adjustment proportionality coefficient are obtained; and then the high-frequency treading amplitude interval is lengthened according to the adjustment proportionality coefficient, so that the change speed of the speed in the treading amplitude interval is reduced, intervention in advance is carried out, the change rate of the vehicle speed when the vehicle has an emergency situation is effectively reduced, and intervention of a braking system is facilitated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle braking control, and in particular to a new energy vehicle braking system and method based on machine learning. Background Art

[0002] At present, the new energy vehicles on the market are mainly pure electric vehicles, among which the total braking system of new energy vehicles is an important part of new energy vehicles. The total braking system of new energy vehicles distributes braking force through the vehicle controller VCU, adheres to the principle of electric braking force first and mechanical braking force second, and ensures that the battery, electric locomotive battery and driving effect reach the optimal state. Through the reasonable coordination of electric braking and mechanical braking, the driver's operating comfort is greatly improved, the damage to the battery system is reduced, and the service life of new energy vehicles is extended.

[0003] At present, a Chinese patent with the existing patent application number "CN201980064030.X" discloses an automatic braking device for a vehicle, and a vehicle that adopts a diagonal method as a braking system is equipped with an automatic braking device. The device is equipped with a yaw rate sensor, a steering angle sensor, a first pressure regulating valve for adjusting the hydraulic pressure of the right front wheel cylinder, a second pressure regulating valve for adjusting the hydraulic pressure of the left front wheel cylinder, and a controller for controlling the first pressure regulating valve and the second pressure regulating valve. The controller calculates the turning amount deviation based on the standard turning amount corresponding to the steering angle and the actual turning amount corresponding to the yaw rate. When the turning amount deviation is greater than the specified amount, the vehicle's deviation direction is determined based on the yaw rate, and the vehicle is corrected by increasing the hydraulic target value of one side and reducing the hydraulic target value of the other side. Although it serves the purpose of vehicle speed control to a certain extent, it can only be activated when the braking system is triggered, and cannot cooperate with the vehicle's power system.

[0004] However, during the implementation of the above technical solution, it was found that there were at least the following technical problems:

[0005] Inability to intervene in advance: The existing vehicle braking system mainly controls the "situations" that have already occurred, such as excessive speed, impending collision, etc., but cannot intervene in advance for the corresponding power system, resulting in the inability to reserve enough time for braking control when the vehicle encounters a "situation" (the braking system control has a delay and requires time to react). When encountering these emergencies, the driver's feet are prone to shaking, causing the vehicle speed to increase for a short period of time, which will also affect the braking effect of the subsequent braking system;

[0006] In addition, since the vehicle's power system is mainly used during driving, and most of the situations that require the assistance of the braking system occur during the operation of the power system. For example, the vehicle reaches a certain speed by stepping on the power pedal. Then, when encountering an emergency or the speed is too fast, it is necessary to step on the brake pedal immediately. The vehicle's braking system controls the vehicle instead of the power system. In this alternating process, the control-related structure of the power system needs to be turned off or disabled first, and then the braking system takes over the control of the braking mechanism on the vehicle. This itself takes a certain amount of time. For this reason, we propose a new energy vehicle braking system and method based on machine learning. Summary of the invention

[0007] 1. Technical issues to be resolved

[0008] In view of the deficiencies in the prior art, the present invention provides a new energy vehicle braking system and method based on machine learning. The high-frequency pedaling amplitude and the corresponding adjustment proportional coefficient of the current vehicle brake pedal are obtained based on the number of vehicle operation times, operating time and speed change rate at different speeds as basic data of vehicle usage habits. Then, the high-frequency pedaling amplitude interval is lengthened according to the adjustment proportional coefficient, thereby reducing the speed change rate within the pedaling amplitude interval, thereby performing early intervention, effectively reducing the speed change rate of the vehicle when an emergency occurs, and facilitating the intervention of the braking system. Similarly, the interval corresponding to the low-frequency pedaling amplitude is adjusted, and thus the high frequency and low frequency are classified according to the user's usage habits. Therefore, when the low-frequency pedaling amplitude interval is shortened, it is convenient for the vehicle to quickly enter the gentle acceleration area, thereby facilitating the subsequent intervention of the braking system, and solving the technical problems that the existing new energy vehicle braking system cannot achieve early intervention and the braking effect of the braking system when in use.

[0009] (II) Technical solution

[0010] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0011] A new energy vehicle braking system based on machine learning, the braking system comprising:

[0012] The power control module obtains the vehicle's usage habits based on the number of times the vehicle runs at different speeds, the running time, and the speed change rate, and then adjusts the vehicle's driving parameters at different pedaling frequencies based on the usage habits;

[0013] The braking performance detection module obtains the braking force, braking force duration and speed change of the vehicle under the current braking force of the brake pedal at different pedaling amplitudes during vehicle driving, generates a vehicle braking influence coefficient, and then compares it with a preset verification threshold interval, which includes an upper threshold and a lower threshold;

[0014] When the vehicle braking influence coefficient is greater than the upper threshold, no response is made;

[0015] When the vehicle braking influence coefficient is between the upper threshold and the lower threshold, a control instruction is issued to execute the control strategy until the vehicle braking influence coefficient is greater than the upper threshold and then stops;

[0016] When the vehicle's braking influence coefficient is less than the lower limit threshold, a brake replacement instruction is issued, the normal effective speed range of the braking force is calculated, and it is uploaded to the vehicle's central control screen for display;

[0017] The braking analysis module, when receiving a control command or a brake replacement command, turns on the simulation unit, increases the braking force of the vehicle according to the preset auxiliary braking force, recalculates the vehicle braking influence coefficient under the current braking force, and then generates an influence index of the auxiliary braking force on the vehicle by comparing the two calculation results;

[0018] The braking estimation module generates an estimated value of the external vehicle impact based on the external vehicle driving parameters detected by the on-board radar during the vehicle's driving process, and then combines it with the auxiliary braking force influence index to obtain an intervention capability evaluation value of the auxiliary braking force. When the intervention capability evaluation value of the auxiliary braking force enters the preset alarm range, a braking command is sent to the on-board control system, and the on-board control system performs emergency braking.

