Vehicle transverse control exit algorithm strategy
By introducing a slow retreat phase when the horizontal control of the vehicle assisted driving system exits, and adjusting the output control amount based on real-time driving status information, the problem of low safety when the horizontal control exits in the prior art is solved, and higher system safety is achieved and adaptability.
Patent Information
- Application Number
- CN202411236363.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-12
AI Technical Summary
When the existing vehicle assisted driving system exits, it is difficult to judge whether it can exit safely based on real-time driving conditions when the horizontal control exits, resulting in low safety of horizontal control.
When the horizontal control system exits, the slow-retreat phase is introduced, the descent coefficient is generated based on the real-time driving status information, the output control amount is adjusted, and the hysteresis phase is judged through the slow-retreat time and safety threshold, ensuring that the system considers various working conditions when exiting.
It effectively improves the safety and adaptability of the horizontal control system, ensures that various driving conditions can be taken into account when exiting, reduces R&D costs and is adapted to different models.
Smart Images

Figure CN118928442A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of intelligent driving technology and relates to intelligent driving technology, specifically a vehicle lateral control exit algorithm strategy. Background Art
[0002] With the rapid development of automobile technology, the level of automobile intelligence is getting higher and higher, and more and more cars have assisted driving functions. Among the assisted driving functions of automobiles, the lateral control of the vehicle mainly assists the vehicle in changing lanes, keeps the vehicle in the middle of the lane, and when the driver unconsciously deviates from the lane, the vehicle will alarm or correct the deviation to return to the center of the lane. The lateral control of the vehicle can improve the lateral stability of the vehicle, reduce the operating burden of the driver, and at the same time avoid traffic accidents caused by the driver's unconscious deviation from the lane, thereby improving the safety performance of the vehicle.
[0003] In the lateral control algorithm, the actuator of the lateral control algorithm is the electric power steering system EPS (Electric Power Steering). Because the lateral control has a great influence on the stability control of the whole vehicle, the auxiliary driving controller needs to have a relatively complex handshake logic with the EPS during the lateral control process. The vehicle can only be controlled laterally after the handshake is successful. The torque and angle limit requirements must be followed during the control process. When exiting, the exit strategy of the lateral control must also be followed to ensure that EPS has no faults after the lateral control is exited. EPS can smoothly execute the driver's operation and ensure that the driver can take over the vehicle normally and drive safely. For vehicle safety considerations, when the EPS controller receives the lateral stage control signal, it needs to detect the requested torque and angle. For signals that exceed the specified limit, when the value exceeds a certain value and the time exceeds a certain value, the EPS system needs to issue a corresponding error state according to the strategy, and give corresponding feedback according to the severity. For some more serious errors, the EPS system needs to exit the lateral control and send the cause of the error state to the auxiliary driving controller through the bus. The fault state of EPS controller exit is basically divided into temporary fault and permanent fault. Temporary fault allows the assisted driving controller to reactivate the lateral control function. A permanent fault means that the assisted driving controller is not allowed to perform lateral control during the current power-on cycle of EPS. The vehicle needs to be stopped and the power is cycled before the assisted driving controller is allowed to request lateral control.
[0004] However, when most existing assisted driving systems determine that the conditions for exiting lateral control are met, they will directly exit lateral control and set the requested torque or steering angle to 0. The driver will directly take over lateral control, and EPS will switch to only responding to the driver's lateral control request. During this process, if the vehicle is traveling at a higher speed or deviates greatly from the lane line, it will exit lateral control when there is a greater risk of collision with the vehicle in the adjacent lane. It is difficult for the driver to make a timely and correct response, and when exiting lateral control, it is difficult to determine whether to exit directly based on the real-time driving conditions. The safety of lateral control is relatively low. Therefore, a vehicle lateral control exit algorithm strategy is needed. Summary of the invention
[0005] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a vehicle lateral control exit algorithm strategy, which is used to solve the problem that the lateral control system in the assisted driving system of the existing vehicle is difficult to judge whether it can be directly exited according to the real-time driving conditions; and the technical problem that the lateral control is less safe when exiting. The present application solves the above problem by adding a slow exit phase when the lateral control system exits operation, effectively ensuring that the assisted driving controller can also consider various working conditions during the exit phase.
