WO2025020025A1 - Dynamic safety filtering control method and domain control architecture for extreme driving functions - Google Patents

Abstract

A dynamic safety filtering control method and domain control architecture for extreme driving functions. The method comprises: S10, obtaining environment state information and function state information; S20, determining whether a vehicle is approaching a real-time physical constraint boundary; if yes, entering step S50; and if not, entering step S30; S30, determining whether the vehicle has exceeded a dynamic safety boundary; if yes, entering step S50; and if not, entering step S40; S40, completely trusting in an extreme driving control output; S50, calculating a target yaw angle and a target vehicle speed of the vehicle; S60, in light of a real-time physical constraint boundary finite time, intervening in the target yaw angle and/or the target vehicle speed, wherein intervening in the target yaw angle ensures that the vehicle is at a safe distance from the physical constraint boundary, and intervening in the vehicle speed ensures the dynamic stability of the vehicle; and S70, according to the target vehicle speed and the target yaw angle, calculating a safety control quantity.

Description

Dynamic safety filtering control method and domain control architecture for extreme driving functions Technical Field

The present invention relates to a dynamic safety filtering control method, device and domain control architecture for extreme driving functions, and relates to the field of vehicle extreme driving.

Background Art

According to statistics from the World Health Organization, more than 90% of traffic accidents are caused by driver errors. Among them, traffic accidents that cause major casualties mainly occur in extreme driving conditions. Under extreme driving conditions, cars are often in a state of high speed and sharp turns, the vehicle control margin is reduced, and it is more likely to become unstable. With the development of autonomous driving technology and chassis intelligence technology, some manufacturers have developed extreme safety functions and extreme driving experience functions for smart cars. At present, extreme driving functions are roughly divided into model-based solutions and data-driven methods. Among them, model-solving methods mostly design cost functions according to performance requirements, and seek the best control strategy within the constraints by minimizing the cost function; data-driven methods collect a large amount of "input-output" data and directly generate control strategies through machine learning and deep learning.

The extreme driving function based on model solving or data-driven focuses on optimizing the preset performance of the function, but ignores the potential safety risks that the extreme driving function itself brings to the vehicle. During the implementation of the extreme driving function, there may be sudden changes in the environment such as obstacles intruding or road changes. At this time, the preset performance indicators cannot meet the safety needs after the sudden change of the environment and the planning control cannot be adjusted in real time, which may lead to unexpected accidents. In addition, the extreme driving function has high requirements for the real-time and accuracy of the input signal. Sensor delays, oscillations or errors may lead to misuse of the function, which in turn causes vehicle body instability.

The current solution for the extreme driving function is to directly make autonomous decisions to take over vehicle control after the extreme driving control module is triggered. However, this solution lacks consideration for the safety of the extreme driving function and does not take into account the vehicle's electrical and electronic systems. The impact of the architecture may cause unexpected serious accidents due to sudden changes in the environment, misuse of functions, etc. The existing research on extreme driving ignores the potential safety risks of the function itself and does not focus on the prevention and control of potential dangers in the case of functional abnormalities, which is prone to unexpected safety accidents.

Summary of the invention

The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, in view of the above problems, the purpose of the present invention is to provide a dynamic safety filtering method, device and control architecture for extreme driving functions, which can monitor and analyze data-driven or model-solved extreme driving functions, and intervene in vehicle operations according to functional abnormalities to ensure that the vehicle is within the physical constraint boundaries.

In order to achieve the above-mentioned object of the invention, the technical solution of the present invention is:

In a first aspect, the present invention provides a dynamic safety filtering control method for an extreme driving function, comprising:

S10, obtaining environmental status information and vehicle function status information;

S20, determining whether the vehicle is close to the real-time physical constraint boundary, if yes, proceeding to step S50; if no, proceeding to step S30;

S30, determining whether the vehicle exceeds the dynamic safety boundary, if yes, proceeding to step S50; if no, proceeding to step S40;

S40, full trust in the extreme driving control output;

S50, calculating a target yaw angle and a target vehicle speed of the vehicle;

S60, intervening in a target yaw angle and/or a target vehicle speed for a limited time in combination with the real-time physical constraint boundary, wherein intervening in the target yaw angle ensures that the vehicle is at a safe distance from the physical constraint boundary, and intervening in the target vehicle speed ensures dynamic stability of the vehicle;

S70: Calculate a safety control amount according to the target vehicle speed and the target yaw angle.

Further, the determining whether the vehicle is close to a real-time physical constraint boundary includes:

The proximity of the vehicle is determined based on the distance between the vehicle and the physical constraint boundary and the relative movement speed. The real-time physical constraint boundary refers to the sudden road boundary or sudden obstacle in the road to which the current function cannot adapt.