[0019] Preferably, the specific analysis process of the power control module is as follows:

[0020] Retrieving the pedaling amplitude of the power pedal and the power output parameters of the vehicle during the previous m driving processes in the database, the power output parameters including the number of operations at different vehicle speeds, the operation duration, and the vehicle speed change rate at each pedaling amplitude;

[0021] With the vehicle's driving speed as the X-axis and the number of operations at the same speed as the Y-axis, a usage frequency diagram at different speeds is established to obtain the maximum number of uses and the corresponding driving speed, which are recorded as H max 、V max ;

[0022] By analyzing the formula Calculate the standard value of the frequency of use Cxe v , where Indicates the preset adjustment ratio, and then uses the standard value of the frequency Cxev As the dividing line, the part exceeding the limit is recorded as the high-frequency speed, and vice versa, it is recorded as the low-frequency speed.

[0023] According to the high-frequency vehicle speed, the pedaling amplitude value at the corresponding vehicle speed is retrieved, and the set is recorded as the high-frequency pedaling amplitude set, and the low-frequency pedaling amplitude set is obtained similarly;

[0024] By analyzing the formula Calculate the adjustment ratio coefficient of the high-frequency pedaling amplitude set Where k represents the total number of times the brake pedal is stepped on during the first m driving times, L i represents the pedaling amplitude of the power pedal for the i-th time, and n represents the pedaling amplitude of L during the first m driving times. i The number of times it is used, α L Represents the weight coefficient of the average pedaling amplitude, V i Indicates that the pedaling amplitude is L i The vehicle speed at time V i-1 Indicates that the pedaling amplitude is L i -Le is the vehicle's speed, T Vi Indicates the driving speed is V i The running time, T 总 represents the total driving time of the first m trips, α θ The weight coefficient representing the rate of change of vehicle speed, when No adjustment is made. No adjustment is made and it is recorded as fault data;

[0025] According to the obtained adjustment scale factor Increase the high-frequency pedaling amplitude set.

[0026] Preferably, the specific analysis process of adjusting the pedaling amplitude is as follows:

[0027] Set the high-frequency pedaling amplitude and its corresponding adjustment ratio Substitute into the analysis formula Calculate the high-frequency pedaling amplitude set after amplification of the pedaling amplitude range Elio 增距 ;

[0028] Get the distance when the vehicle's power pedal is fully depressed, and record it as L 总 , by analyzing the formula Calculate the pedaling amplitude range after the low-frequency pedaling amplitude set is shortened. 缩距 And the adjustment ratio coefficient of the low-frequency pedaling amplitude set

[0029] Adjust the proportional coefficient based on the calculated low-frequency pedaling amplitude set Adjust the pedaling amplitude corresponding to the low-frequency vehicle speed.

[0030] Preferably, the specific analysis process of the braking performance detection module is as follows;

[0031] Obtain the braking force generated by the braking amplitude, the duration of the braking force, and the speed of the vehicle under the current braking force at different braking amplitudes during vehicle driving, which are recorded as N L 、T L , S L ;

[0032] By analyzing the formula Calculate the vehicle speed change Kjsm under the current braking force L , where (S L ) j represents the speed of the vehicle at time j;

[0033] N L 、T L 、Kjsm L Substitute into the analysis formula Calculate the vehicle braking influence coefficient Ysxz L , where β represents the correction factor of the preset vehicle braking influence coefficient, e represents the natural constant, f1, f2, f3 represent the weights of the preset braking force, braking force duration and vehicle speed change under the current braking force, respectively, f1+f2+f3=1;

[0034] By analyzing the formula Get the normal effective speed range of braking force Qevi v , where Qz V Indicates the preset speed adjustment ratio.

[0035] Preferably, before comparing the vehicle braking influence coefficient with the preset verification threshold interval, a verification is required first, and the specific verification process is as follows:

[0036] Obtain the vehicle braking influence coefficient of the last time the vehicle reached the same braking force in the database, and compare it with the vehicle braking influence coefficient currently obtained. When the difference between the two is within the preset error allowable range, no response is made, otherwise, an alarm is issued;

[0037] Calculate the vehicle braking influence coefficient at each moment during the current braking force duration, calculate the difference between two adjacent vehicle braking influence coefficients, and then compare the difference with the preset fluctuation threshold. When the difference is within the fluctuation threshold, no response is made, otherwise an alarm is issued.

[0038] Preferably, when a control instruction is received, a control strategy is executed, and the process of executing the control strategy is as follows:

[0039] S211, set the braking force adjustment range Qed N ;

[0040] S212, adjust the amplitude Qed according to the braking force N Add auxiliary braking force on the basis of vehicle braking force and recalculate vehicle braking influence coefficient Ysxz L ;

[0041] S213, the vehicle braking influence coefficient Ysxz obtained in the previous step L Compare with the preset check threshold interval to detect whether it is greater than the upper threshold. When the vehicle braking influence coefficient is greater than the upper threshold, stop adjusting. Otherwise, re-execute the previous step.

[0042] S214. According to the above steps S211 to S213, the auxiliary braking force at different vehicle speeds is calculated to obtain an auxiliary braking force adjustment set of the vehicle.