[0006] To achieve the above objectives, the first aspect of the present application provides a vehicle lateral control exit algorithm strategy, comprising the following steps:
[0007] Step 1: obtaining a lateral control request; controlling the lateral control system to enter an operation phase according to the lateral control request;
[0008] Step 2: acquiring driving state information and driver control information in real time; generating a lateral stage control signal according to the driving state information and the driver control information; when the lateral stage control signal is a lateral control exit signal, controlling the lateral control system to enter a slow retreat stage and proceeding to step 3; otherwise, the lateral control system continues to remain in the operating stage and proceeds to step 2;
[0009] Step 3: Generate a drop coefficient based on real-time driving status information;
[0010] Step 4: Generate the final output control amount according to the drop coefficient;
[0011] Step 5: Get the slow-down time and determine whether the slow-down time is greater than the maximum time threshold. If yes, proceed to step 6; if no, obtain the final output control amount. When the absolute value of the final output control amount is less than the safety threshold, proceed to step 6; otherwise, proceed to step 3;
[0012] Step 6: The lateral control system enters the hysteresis stage and obtains the hysteresis time for entering the hysteresis stage. When the hysteresis time is greater than the set time threshold, the lateral control system enters the activation permission stage and jumps to step 1.
[0013] The present application introduces a slow retreat stage when exiting the lateral control stage. During the slow retreat stage, a descent coefficient is generated according to real-time driving status information; a final output control amount is generated according to the descent coefficient; corresponding parameters are adjusted according to the final output control amount; a slow retreat time is obtained to determine whether the slow retreat time is greater than a maximum time threshold; if yes, the lateral control system enters a hysteresis stage; if no, the final output control amount is obtained, and when the absolute value of the final output control amount is less than a safety threshold; the lateral control system enters a hysteresis stage; otherwise, a descent coefficient is continued to be generated according to real-time driving status information; a final output control amount is generated according to the descent coefficient; corresponding parameters are adjusted according to the final output control amount; it is determined whether the lateral control system can exit the slow retreat stage and enter the hysteresis stage; by introducing the slow retreat stage after the lateral control system meets the exit conditions, various operating conditions at the time of exit can be taken into account, and corresponding parameters can be adjusted according to the driving status information at the time of exit and the driver's control information, which can not only ensure the normal exit of the system, but also consider various operating conditions at the exit stage, and effectively improve the system safety.
[0014] Preferably, the generating of the lateral stage control signal according to the driving state information and the driver control information includes:
[0015] Extract the turn signal status, double flash status, brake pedal status and throttle status from the driver's control information; when any one of the turn signal status, double flash status, brake pedal status and throttle status is activated; generate an active exit signal; otherwise, extract the steering wheel torque from the driver's control information; determine whether the steering wheel torque is greater than the exit torque threshold,
[0016] If yes, an active exit signal is generated;
[0017] If not, then extract the lane line detection image in the driving status information; input the lane line detection image into the lane line recognition model to obtain the lane line label; when the lane line label is qualified, generate a lateral control maintenance signal; otherwise, generate a passive exit signal;
[0018] Among them, the lane line recognition model is obtained through artificial intelligence model training; the lateral stage control signal is a lateral control exit signal or a lateral control maintenance signal; the lateral control exit signal is an active exit signal or a passive exit signal.
[0019] Preferably, the lane line recognition model is obtained by training an artificial intelligence model, including:
[0020] Acquire a number of lane line detection images and their corresponding lane line labels; integrate the number of lane line detection images and lane line labels into a number of sets of training data and test data;
[0021] The artificial intelligence model is trained using training data, and the artificial intelligence model is tested using test data; finally, a lane line recognition model is obtained, whose input is a lane line detection image and whose output is a corresponding lane line label; wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.