Further, the determining whether the vehicle exceeds the dynamic safety boundary includes:

The dynamic safety boundary refers to the preset boundary of the vehicle's yaw rate and center of mass sideslip angle in the phase diagram. If the point of the vehicle's current state is not within the boundary, the vehicle is considered to have exceeded the dynamic safety boundary.

Furthermore, the calculation of the target yaw angle and target vehicle speed of the vehicle includes:

Establish a vehicle dynamics model to predict the next vehicle state based on the current state of the vehicle and the action of the extreme driving control module;

The vehicle speed and yaw angle are selected from the next vehicle state as the target tracking quantity, and the lateral deviation of the Frenet coordinate system is selected as the tracking reference quantity:

In the formula, are the expected vehicle speed, expected yaw rate, and expected normal speed along the road, s _t is the current vehicle state, a _input is the input before filtering, and s _t+1 is the state at the next moment.

Furthermore, the limited time intervention of the target yaw angle and/or target vehicle speed in combination with the real-time physical constraint boundary includes:

Combined control barrier function The target yaw angle is intervened by the increasing time scaling function Φ _T , which limits the influence of the approaching physical constraint boundary on the extreme driving control module to a finite time T:

Determine the risk of vehicle instability in real time and reduce the target speed v _d to obtain v _safe ;

In the formula, is the estimated yaw rate, l _d is the estimated normal velocity along the road, l is the actual normal velocity along the road, v is the total vehicle speed, ∈, b is a positive parameter, s _p is the maximum distance for controlling the intervention of the safety filter, β is the sideslip angle of the vehicle's center of mass, T is the preset finite time, t is the actual time after the filter is triggered, and n is the order.

In a second aspect, the present invention further provides a dynamic safety filtering control system for extreme driving functions, the system comprising:

An information acquisition module, configured to obtain environmental status information and functional status information;

A physical constraint boundary judgment module determines whether the vehicle is close to the real-time physical constraint boundary;

a dynamic safety boundary determination module, configured to determine whether the vehicle exceeds the dynamic safety boundary;

The output module is configured to fully trust the extreme driving output if the vehicle is not close to the real-time physical constraint boundary;

A target quantity calculation module is configured to calculate a target yaw angle and a target vehicle speed of the vehicle;

A target quantity intervention module is configured to intervene in the target quantity for a limited time in combination with the boundary information, wherein the target yaw angle is intervened to ensure that the vehicle has a safe distance from the physical constraint, and the vehicle speed is intervened to ensure the dynamic stability of the vehicle. If the vehicle state has exceeded the dynamic control safety boundary, the reverse steering and deceleration are simultaneously performed to stabilize the vehicle;

The control signal output module is configured to calculate a safety control amount according to a target vehicle speed and a target yaw angle.

In a third aspect, the present invention provides an intelligent vehicle domain control structure for extreme driving functions, the domain control structure comprising an automatic driving system and an intelligent chassis system; the automatic driving system is provided with an extreme driving comprehensive decision-making module and a fusion perception module, and the intelligent chassis system is provided with the dynamic safety filtering control system, the extreme driving control module and the conventional chassis dynamics controller described in the second aspect of the present invention; wherein,

The extreme driving comprehensive decision module is configured to shield the original stability control function of the vehicle, analyze whether there are unexpected changes in the environment that are not considered by the executed function based on the current and historical information obtained by the fusion perception module, and feed back the function status to the dynamic safety filtering control system;

The extreme driving control module, after being triggered by the extreme driving comprehensive decision module, inputs the control amount into the dynamic safety filter control system for intervention, and takes over the conventional chassis dynamics controller in a linear or inertial transition manner;

The dynamic safety filtering control system evaluates the behavior and status of the extreme driving function and determines the adaptability of the function to the current environment. If the vehicle is not close to the physical constraints, it will be directly output through the extreme driving control module; otherwise, it will filter out operations that may cause vehicle instability and violate physical constraints through dynamic adaptive filtering.

Furthermore, the automatic driving system also includes an automatic driving system decision planning and trajectory tracking module. When the fusion perception module feedbacks that the vehicle begins to move away from the newly added physical constraint boundary, the extreme driving comprehensive decision module transmits the normal environment and functional status to the dynamic safety filtering control system, and directly trusts the output of the extreme driving control module until the extreme driving control module completes the extreme driving task with predetermined performance. The extreme driving comprehensive decision module returns the control authority to the automatic driving system decision planning and trajectory tracking module, and the conventional chassis dynamics controller controls the steering system, drive system and braking system to perform the target operation, at which time the extreme driving function exits.

Furthermore, the extreme driving control module includes sequential decision-making and optimal control problems for optimizing preset extreme driving performance, optimizing preset performance indicators in the form of reward functions or cost functions, and solving front wheel steering angles and driving and braking targets driven by data or models.