[0043] Preferably, the specific analysis process of the braking analysis module is as follows:

[0044] Setting auxiliary braking force Eed N ;

[0045] Based on auxiliary braking force Eed N Add auxiliary braking force based on the current braking force and recalculate the vehicle braking influence coefficient Ysxz L ;

[0046] Substitute the results of the two calculations into the analysis formula Calculate the influence index Zbno of auxiliary braking force on the vehicle L ,Ysxz S Indicates the vehicle braking influence coefficient under the current braking force, Ysxz D Indicates the vehicle braking influence coefficient after adding auxiliary braking force on the basis of current braking force, L t Indicates the distance required for the vehicle to stop under the current braking force, L′ t It indicates the distance required for the vehicle to stop after the auxiliary braking force is added to the current braking force. σ1 and σ2 respectively represent the weights of the vehicle braking influence coefficient ratio and the braking distance ratio. σ1>0, σ2>0.

[0047] Preferably, the external vehicle driving parameters include the driving speed of the external vehicle and the distance between the external vehicle and the vehicle, and the external vehicle impact estimation value Brwq is generated. L The formula is as follows:

[0048]

[0049] Where V 外 represents the speed of the external vehicle, L s Indicates the distance between the vehicle and the external vehicle, A1 and A2 respectively indicate the speed of the external vehicle and the proportional coefficient of the distance between the vehicle and the external vehicle, A1+A2=1, and 1>A1>0, 1>A2>0;

[0050] The calculated external vehicle impact estimate Brwq L Combined with the auxiliary braking force influence index, the intervention capability evaluation value Bewe of the auxiliary braking force is generated L , generate the intervention capability evaluation value Bewe of the auxiliary braking force L The formula used is as follows:

[0051]

[0052] Where ω1, ω2, and ω3 are the proportional coefficients of the auxiliary braking force influence ratio, influence coefficient deviation, and external vehicle influence estimation value, respectively.

[0053] Preferably, the alarm range includes a high-risk interval and a low-risk interval; when the intervention capability evaluation value of the auxiliary braking force Bewe L When the vehicle is in the low-risk zone, a low-risk warning is issued, and the amount of auxiliary braking force applied is adjusted according to the degree of brake pedal depression;

[0054] When the intervention capability evaluation value of the auxiliary braking force Bewe L When the vehicle is in a high-risk area, a high-risk warning is issued and a braking command is sent to the vehicle control system, which performs emergency braking.

[0055] Preferably, the new energy vehicle braking method based on machine learning as claimed in claim 1 comprises:

[0056] According to the number of times the vehicle runs at different speeds, the running time and the speed change rate, the usage habits of the vehicle are obtained, and then the vehicle driving parameters at different pedaling frequencies are adjusted according to the usage habits;

[0057] The braking force, braking force duration and speed change of the vehicle under the current braking force of the brake pedal at different pedaling amplitudes during vehicle driving are obtained to generate a vehicle braking influence coefficient, which is then compared with a preset verification threshold interval, where the verification threshold interval includes an upper threshold and a lower threshold;

[0058] When the vehicle braking influence coefficient is greater than the upper threshold, no response is made;

[0059] When the vehicle braking influence coefficient is between the upper threshold and the lower threshold, a control instruction is issued to execute the control strategy until the vehicle braking influence coefficient is greater than the upper threshold and then stops;

[0060] When the vehicle's braking influence coefficient is less than the lower limit threshold, a brake replacement instruction is issued, the normal effective speed range of the braking force is calculated, and it is uploaded to the vehicle's central control screen for display;

[0061] When receiving a control command or a brake replacement command, the simulation unit is turned on, the braking force of the vehicle is increased according to the preset auxiliary braking force, the vehicle braking influence coefficient of the vehicle under the current braking force is recalculated, and then the influence index of the auxiliary braking force on the vehicle is generated by comparing the two calculation results;

[0062] Based on the external vehicle driving parameters detected by the on-board radar during the vehicle's driving process, an estimated value of the external vehicle impact is generated, and then combined with the auxiliary braking force influence index to obtain an intervention capability evaluation value of the auxiliary braking force. When the intervention capability evaluation value of the auxiliary braking force enters the preset alarm range, a braking command is sent to the on-board control system, and the on-board control system performs emergency braking.

[0063] (III) Beneficial effects

[0064] 1. Based on the number of times the vehicle runs at different speeds, the running time and the speed change rate as the basic data of the vehicle usage habits, the high-frequency pedaling amplitude of the current vehicle brake pedal and the corresponding adjustment ratio coefficient are obtained, and then the high-frequency pedaling amplitude interval is lengthened according to the adjustment ratio coefficient, thereby reducing the speed change rate within the pedaling amplitude interval, so as to intervene in advance, effectively reduce the speed change rate of the vehicle when an emergency occurs, and facilitate the intervention of the braking system. Similarly, the interval corresponding to the low-frequency pedaling amplitude is adjusted, and thus the high frequency and low frequency are classified according to the user's usage habits, so when the low-frequency pedaling amplitude interval is shortened, it is convenient for the vehicle to quickly enter the gentle acceleration area, thereby facilitating the subsequent intervention of the braking system.

[0065] 2. First, determine the braking capacity of the current vehicle by collecting relevant braking data during the vehicle's driving process, and then add auxiliary braking force on the basis of the existing braking force to determine the impact of the auxiliary braking force on the vehicle. In this way, a specific braking force can be applied to the vehicle to assist the vehicle in braking in response to the lack of braking force of the current vehicle. Moreover, since the auxiliary braking force is adjusted according to the results obtained from actual detection, the added auxiliary braking force is more accurate.