[0022] Preferably, generating the drop coefficient according to the real-time driving status information includes:
[0023] Extracting the driving speed XV, the lateral vehicle distance HJ and the turning angle value QZ from the real-time driving status information;
[0024] The drop coefficient influencing factor g1 is calculated by the formula g1=exp(-k1×XV / ZV), where ZV is the maximum speed of the corresponding vehicle; k1 is the proportional coefficient; and k1>0;
[0025] The second factor of the reduction coefficient g2 is calculated by the formula g2=k2×ln(HJ / AJ+1), where AJ is the safe lateral distance; k2 is the proportionality coefficient; and 1>k2>0;
[0026] The third factor of the decline coefficient g3 is calculated by the formula g3=exp(-k3×QZ), where k3 is the proportional coefficient; and k3>0;
[0027] When the lateral control exit signal is an active exit signal, the descending coefficient influence factor g4=dk is set; otherwise, the descending coefficient influence factor g4=dm is set; where dk is the fast adjustment coefficient, dm is the slow adjustment speed; and 0 <dm<dk<1;
[0028] The reduction coefficient G is calculated by the formula G=g1×g2×g3×g4.
[0029] Preferably, generating the final output control amount according to the drop coefficient includes:
[0030] Obtain the requested turning angle value Tlast before the slow retreat phase, the maximum turning angle value change rate ΔTrate, the time t to enter the slow retreat phase, and the drop coefficient G;
[0031] The final output control amount Ttorque is calculated by the formula Ttorque=Tlast-ΔTrate×G×t.
[0032] Preferably, the safety threshold is calculated by the formula Taq=ΔTrate×Δt; wherein Δt is the operating cycle of the auxiliary driving system.
[0033] Compared with the prior art, the beneficial effects of this application are:
[0034] 1. The present application introduces a slow retreat stage when exiting the lateral control stage. In the slow retreat stage, a descent coefficient is generated according to the real-time driving status information; a final output control amount is generated according to the descent coefficient; corresponding parameters are adjusted according to the final output control amount; a slow retreat time is obtained to determine whether the slow retreat time is greater than a maximum time threshold; if yes, the lateral control system enters a hysteresis stage; if not, the final output control amount is obtained, and when the absolute value of the final output control amount is less than a safety threshold, the lateral control system enters a hysteresis stage; otherwise, a descent coefficient is continued to be generated according to the real-time driving status information; a final output control amount is generated according to the descent coefficient; corresponding parameters are adjusted according to the final output control amount; it is determined whether the lateral control system can exit the slow retreat stage and enter the hysteresis stage; by introducing the slow retreat stage after the lateral control system meets the exit conditions, various working conditions at the time of exit can be taken into account, and corresponding parameters can be adjusted according to the driving status information at the time of exit and the driver's control information, which can not only ensure the normal exit of the system, but also consider various working conditions at the exit stage, and effectively improve the system safety.
[0035] 2. In this embodiment, a descent coefficient is used to control the output in the slow retreat phase, which can effectively ensure that the assisted driving controller can also consider various working conditions in the exit phase, and fewer calibration parameters are required, thereby reducing R&D costs; at the same time, it can be quickly adapted to different vehicle models and has high generalization performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0037] Figure 1 A schematic diagram of the method steps of this application;
[0038] Figure 2 This is a flowchart of the present application. DETAILED DESCRIPTION
[0039] The technical solution of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.