Furthermore, the condition for triggering the extreme driving control module by the extreme driving comprehensive decision-making module is: the extreme driving comprehensive decision-making module monitors in real time during normal driving whether the vehicle's operating state and dynamic state are close to the safety boundary, and determines whether the function currently performed by the vehicle involves the adhesion limit, and triggers the extreme driving control module if it involves the adhesion limit.

The present invention adopts the above technical solution, and has the following characteristics:

1. The present invention focuses on the potential safety risks of extreme driving functions themselves, and can monitor and analyze data driven or The model solves the extreme driving functions and intervenes in vehicle operations based on functional abnormalities to ensure that the vehicle is within the physical constraint boundaries, thereby avoiding excessive sacrifice of preset performance to the greatest extent possible.

2. The emergence and development of the extreme driving function of the present invention has broadened the dynamic control boundary for the vehicle, greatly improved driving safety and driving experience, and at the same time improved the reliability and adaptability of the function itself.

3. The present invention is based on the idea of functional safety, and aims to ensure that when the extreme driving function is abnormal or faces unknown dangers, it can operate according to the preset performance degradation under the premise of ensuring safety, thereby ensuring operational safety and having broad application prospects.

4. The dynamic safety filtering method proposed in the present invention limits the intervention on the function to a limited time, thereby reducing the impact on the overall driving performance.

In summary, the present invention can be widely applied to vehicle extreme driving.

BRIEF DESCRIPTION OF THE DRAWINGS

Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Throughout the accompanying drawings, the same reference numerals are used to represent the same components. In the accompanying drawings:

FIG1 is a flow chart of a dynamic safety filtering control method for extreme driving functions provided by an embodiment of the present invention.

FIG2 is a schematic diagram of an intelligent vehicle domain control architecture that adapts to extreme driving functions provided by an embodiment of the present invention.

DETAILED DESCRIPTION

It should be understood that the terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms "a,""an," and "the" may also be intended to include the plural forms unless the context clearly indicates otherwise. The terms "include,""comprises,""contains," and "have" are inclusive and thus specify that the stated features, steps, operations, elements, and/or components are intended to include at least one embodiment of the present invention. The existence of a feature, step, operation, element, component, and/or combination thereof is not excluded, but the existence or addition of one or more other features, steps, operations, elements, components, and/or combinations thereof is not excluded. The method steps, processes, and operations described herein are not to be interpreted as requiring them to be performed in the specific order described or illustrated, unless the execution order is explicitly indicated. It should also be understood that additional or alternative steps may be used.

Although the terms first, second, third, etc. can be used in the text to describe multiple elements, components, regions, layers and/or sections, these elements, components, regions, layers and/or sections should not be limited by these terms. These terms can only be used to distinguish an element, component, region, layer or section from another region, layer or section. Unless the context clearly indicates, terms such as "first", "second" and other numerical terms do not imply order or sequence when used in the text. Therefore, the first element, component, region, layer or section discussed below can be referred to as the second element, component, region, layer or section without departing from the teaching of the example embodiments.

For ease of description, spatially relative terms may be used herein to describe the relationship of one element or feature relative to another element or feature as shown in the figures, such as "inside", "outside", "inner side", "outer side", "below", "above", etc. Such spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation depicted in the figures.

Since the extreme driving function based on model solving or data driving in the prior art focuses on optimizing the performance of the preset function, but does not pay attention to the safety risks brought by the extreme driving function itself to the vehicle, and does not pay attention to the prevention and control of potential dangers in the case of functional abnormalities, unexpected safety accidents are prone to occur. The present invention provides a dynamic safety filtering method, device and intelligent vehicle domain control architecture for extreme driving functions, which can monitor and analyze data-driven or model-solved extreme driving functions, and intervene in vehicle operations according to functional abnormalities to ensure that the vehicle is within the physical constraint boundaries.

Dynamic safety filtering method, system and intelligent vehicle domain control for extreme driving functions proposed in the present invention The architecture is oriented to data-driven or model-solving extreme driving functions and has the following characteristics:

(1) Behavior monitoring and analysis: By combining the cloud control system, autonomous driving system, and intelligent chassis system, behavior monitoring and analysis of extreme driving functions are performed to understand the system's working status and driving behavior in real time;

(2) Dynamic safety filtering: Adaptive hybrid control is combined with classic switching control to intervene in the output and behavior of the extreme driving module in the form of dynamic safety filtering, which can solve the functional safety challenges caused by sudden changes in the environment, misuse of functions, and other factors.

(3) Limited time intervention: By combining the control obstacle function with the increasing time scale function, the intervention of the dynamic safety filter on the main control module is limited to a limited time, so as to reduce the impact of the dynamic safety filter on the overall driving performance.