[0066] 3. Based on the collected driving parameters of external vehicles, calculate an estimated external vehicle impact value Brwq that can intuitively evaluate the impact of external vehicles on the vehicleL , and then combined with the auxiliary braking force influence index to generate the intervention capability evaluation value Bewe of the auxiliary braking force L , so that when the user urgently steps on the brake pedal, auxiliary braking force can be applied to the vehicle, thereby assisting the vehicle to decelerate faster. In addition, due to the intervention capability evaluation value Bewe L is the estimated value Brwq through external vehicles L It is obtained by combining with the auxiliary braking force influence index. Therefore, it can adaptively adjust the auxiliary braking force of the vehicle according to the braking force required during the vehicle's driving process, so that it can control the vehicle speed more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0068] Figure 1 It is a flowchart of the overall process of an embodiment of the present invention;

[0069] Figure 2 A structural diagram of a vehicle braking system in an embodiment of the present invention;

[0070] Figure 3 1 is a usage frequency diagram at different vehicle speeds in an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The embodiment of the present application provides a new energy vehicle braking system and method based on machine learning to solve the technical problems that the existing new energy vehicle braking system cannot achieve early intervention and the braking effect of the braking system when in use. During driving, the high-frequency pedaling amplitude of the current vehicle brake pedal and the corresponding adjustment proportional coefficient are obtained based on the number of vehicle operations, operating time and speed change rate at different speeds as basic data of vehicle usage habits. Then, the high-frequency pedaling amplitude interval is lengthened according to the adjustment proportional coefficient, thereby reducing the speed change rate within the pedaling amplitude interval, so as to intervene in advance and effectively reduce the speed change rate of the vehicle when an emergency occurs, thereby facilitating the intervention of the braking system. Similarly, the interval corresponding to the low-frequency pedaling amplitude is adjusted, and thus the high frequency and low frequency are classified according to the user's usage habits. Therefore, when the low-frequency pedaling amplitude interval is shortened, it is convenient for the vehicle to quickly enter the gentle acceleration area, thereby facilitating the subsequent intervention of the braking system.

[0072] Example 1

[0073] The technical solution in the embodiment of the present application is to solve the technical problem that the existing new energy vehicle braking system cannot achieve early intervention and the braking effect of the braking system when in use. The overall idea is as follows:

[0074] In view of the problems existing in the prior art, the present invention provides a new energy vehicle braking system based on machine learning, wherein the braking system is mainly divided into a power control unit and a braking control unit, wherein the braking control unit mainly performs auxiliary intervention with an auxiliary braking unit, and in addition, an autonomously adjustable power control unit is provided. The reason for adding a power control unit to the braking system is to avoid mutual interference between the braking system and the original power control system during the braking process, and when the braking system is working, a certain reaction time is required (that is, the braking system takes over the power system to control the vehicle's braking system), and within this reaction time, if the vehicle speed changes over a distance, the originally detected data will become inapplicable, and the current data needs to be recollected, which further increases the reaction time.

[0075] For example, when a "situation" occurs during vehicle driving, the human reaction speed is between 0.2 and 0.3 seconds. When encountering a "situation", the human body will subconsciously enter a "rigid" state (muscles tense), and the driver's foot is usually bent when stepping on the pedal, such as Figure 2 As shown, before the driver reacts, the foot has already stepped on the power pedal (i.e., the accelerator), causing the vehicle to accelerate for a while, resulting in a change in the vehicle's speed in a short period of time. As for the braking system, in order to intervene in the vehicle's braking structure as soon as possible, the vehicle's data is usually collected when the vehicle starts. The braking system will not intervene until the vehicle data is abnormal, and a braking strategy is given based on the current abnormal vehicle data. It can be seen that the vehicle's braking has "information lag";

[0076] Moreover, due to the above reasons (the vehicle speed will change during the reaction phase), the previously detected data is inaccurate, and the braking strategy obtained based on the data cannot achieve the most effective braking. If the current vehicle speed that has changed is detected, the strategy is generated, and the braking is executed again, it will seriously affect the timeliness of the braking.

[0077] Therefore, in order to be able to respond in time and ensure the accuracy of the braking strategy provided, the power pedal should be adjusted before the braking system intervenes, so that the above problems will not occur. To this end, we need to coordinate the power control unit and the brake control unit.

[0078] Power control unit

[0079] According to the number of times the vehicle runs at different speeds, the running time and the speed change rate, the vehicle usage habits are obtained, and then the vehicle driving parameters at different pedaling frequencies are adjusted according to the usage habits. Among them, when the vehicle maintains a certain speed and lasts for 1-3 seconds, it is recorded as running once at the current speed, and the set time can be set in advance as needed. For example, if the vehicle runs at a speed of 120km / h for more than 3s, it is recorded that the vehicle is used once at a speed of 120km / h. This application takes 3s as an example. The specific analysis process is as follows:

[0080] S101, retrieve the power pedal's pedaling amplitude and the vehicle's power output parameters from the database during several previous driving processes. During the process of loading the brake system, freely select the required amount of driving data to be retrieved from the database according to the computing power of the vehicle control system. Generally, the vehicle system takes the previous 50 times as the reference. The power output parameters include the number of operations at different vehicle speeds, the operating time, and the vehicle speed change rate at each pedaling amplitude. The driver's usage habits can be seen based on the maximum number of times the vehicle is used and the operating time. The vehicle speed change rate at each pedaling amplitude refers to the degree of change in the vehicle speed when the power pedal is stabilized at a certain pedaling amplitude and the pedaling amplitude of the power pedal changes slightly. It is recorded as the vehicle speed change rate at the current pedaling amplitude.