[0040] See also Figure 1-Figure 2 The first aspect of the present application provides a vehicle lateral control exit algorithm strategy, including the following steps:
[0041] Step 1: Obtain a lateral control request; the lateral control request is a lateral control start request input by the driver through the control panel; after receiving the lateral control request, the lateral control system is controlled to enter the operation stage according to the lateral control request; the lateral control system has an activation permission stage, an operation stage, a slow retreat stage and a hysteresis stage, and the activation permission stage can be understood as a stop operation stage; the slow retreat stage is a buffer stage when the lateral control system exits; the hysteresis stage is a buffer stage after the lateral control system exits;
[0042] Step 2: Acquire driving status information and driver control information in real time; the driving status information includes driving speed, turning angle value, lateral vehicle distance, lane line detection image, etc.; the lateral vehicle distance is obtained by ranging radar, and the detection image is obtained by image acquisition equipment, such as high-definition camera; the driver control information includes turn signal status, double flash status, brake pedal status, throttle status and steering wheel torque; the turn signal status, double flash status, brake pedal status and throttle status all include two states of start or close; when the turn signal is on, its corresponding state is start, otherwise its corresponding state is close; when the double flash is on, its corresponding state is start, otherwise its corresponding state is close; when the brake pedal is pressed, its corresponding state is start, otherwise its corresponding state is close; when the throttle is pressed, its corresponding state is start, otherwise its corresponding state is close; generate a lateral stage control signal according to the driving status information and the driver control information; when the lateral stage control signal is a lateral control exit signal, the lateral control system enters the slow retreat stage and enters step three: otherwise, the lateral control system continues to remain in the running stage and enters step two;
[0043] In step 2, it is determined whether the conditions for exiting lateral control are met through the driving status information and the driver's control information. If so, a lateral control exit signal is generated to make the lateral control system enter the slow retreat stage; if not, a lateral control maintenance signal is generated to control the lateral control system to continue to remain in the operating stage, and continue to obtain the driving status information and the driver's control information to determine whether the driving status information and the driver's control information meet the conditions for exiting lateral control.
[0044] Step 3: Generate a descent coefficient according to the real-time driving state information; the descent coefficient is a coefficient used to control the output control amount in the slow retreat stage, which can make the output control amount in the slow retreat stage change within a safe range;
[0045] Step 4: Generate the final output control amount according to the drop coefficient; specifically including:
[0046] Obtain the requested steering angle value Tlast before the slow retreat phase, the maximum steering angle value change rate ΔTrate, the time t for entering the slow retreat phase, and the drop coefficient G; the requested steering angle value is the amplitude of the steering angle requested by the auxiliary driving controller at the moment before the lateral control system enters the slow retreat phase; it can be the amplitude of the steering angle requested by the lateral control system or the driver; the maximum steering angle change rate is the maximum change rate of the maximum steering angle value allowed by the EPS system;
[0047] The final output control amount Ttorque is calculated by the formula Ttorque=Tlast-ΔTrate×G×t; the final output control amount is used to control the steering angle value of the car; it can be understood that the above method is also applicable to converting all steering angle values into torque values.
[0048] Step 5: Obtain the slow retreat time, which is the time to enter the slow retreat stage; determine whether the slow retreat time is greater than the maximum time threshold; if yes, proceed to step 6; if no, obtain the final output control amount, and when the absolute value of the final output control amount is less than the safety threshold, proceed to step 6; otherwise, proceed to step 3;
[0049] In step five, by setting the maximum time threshold, the lateral control system will not be trapped in the slow retreat stage for a long time, ensuring the normal operation of the system; at the same time, when the final output control amount has been reduced to the required value, the lateral control system is exited to ensure the safety of the lateral control system when it can be exited; the safety threshold is calculated by the formula Taq=ΔTrate×Δt; where Δt is the operating cycle of the assisted driving system; when the assisted driving controller has reduced the requested torque or steering angle to the required value, exiting at this time will not cause the EPS system to report a fault.
[0050] Step 6: The lateral control system enters the hysteresis stage and obtains the hysteresis time for entering the hysteresis stage. When the hysteresis time is greater than the set time threshold, the lateral control system enters the activation permission stage and jumps to step 1.
[0051] The present embodiment obtains a lateral control request; controls the lateral control system to enter an operating stage; obtains driving state information and driver control information in real time, and generates a lateral stage control signal according to the driving state information and the driver control information; when the lateral stage control signal is a lateral control maintenance signal, controls the lateral control system to continue to remain in the operating stage, and continues to obtain driving state information and driver control information, and determines whether the driving state information and the driver control information meet the conditions for exiting lateral control; when the lateral stage control signal is a lateral control exit signal, the lateral control system enters a slow retreat stage; generates a descent coefficient according to real-time driving state information; generates a final output control amount according to the descent coefficient; adjusts corresponding parameters according to the final output control amount; obtains the slow retreat time, and determines whether the slow retreat time is greater than a maximum time threshold; if yes, the lateral control system enters a hysteresis stage; if not, obtains the final output control amount , when the absolute value of the final output control amount is less than the safety threshold; the lateral control system enters the hysteresis stage; otherwise, continue to generate the descent coefficient according to the real-time driving status information; generate the final output control amount according to the descent coefficient; adjust the corresponding parameters according to the final output control amount; judge whether the lateral control system can exit the slow retreat stage and enter the hysteresis stage; after entering the hysteresis stage, obtain the hysteresis time for entering the hysteresis stage, and when the hysteresis time is greater than the set time threshold, the lateral control system enters the activation permission stage; at this time, the lateral control system can receive lateral control requests; by introducing the slow retreat stage after the lateral control system meets the exit conditions, various working conditions at the time of exit can be taken into account, and the corresponding parameters can be adjusted according to the driving status information at the time of exit and the driver's control information, which can not only ensure the normal exit of the system, but also consider various working conditions in the exit stage, effectively improve the system safety, but also improve the comfort.
[0052] Generate a lateral stage control signal according to the driving state information and the driver's control information, including: extracting the turn signal state, double flash state, brake pedal state and throttle state in the driver's control information; when any one of the turn signal state, double flash state, brake pedal state and throttle state is activated; generate an active exit signal; otherwise, extract the steering wheel torque in the driver's control information; determine whether the steering wheel torque is greater than the exit torque threshold,
[0053] If yes, an active exit signal is generated;
[0054] If no, then extract the lane line detection image in the driving status information; input the lane line detection image into the lane line recognition model to obtain the lane line label; when the lane line label is qualified, generate a lateral control maintenance signal; otherwise, generate a passive exit signal; wherein, the lane line recognition model is obtained through artificial intelligence model training; the lateral stage control signal is a lateral control exit signal or a lateral control maintenance signal; the lateral control exit signal is an active exit signal or a passive exit signal; the active exit signal is a signal for the driver to exit the lateral control system due to manual operation; the passive exit signal is a signal for the lateral control system to exit when the reference lane line of the lateral control system is lost, or the lane line is unclear, resulting in the lateral control system being unable to accurately perform lateral control; such as no lane line is detected in the detection image, or the lane line does not meet the standards, etc.
[0055] The lane recognition model is trained through an artificial intelligence model, including:
[0056] Acquire several lane line detection images and corresponding lane line labels; integrate the several lane line detection images and lane line labels into several groups of training data and test data; set lane line label experts according to the clarity and completeness of the lane lines in each lane line detection image; since the lateral control instructions generated according to the clear and complete lane lines have high accuracy, when the lane lines reach a certain clarity and completeness, the corresponding lane line labels are set as qualified, and the others are set as unqualified; since the lateral control instructions generated according to the clear and complete lane lines have high accuracy.
[0057] The artificial intelligence model is trained using training data, and the artificial intelligence model is tested using test data; finally, a lane line recognition model is obtained, whose input is a lane line detection image and whose output is a corresponding lane line label; wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.