[0081] S102, with the vehicle's driving speed as the X-axis and the number of operations at the same speed as the Y-axis, establish a usage frequency diagram at different speeds, obtain the maximum number of uses and the corresponding driving speed, and record them as H max 、V max ,like Figure 3 As shown;

[0082] S103, by analyzing the formula Calculate the standard value of the frequency of use Cxe v , where Indicates the preset adjustment ratio, and then uses the standard value of the frequency Cxe v As the dividing line, the excess part is recorded as the high-frequency speed, and vice versa, it is recorded as the low-frequency speed. Figure 3 As shown;

[0083] S104, according to the high-frequency vehicle speed, retrieve the pedaling amplitude values ​​at the corresponding vehicle speed, record the set of values ​​as the high-frequency pedaling amplitude set, and similarly obtain the low-frequency pedaling amplitude set;

[0084] S105, by analyzing the formula Calculate the adjustment ratio coefficient of the high-frequency pedaling amplitude set Where k represents the total number of times the brake pedal is stepped on during the first m driving times, L irepresents the pedaling amplitude of the power pedal for the i-th time, and n represents the pedaling amplitude of L during the first m driving times. i The number of times it is used, α L Represents the weight coefficient of the average pedaling amplitude, V i Indicates that the pedaling amplitude is L i The vehicle speed at time V i-1 Indicates that the pedaling amplitude is L i -The vehicle's speed at Le, T Vi Indicates the driving speed is V i The running time, T 总 represents the total driving time of the first m trips, α θ The weight coefficient representing the rate of change of vehicle speed, It indicates the efficiency of vehicle speed change when the pedal depression distance is Le.

[0085] Adjust the scale factor When No adjustment is made. No adjustment is made and it is recorded as fault data;

[0086] S106, adjusting the proportional coefficient according to the obtained Increase the high-frequency pedaling amplitude set. The adjustment of pedaling amplitude is divided into the following two parts:

[0087] High frequency pedaling amplitude collection:

[0088] Set the high-frequency pedaling amplitude and its corresponding adjustment ratio Substitute into the analysis formula Calculate the high-frequency pedaling amplitude set after amplification of the pedaling amplitude range Elio 增距 .

[0089] In this way, the length corresponding to the high-frequency pedaling amplitude increases. For example, before adjustment, the speed increases by 20km / h on the original basis for every 1cm increase in the pedaling amplitude; after the amplification, the speed increases by 15km / h on the original basis for every 1cm increase in the pedaling amplitude. Therefore, adjustments can be made based on the original pedaling amplitude to avoid excessive changes in speed when a "situation" occurs, providing sufficient reaction time for the braking system to take effect. At the same time, the data obtained when the braking system intervenes is stable, making the corresponding braking strategy of the braking system more accurate.

[0090] Low frequency pedaling amplitude collection:

[0091] Get the distance when the vehicle's power pedal is fully depressed, and record it as L 总 , by analyzing the formula Calculate the pedaling amplitude range after the low-frequency pedaling amplitude set is shortened. 缩距And the adjustment ratio coefficient of the low-frequency pedaling amplitude set

[0092] According to the result of the high-frequency pedaling amplitude adjustment, the low-frequency pedaling amplitude set is adjusted, so that the time to quickly enter the controllable range can be shortened. At the same time, this method can be used to compress the unused speed range. For example, the upper limit of the vehicle speed is generally 200km / h, and the speed range that is actually used is 0-150km / h. The unusable 150-200km / h occupies 1 / 4 of the pedal available range. In this way, 0-150km / h can be increased from the original 3 / 4 to 9 / 10, etc., which can slow down the efficiency of vehicle speed changes.

[0093] Adjust the proportional coefficient based on the calculated low-frequency pedaling amplitude set Adjust the pedaling amplitude corresponding to the low-frequency vehicle speed.

[0094] Brake adjustment unit

[0095] (1) Braking performance detection: The braking force, braking force duration, and speed change of the vehicle under the current braking force at different braking amplitudes during vehicle driving are obtained to generate the vehicle braking influence coefficient. The specific analysis process of the braking performance detection module is as follows;

[0096] S201, obtaining the braking force generated by the braking amplitude at different braking amplitudes during vehicle driving, the duration of the braking force, and the speed of the vehicle under the current braking force, respectively denoted as N L 、T L , S L ;

[0097] S202, by analyzing the formula Calculate the vehicle speed change Kjsm under the current braking force L , where (S L ) j represents the speed of the vehicle at time j, Indicates the vehicle is braking force N L The average vehicle speed under the action of the braking time is then compared with the braking time to obtain the speed change Kjsm during the entire braking process. L ;

[0098] S203, N L 、T L 、Kjsm L Substitute into the analysis formula Calculate the vehicle braking influence coefficient Ysxz L, where β represents the correction factor of the preset vehicle braking influence coefficient, e represents the natural constant, f1, f2, f3 represent the weights of the preset braking force, braking force duration and vehicle speed change under the current braking force, respectively, f1+f2+f3=1;

[0099] S204, by analyzing the formula Get the normal effective speed range of braking force Qevi v , where Qz V Indicates the preset speed adjustment ratio. In order to improve the accuracy of the detection, the obtained data needs to be verified. Common verifications are divided into the following two types:

[0100] Vehicle braking influence coefficient verification: obtain the vehicle braking influence coefficient of the last vehicle under the same braking force in the database, and compare it with the current vehicle braking influence coefficient. When the difference between the two is within the preset error allowable range, no response is made, otherwise, an alarm is issued;

[0101] Fluctuation range check: Calculate the vehicle braking influence coefficient at each moment during the current braking force duration, calculate the difference between two adjacent vehicle braking influence coefficients, and then compare the difference with the preset fluctuation threshold. When the difference is within the fluctuation threshold, no response is made, otherwise an alarm is issued.

[0102] After the verification is completed, the vehicle braking influence coefficient is compared with a preset verification threshold interval, which includes an upper threshold and a lower threshold;

[0103] When the vehicle braking influence coefficient is greater than the upper threshold, no response is made;

[0104] When the vehicle braking influence coefficient is between the upper threshold and the lower threshold, a control instruction is issued to execute the control strategy until the vehicle braking influence coefficient is greater than the upper threshold and then stops;

[0105] When the vehicle's braking influence coefficient is less than the lower limit threshold, a brake replacement command is issued, the speed range in which the braking force is normally effective is calculated, and it is uploaded to the vehicle's central control screen for display.