[0058] Generating a drop coefficient according to the real-time driving state information, including: extracting the driving speed XV, the lateral vehicle distance HJ and the front wheel turning angle value QZ from the real-time driving state information;
[0059] The drop coefficient influencing factor g1 is calculated by the formula g1=exp(-k1×XV / ZV), where ZV is the maximum speed of the corresponding vehicle; k1 is the proportional coefficient; and k1>0;
[0060] In this embodiment, the reduction coefficient influence factor 1 is calculated through the above formula. When the vehicle's driving speed is higher, the corresponding steering adjustment safety margin is smaller, and the change of the steering angle value or torque value request needs to be smoother, that is, the corresponding reduction coefficient influence factor 1 is set smaller; so as to improve the comfort of the whole vehicle.
[0061] The second factor of the reduction coefficient g2 is calculated by the formula g2=k2×ln(HJ / AJ+1), where AJ is the safe lateral distance; k2 is the proportionality coefficient; and 1>k2>0;
[0062] In this embodiment, the second factor affecting the reduction coefficient is calculated through the above formula. When the lateral distance between the vehicle and the parallel vehicle is smaller, the risk of large adjustments is higher; the probability of collision is greater; in order to ensure safety, the corresponding steering adjustment amplitude is smaller; such an assisted driving controller can avoid risks to a greater extent in the slow retreat stage and improve the safety of the vehicle.
[0063] The third factor of the drop coefficient g3 is calculated by the formula g3=exp(-k3×QZ), where k3 is the proportional coefficient; and k3>0; it can be understood that the front wheel angle value can also be replaced by the steering wheel angle value, but the steering wheel angle value needs to be converted into the front wheel angle value; at the same time, the third factor of the drop coefficient can also be calculated by the torque value, and the third factor of the drop coefficient g3 is calculated by the formula g3=exp(-k31×QZ); where k31 is the proportional coefficient; and k31>0, k31≠k3;
[0064] In this embodiment, the third factor affecting the reduction coefficient is calculated through the above formula; when the requested steering angle value or torque value is larger, it often means that the vehicle deviates more from the target position. At this time, a relatively small change rate can ensure that the vehicle is corrected as much as possible during the slow retreat stage, thereby ensuring the safety of the vehicle.
[0065] When the lateral control exit signal is an active exit signal, the descending coefficient influence factor g4=dk is set; otherwise, the descending coefficient influence factor g4=dm is set; where dk is the fast adjustment coefficient, dm is the slow adjustment speed; and 0 <dm<dk<1;
[0066] In this embodiment, the fourth influencing factor of the reduction coefficient is calculated by the above formula; when it is detected that the lateral control system is exited manually, the authority of the lateral control should be handed over to the driver as soon as possible to ensure that the driver's operation can be responded to in a timely manner. When it is detected that the lateral control system is exited unintentionally, the lateral control system should be exited slowly, and the driver should be reminded of the reason for exiting the lateral control system, so that the driver can take over the lateral control of the vehicle in time.
[0067] It is worth noting that the specific values of the proportional coefficients k1, k2, k3 and k31 corresponding to different models are different. They can be set according to the total weight of the vehicle. The heavier the vehicle, the smaller the adjustment range of the corresponding turning angle value should be, and the larger the k1, k2, k3 and k31 should be set.
[0068] The descent coefficient G is calculated by the formula G = g1 × g2 × g3 × g4, and the calculated descent coefficient G satisfies 0 < G < 1; the descent coefficient is used to perform lateral control on the vehicle when the lateral control system is about to exit.
[0069] In this embodiment, the descent coefficient is used to control the output in the slow retreat stage, which can effectively ensure that the auxiliary driving controller can also consider various working conditions in the exit stage, and the required calibration parameters are less, reducing the R & D cost; at the same time, it can be quickly adapted to different vehicle models and has high generalization performance.
[0070] In another embodiment, each descent coefficient influencing factor is obtained by looking up a table; the driving speed, lateral vehicle distance, torque value or steering angle value in the real-time driving state information is extracted; the steering angle value includes the front wheel steering angle value or the steering wheel steering angle value; the descent coefficient influencing factor one is queried in the descent coefficient influencing factor one look-up table according to the driving speed; the descent coefficient influencing factor one look-up table is composed of several driving speeds and their corresponding descent coefficient influencing factors one, and the descent coefficient influencing factor one is set by an expert according to the driving speed; the descent coefficient influencing factor two is queried in the descent coefficient influencing factor two look-up table according to the lateral vehicle distance; the descent coefficient influencing factor two look-up table is composed of several lateral vehicle distances and their corresponding descent coefficient influencing factors two, and the descent coefficient influencing factor two is set by an expert according to the lateral vehicle distance; the descent coefficient influencing factor three is queried in the descent coefficient influencing factor three look-up table according to the torque value or the steering angle value; the descent coefficient influencing factor three look-up table is composed of several torque values and their corresponding descent coefficient influencing factors three, or the steering angle value and their corresponding descent coefficient influencing factors three, and the descent coefficient influencing factor three is set by an expert according to the torque value or the steering angle value; when the lateral control exit signal is an active exit signal, the descent coefficient influencing factor four g4 = dk is set; when the lateral control exit signal is a passive exit signal, the descent coefficient influencing factor four g4 = dm is set; dk is the fast adjustment coefficient, and dm is the slow adjustment speed; and 0 < dm < dk < 1; and then the descent coefficient G is calculated therefrom.
[0071] This embodiment has a good improvement in the safety and comfort of the vehicle lateral control exit stage, and can meet the lateral control system exit requirements of various vehicle models; after adding a slow retreat stage to the lateral control, when the lateral control system exits the lateral control, the control of the vehicle is smoother, and various working conditions in the exit stage can be considered, which can effectively cover the scenarios of exiting the lateral control of the vehicle at a relatively high speed, when deviating from the lane line relatively large, and when there is a relatively high collision risk with adjacent lane vehicles in the exit stage.
[0072] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.
[0073] How this application works:
[0074] The present application obtains a lateral control request; controls the lateral control system to enter the operation stage; obtains driving status information and driver control information in real time, and generates a lateral stage control signal according to the driving status information and the driver control information; when the lateral stage control signal is a lateral control maintenance signal, controls the lateral control system to continue to remain in the operation stage, and continues to obtain driving status information and driver control information, and determines whether the driving status information and the driver control information meet the conditions for exiting lateral control; when the lateral stage control signal is a lateral control exit signal, the lateral control system enters the slow retreat stage; generates a descent coefficient according to the real-time driving status information; generates a final output control amount according to the descent coefficient; adjusts the corresponding parameters according to the final output control amount; obtains the slow retreat time, and determines whether the slow retreat time is greater than the maximum time threshold; if so, the lateral control system enters the hysteresis stage; if not, obtains the final output Control quantity, when the absolute value of the final output control quantity is less than the safety threshold; the lateral control system enters the hysteresis stage; otherwise, continue to generate the descent coefficient according to the real-time driving status information; generate the final output control quantity according to the descent coefficient; adjust the corresponding parameters according to the final output control quantity; judge whether the lateral control system can exit the slow retreat stage and enter the hysteresis stage; after entering the hysteresis stage, obtain the hysteresis time for entering the hysteresis stage, and when the hysteresis time is greater than the set time threshold, the lateral control system enters the activation permission stage; at this time, the lateral control system can receive lateral control requests; by introducing the slow retreat stage after the lateral control system meets the exit conditions, various working conditions at the time of exit can be taken into account, and the corresponding parameters can be adjusted according to the driving status information at the time of exit and the driver's control information, which can not only ensure the normal exit of the system, but also take into account various working conditions at the exit stage, and effectively improve the system safety.
[0075] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.