[0106] When a control instruction is received, the control strategy is executed. The process of executing the control strategy is as follows:

[0107] S211, set the braking force adjustment range Qed N ;

[0108] S212, adjust the amplitude Qed according to the braking force N Add auxiliary braking force on the basis of vehicle braking force and recalculate vehicle braking influence coefficient Ysxz L ;

[0109] S213, the vehicle braking influence coefficient Ysxz obtained in the previous step L Compare with the preset check threshold interval to detect whether it is greater than the upper threshold. When the vehicle braking influence coefficient is greater than the upper threshold, stop adjusting. Otherwise, re-execute the previous step.

[0110] S214. According to the above steps S211 to S213, the auxiliary braking force at different vehicle speeds is calculated to obtain an auxiliary braking force adjustment set of the vehicle.

[0111] (2) Braking capacity analysis: When receiving a control command or brake replacement command, the simulation unit is turned on, the vehicle's braking force is increased according to the preset auxiliary braking force, and the vehicle's braking influence coefficient under the current braking force is recalculated. Then, the two calculation results are compared to generate the auxiliary braking force's influence index on the vehicle. The specific process is as follows:

[0112] S221, setting auxiliary braking force Eed N ;

[0113] S222, based on auxiliary braking force Eed N Add auxiliary braking force based on the current braking force and recalculate the vehicle braking influence coefficient Ysxz L ;

[0114] S223. Substitute the results of the two calculations into the analysis formula Calculate the influence index Zbno of auxiliary braking force on the vehicle L ,Ysxz S Indicates the vehicle braking influence coefficient under the current braking force, Ysxz D Indicates the vehicle braking influence coefficient after adding auxiliary braking force on the basis of current braking force, L t Indicates the distance required for the vehicle to stop under the current braking force, L′ t It indicates the distance required for the vehicle to stop after the auxiliary braking force is added to the current braking force. σ1 and σ2 respectively represent the weights of the vehicle braking influence coefficient ratio and the braking distance ratio. σ1>0, σ2>0.

[0115] (3) Braking prediction module: Generates an estimated value of the impact of external vehicles based on the driving parameters of external vehicles detected by the on-board radar during the vehicle's driving process. The driving parameters of external vehicles include the driving speed of the external vehicle and the distance between the vehicle and the vehicle, and generates an estimated value of the impact of external vehicles Brwq L The formula is as follows:

[0116]

[0117] Where V 外 represents the speed of the external vehicle, L s Indicates the distance between the vehicle and the external vehicle, A1 and A2 respectively indicate the speed of the external vehicle and the proportional coefficient of the distance between the vehicle and the external vehicle, A1+A2=1, and 1>A1>0, 1>A2>0;

[0118] The calculated external vehicle impact estimate Brwq L Combined with the auxiliary braking force influence index, the intervention capability evaluation value Bewe of the auxiliary braking force is generated L , generate the intervention capability evaluation value Bewe of the auxiliary braking force L The formula used is as follows:

[0119]

[0120] Where ω1, ω2, and ω3 are the proportional coefficients of the auxiliary braking force influence ratio, influence coefficient deviation, and external vehicle influence estimation value, respectively.

[0121] The external vehicle impact estimate Brwq will be generated L Combined with the auxiliary braking force influence index, the intervention capability evaluation value of the auxiliary braking force is obtained. When the intervention capability evaluation value of the auxiliary braking force enters the preset alarm range, a braking command is sent to the vehicle control system, and the vehicle control system performs emergency braking.

[0122] The warning range includes a high-risk interval and a low-risk interval; when the intervention capability evaluation value of the auxiliary braking force is Bewe L When the vehicle is in the low-risk zone, a low-risk warning is issued, and the amount of auxiliary braking force applied is adjusted according to the degree of brake pedal depression;

[0123] When the intervention capability evaluation value of the auxiliary braking force Bewe L When the vehicle is in a high-risk zone, a high-risk warning is issued and a braking command is sent to the vehicle control system, which then performs emergency braking.

[0124] Example 2

[0125] Based on Example 1, in order to ensure that the device remains stable when folded or unfolded, the overall idea of ​​this application embodiment is as follows:

[0126] A new energy vehicle braking method based on machine learning, the braking method comprising:

[0127] According to the number of times the vehicle runs at different speeds, the running time and the speed change rate, the usage habits of the vehicle are obtained, and then the vehicle driving parameters at different pedaling frequencies are adjusted according to the usage habits;

[0128] The braking force, braking force duration and speed change of the vehicle under the current braking force of the brake pedal at different pedaling amplitudes during vehicle driving are obtained to generate a vehicle braking influence coefficient, which is then compared with a preset verification threshold interval, where the verification threshold interval includes an upper threshold and a lower threshold;

[0129] When the vehicle braking influence coefficient is greater than the upper threshold, no response is made;

[0130] When the vehicle braking influence coefficient is between the upper threshold and the lower threshold, a control instruction is issued to execute the control strategy until the vehicle braking influence coefficient is greater than the upper threshold and then stops;

[0131] When the vehicle's braking influence coefficient is less than the lower limit threshold, a brake replacement instruction is issued, the normal effective speed range of the braking force is calculated, and it is uploaded to the vehicle's central control screen for display;

[0132] When receiving a control command or a brake replacement command, the simulation unit is turned on, the braking force of the vehicle is increased according to the preset auxiliary braking force, the vehicle braking influence coefficient of the vehicle under the current braking force is recalculated, and then the influence index of the auxiliary braking force on the vehicle is generated by comparing the two calculation results;

[0133] Based on the external vehicle driving parameters detected by the on-board radar during the vehicle's driving process, an estimated value of the external vehicle impact is generated, and then combined with the auxiliary braking force influence index to obtain an intervention capability evaluation value of the auxiliary braking force. When the intervention capability evaluation value of the auxiliary braking force enters the preset alarm range, a braking command is sent to the on-board control system, and the on-board control system performs emergency braking.

[0134] Finally, it should be noted that: Obviously, the above embodiments are only examples for clearly explaining the present invention, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from this are still within the scope of protection of the present invention.

Claims

1. A new energy vehicle braking system based on machine learning, characterized in that: The braking system includes: The power control module obtains the vehicle's usage habits based on the number of times the vehicle runs at different speeds, the running time, and the speed change rate, and then adjusts the vehicle's driving parameters at different pedaling frequencies based on the usage habits; The braking performance detection module obtains the braking force, braking force duration and speed change of the vehicle under the current braking force of the brake pedal at different pedaling amplitudes during vehicle driving, generates a vehicle braking influence coefficient, and then compares it with a preset verification threshold interval, which includes an upper threshold and a lower threshold; When the vehicle braking influence coefficient is greater than the upper threshold, no response is made; When the vehicle braking influence coefficient is between the upper threshold and the lower threshold, a control instruction is issued to execute the control strategy until the vehicle braking influence coefficient is greater than the upper threshold and then stops; When the vehicle's braking influence coefficient is less than the lower limit threshold, a brake replacement instruction is issued, the normal effective speed range of the braking force is calculated, and it is uploaded to the vehicle's central control screen for display; The braking analysis module, when receiving a control command or a brake replacement command, turns on the simulation unit, increases the braking force of the vehicle according to the preset auxiliary braking force, recalculates the vehicle braking influence coefficient under the current braking force, and then generates an influence index of the auxiliary braking force on the vehicle by comparing the two calculation results; The braking estimation module generates an estimated value of the external vehicle impact based on the external vehicle driving parameters detected by the on-board radar during the vehicle's driving process, and then combines it with the auxiliary braking force influence index to obtain an intervention capability evaluation value of the auxiliary braking force. When the intervention capability evaluation value of the auxiliary braking force enters the preset alarm range, a braking command is sent to the on-board control system, and the on-board control system performs emergency braking.

2. A new energy vehicle braking system based on machine learning as claimed in claim 1, characterized in that: The specific analysis process of the power control module is as follows: Retrieving the pedaling amplitude of the power pedal and the power output parameters of the vehicle during the previous m driving processes in the database, the power output parameters including the number of operations at different vehicle speeds, the operation duration, and the vehicle speed change rate at each pedaling amplitude; With the vehicle's driving speed as the X-axis and the number of operations at the same speed as the Y-axis, a usage frequency diagram at different speeds is established to obtain the maximum number of uses and the corresponding driving speed, which are recorded as H max 、V max ; By analyzing the formula Calculate the standard value of the frequency of use Cxe v , where Indicates the preset adjustment ratio, and then uses the standard value of the frequency Cxe v As the dividing line, the part exceeding the limit is recorded as the high-frequency speed, and vice versa, it is recorded as the low-frequency speed. According to the high-frequency vehicle speed, the pedaling amplitude value at the corresponding vehicle speed is retrieved, and the set is recorded as the high-frequency pedaling amplitude set, and the low-frequency pedaling amplitude set is obtained similarly; By analyzing the formula Calculate the adjustment ratio coefficient of the high-frequency pedaling amplitude set Where k represents the total number of times the brake pedal is stepped on during the first m driving times, L i represents the pedaling amplitude of the power pedal for the i-th time, and n represents the pedaling amplitude of L during the first m driving times. i The number of times it is used, α L Represents the weight coefficient of the average pedaling amplitude, V i Indicates that the pedaling amplitude is L i The vehicle speed at time V i-1 Indicates that the pedaling amplitude is L i -The vehicle's speed at the time of Le, Indicates the driving speed is V i The running time, T 总 represents the total driving time of the first m trips, α θ The weight coefficient representing the rate of change of vehicle speed, when No adjustment is made. No adjustment is made and it is recorded as fault data; According to the obtained adjustment scale factor Increase the high-frequency pedaling amplitude set.

3. A new energy vehicle braking system based on machine learning as claimed in claim 2, characterized in that: The specific analysis process of adjusting the pedaling amplitude is as follows: Set the high-frequency pedaling amplitude and its corresponding adjustment ratio Substitute into the analysis formula Calculate the high-frequency pedaling amplitude set after amplification of the pedaling amplitude range Elio 增距 ; Get the distance when the vehicle's power pedal is fully depressed, and record it as L 总 , by analyzing the formula Calculate the pedaling amplitude range after the low-frequency pedaling amplitude set is shortened. 缩距 And the adjustment ratio coefficient of the low-frequency pedaling amplitude set Adjust the proportional coefficient based on the calculated low-frequency pedaling amplitude set Adjust the pedaling amplitude corresponding to the low-frequency vehicle speed.

4. A new energy vehicle braking system based on machine learning as claimed in claim 1, characterized in that: The specific analysis process of the braking performance detection module is as follows; Obtain the braking force generated by the braking amplitude, the duration of the braking force, and the speed of the vehicle under the current braking force at different braking amplitudes during vehicle driving, which are recorded as N L , T L , S L ; By analyzing the formula Calculate the vehicle speed change Kjsm under the current braking force L , where (S L ) j represents the speed of the vehicle at time j; N L , T L 、Kjsm L Substitute into the analysis formula Calculate the vehicle braking influence coefficient Ysxz L , where β represents the correction factor of the preset vehicle braking influence coefficient, e represents the natural constant, f1, f2, f3 represent the weights of the preset braking force, braking force duration and vehicle speed change under the current braking force, respectively, f1+f2+f3=1; By analyzing the formula Get the normal effective speed range of braking force Qevi v , where Qz v Indicates the preset speed adjustment ratio.