Claims
1. A vehicle lateral control exit algorithm strategy, characterized in that: The following steps are involved: Step 1: obtaining a lateral control request; controlling the lateral control system to enter an operation phase according to the lateral control request; Step 2: Acquire driving state information and driver control information in real time; generate a lateral stage control signal according to the driving state information and the driver control information; when the lateral stage control signal is a lateral control exit signal, the lateral control system enters a slow retreat stage and enters step 3; otherwise, the lateral control system continues to remain in the operating stage and enters step 2; Step 3: Generate a drop coefficient based on real-time driving status information; Step 4: Generate the final output control amount according to the drop coefficient; Step 5: Get the slow-down time and determine whether the slow-down time is greater than the set maximum time threshold; if yes, proceed to step 6; if no, get the final output control amount. When the absolute value of the final output control amount is less than the safety threshold, proceed to step 6; otherwise, proceed to step 3; Step 6: The lateral control system enters the hysteresis stage and obtains the hysteresis time for entering the hysteresis stage. When the hysteresis time is greater than the set time threshold, the lateral control system enters the activation permission stage and jumps to step 1.
2. A vehicle lateral control exit algorithm strategy according to claim 1, characterized in that: The generating of the lateral stage control signal according to the driving state information and the driver control information comprises: Extract the turn signal status, double flash status, brake pedal status and throttle status from the driver's control information; when any one of the turn signal status, double flash status, brake pedal status and throttle status is activated; generate an active exit signal; otherwise, extract the steering wheel torque from the driver's control information; determine whether the steering wheel torque is greater than the exit torque threshold, If yes, an active exit signal is generated; If not, then extract the lane line detection image in the driving status information; input the lane line detection image into the lane line recognition model to obtain the lane line label; when the lane line label is qualified, generate a lateral control maintenance signal; otherwise, generate a passive exit signal; Among them, the lane line recognition model is obtained through artificial intelligence model training; the lateral stage control signal is a lateral control exit signal or a lateral control maintenance signal; the lateral control exit signal is an active exit signal or a passive exit signal.
3. A vehicle lateral control exit algorithm strategy according to claim 2, characterized in that: The lane line recognition model is obtained through artificial intelligence model training, including: Acquire a number of lane line detection images and their corresponding lane line labels; integrate the number of lane line detection images and lane line labels into a number of sets of training data and test data; The artificial intelligence model is trained using training data, and the artificial intelligence model is tested using test data; finally, a lane line recognition model is obtained, whose input is a lane line detection image and whose output is a corresponding lane line label; wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.
4. A vehicle lateral control exit algorithm strategy according to claim 1, characterized in that: The step of generating a drop coefficient according to the real-time driving status information includes: Extracting the driving speed XV, the lateral vehicle distance HJ and the turning angle value QZ from the real-time driving status information; The drop coefficient influencing factor g1 is calculated by the formula g1=exp(-k1×XV / ZV), where ZV is the maximum speed of the corresponding vehicle; k1 is the proportional coefficient; and k1>0; The second factor of the reduction coefficient g2 is calculated by the formula g2=k2×ln(HJ / AJ+1), where AJ is the safe lateral distance; k2 is the proportionality coefficient; and 1>k2>0; The third factor of the decline coefficient g3 is calculated by the formula g3=exp(-k3×QZ), where k3 is the proportional coefficient; and k3>0; When the lateral control exit signal is an active exit signal, the descending coefficient influence factor g4=dk is set; otherwise, the descending coefficient influence factor g4=dm is set; where dk is the fast adjustment coefficient, dm is the slow adjustment speed; and 0 <dm<dk<1; The reduction coefficient G is calculated by the formula G=g1×g2×g3×g4.
5. The vehicle lateral control exit algorithm strategy according to claim 1, characterized in that: The generating of the final output control amount according to the drop coefficient comprises: Obtain the requested turning angle value Tlast before the slow retreat phase, the maximum turning angle value change rate ΔTrate, the time t to enter the slow retreat phase, and the drop coefficient G; The final output control amount Ttorque is calculated by the formula Ttorque=Tlast-ΔTrate×G×t.
6. A vehicle lateral control exit algorithm strategy according to claim 1, characterized in that: The safety threshold is calculated by the formula Taq=ΔTrate×Δt; wherein Δt is the operation cycle of the auxiliary driving system.