5. A new energy vehicle braking system based on machine learning as claimed in claim 4, characterized in that: Before comparing the vehicle braking influence coefficient with the preset verification threshold range, a verification is required. The specific verification process is as follows: Obtain the vehicle braking influence coefficient of the last time the vehicle reached the same braking force in the database, and compare it with the vehicle braking influence coefficient currently obtained. When the difference between the two is within the preset error allowable range, no response is made, otherwise, an alarm is issued; Calculate the vehicle braking influence coefficient at each moment during the current braking force duration, calculate the difference between two adjacent vehicle braking influence coefficients, and then compare the difference with the preset fluctuation threshold. When the difference is within the fluctuation threshold, no response is made, otherwise an alarm is issued.

6. A new energy vehicle braking system based on machine learning as claimed in claim 1, characterized in that: When a control instruction is received, the control strategy is executed. The process of executing the control strategy is as follows: S211, set the braking force adjustment range Qed N ; S212, adjust the amplitude Qed according to the braking force N Add auxiliary braking force on the basis of vehicle braking force and recalculate vehicle braking influence coefficient Ysxz L ; S213, the vehicle braking influence coefficient Ysxz obtained in the previous step L Compare with the preset check threshold interval to detect whether it is greater than the upper threshold. When the vehicle braking influence coefficient is greater than the upper threshold, stop adjusting. Otherwise, re-execute the previous step. S214. According to the above steps S211 to S213, the auxiliary braking force at different vehicle speeds is calculated to obtain an auxiliary braking force adjustment set of the vehicle.

7. A new energy vehicle braking system based on machine learning as claimed in claim 1, characterized in that: The specific analysis process of the brake analysis module is as follows: Setting auxiliary braking force Eed N ; Based on auxiliary braking force Eed N Add auxiliary braking force based on the current braking force and recalculate the vehicle braking influence coefficient Ysxz L ; Substitute the results of the two calculations into the analysis formula Calculate the influence index Zbno of auxiliary braking force on the vehicle L ,Ysxz S Indicates the vehicle braking influence coefficient under the current braking force, Ysxz D Indicates the vehicle braking influence coefficient after adding auxiliary braking force on the basis of current braking force, L t Indicates the distance required for the vehicle to stop under the current braking force, L′ t It indicates the distance required for the vehicle to stop after the auxiliary braking force is added to the current braking force. σ1 and σ2 respectively represent the weights of the vehicle braking influence coefficient ratio and the braking distance ratio. σ1>0, σ2>0.

8. A new energy vehicle braking system based on machine learning as claimed in claim 7, characterized in that: The external vehicle driving parameters include the driving speed of the external vehicle and the distance between the vehicle and the host vehicle, and generate the estimated value of the external vehicle impact Brwq L The formula is as follows: Where V 外 represents the speed of the external vehicle, L s Indicates the distance between the vehicle and the external vehicle, A1 and A2 respectively represent the speed of the external vehicle and the proportional coefficient of the distance between the vehicle and the external vehicle, A1+A2=1, and 1>A1>0, 1>A2>0; The calculated external vehicle impact estimate Brwq L Combined with the auxiliary braking force influence index, the intervention capability evaluation value Bewe of the auxiliary braking force is generated L , generate the intervention capability evaluation value Bewe of the auxiliary braking force L The formula is as follows: Where ω1, ω2, and ω3 are the proportional coefficients of the auxiliary braking force influence ratio, influence coefficient deviation, and external vehicle influence estimation value, respectively.

9. The new energy vehicle braking system based on machine learning as claimed in claim 1, characterized in that: The alert range includes high-risk interval and low-risk interval; When the intervention capability evaluation value of the auxiliary braking force Bewe L When the vehicle is in the low-risk zone, a low-risk warning is issued, and the amount of auxiliary braking force applied is adjusted according to the degree of brake pedal depression; When the intervention capability evaluation value of the auxiliary braking force Bewe L When the vehicle is in a high-risk area, a high-risk warning is issued and a braking command is sent to the vehicle control system, which performs emergency braking.

10. A new energy vehicle braking method based on machine learning as claimed in claim 1, using any one of the systems in claims 1 to 9, characterized in that: The braking method includes: According to the number of times the vehicle runs at different speeds, the running time and the speed change rate, the usage habits of the vehicle are obtained, and then the vehicle driving parameters at different pedaling frequencies are adjusted according to the usage habits; The braking force, braking force duration and speed change of the vehicle under the current braking force of the brake pedal at different pedaling amplitudes during vehicle driving are obtained to generate a vehicle braking influence coefficient, which is then compared with a preset verification threshold interval, where the verification threshold interval includes an upper threshold and a lower threshold; When the vehicle braking influence coefficient is greater than the upper threshold, no response is made; When the vehicle braking influence coefficient is between the upper threshold and the lower threshold, a control instruction is issued to execute the control strategy until the vehicle braking influence coefficient is greater than the upper threshold and then stops; When the vehicle's braking influence coefficient is less than the lower limit threshold, a brake replacement instruction is issued, the normal effective speed range of the braking force is calculated, and it is uploaded to the vehicle's central control screen for display; When receiving a control command or a brake replacement command, the simulation unit is turned on, the braking force of the vehicle is increased according to the preset auxiliary braking force, the vehicle braking influence coefficient of the vehicle under the current braking force is recalculated, and the influence index of the auxiliary braking force on the vehicle is generated by comparing the two calculation results; Based on the external vehicle driving parameters detected by the on-board radar during the vehicle's driving process, an estimated value of the external vehicle impact is generated, and then combined with the auxiliary braking force influence index to obtain an intervention capability evaluation value of the auxiliary braking force. When the intervention capability evaluation value of the auxiliary braking force enters the preset alarm range, a braking command is sent to the on-board control system, and the on-board control system performs emergency braking.