In summary, the present invention fully considers the impact on the overall driving performance while ensuring the safety and stability of the extreme driving function, and provides an effective safety protection solution for the extreme driving function.

The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

Embodiment 1: As shown in FIG1 , the dynamic safety filtering control method for extreme driving function provided in this embodiment includes:

S10, Environmental status and extreme driving function status analysis

In this embodiment, environmental status information such as pedestrian approach information, obstacle information, road adhesion coefficient, vehicle position and posture information, and extreme driving function status information are obtained, including but not limited to the currently executed automatic driving functions, active safety functions and stability functions, and it is analyzed whether there are unexpected changes in the environment that are not taken into account by the functions executed by extreme driving.

S20, determining whether the vehicle is close to the real-time physical constraint boundary, if yes, proceeding to step S50; if no, proceeding to step S30;

In this embodiment, the physical constraint boundary refers to a sudden change in the road boundary or an emergent obstacle in the road to which the currently executed function cannot adapt.

In this embodiment, the extreme driving fusion perception module and the V2X collaborative environmental perception system determine the physical constraints in real time, and judge the degree of vehicle proximity based on factors such as the distance between the vehicle and the physical constraint boundary and the relative movement speed.

S30, determining whether the vehicle exceeds the dynamic safety boundary, if yes, proceeding to step S50; if no, proceeding to step S40;

In this embodiment, the dynamic safety boundary refers to the preset boundary of the vehicle's yaw rate and center of mass sideslip angle in the power system phase diagram. If the point of the vehicle's current state is not within the boundary, it is considered that the vehicle exceeds the dynamic safety boundary.

S40, full trust in the extreme driving control output.

S50, calculating the target speed, yaw angle and lateral displacement of the vehicle;

In this embodiment, the target speed, yaw angle and lateral displacement of the vehicle are calculated, including:

Establish a high-precision vehicle dynamics model to predict the next vehicle state based on the current state of the vehicle and the action of the extreme driving control module;

The vehicle speed and yaw angle are selected from the next vehicle state as the target tracking quantity, and the lateral deviation of the Frenet coordinate system is selected as the tracking reference quantity:

In the formula, are the expected vehicle speed, expected yaw rate, and expected normal speed along the road, s _t is the current vehicle state, a _input is the input before filtering, and s _t+1 is the state at the next moment.

It should be noted that the vehicle dynamics model is applicable to but not limited to 2-15 degree of freedom vehicle dynamics models, vehicle models built into commercial software such as Carsim, etc., and can be selected according to actual needs without limitation here.

S60, combining boundary information with limited time intervention target quantity;

In this embodiment, dynamic safety intervention is performed on the target tracking amount, including intervention on the target yaw angle to ensure that the vehicle is at a safe distance from the physical constraint boundary, and intervention on the vehicle speed to ensure the dynamic stability of the vehicle. If the vehicle state has exceeded the dynamic control safety boundary, reverse steering and deceleration are performed simultaneously to stabilize the vehicle.

First, combined with the control barrier function The target yaw angle is intervened by the increasing time scaling function Φ _T , which limits the influence of the approaching physical constraint boundary on the extreme driving control module within a limited time T, thereby minimizing the negative impact on the overall driving performance.

In the formula, is the estimated yaw rate, l _d is the estimated normal velocity along the road, l is the actual normal velocity along the road, v is the total vehicle speed, ∈, b is a positive parameter, s _p is the maximum distance for controlling the intervention of the safety filter, β is the sideslip angle of the vehicle's center of mass, T is the preset finite time, t is the actual time after the filter is triggered, and n is the order.

Secondly, for the risk of dynamic instability caused by factors such as sudden changes in the road surface, the risk of vehicle instability is determined in real time, and limiting measures such as lowering the target vehicle speed v _d to obtain v _safe and limiting the front wheel turning angle are taken.

Finally, the safety control quantity a _safe is calculated through feedback control based on the target quantity and the current state:

Where a _st is the deceleration and steering input applied to stabilize the vehicle, which can usually be obtained by looking up the table; It is the input obtained by tracking the corrected target quantity.

S70, tracking the target quantity to obtain a safety control signal.

In this embodiment, the target vehicle speed and the target yaw angle are tracked, the control quantity is calculated through feedback control such as PID control and sliding mode control, and the control signal is sent to the vehicle braking system, drive system, steering system and suspension system.

Embodiment 2: Embodiment 1 above provides a dynamic safety filtering control method for extreme driving functions. Correspondingly, this embodiment provides a dynamic safety filtering control system for extreme driving functions. The device provided in this embodiment can implement the dynamic safety filtering control method for extreme driving functions of embodiment 1, and the device can be implemented by software, hardware, or a combination of software and hardware. For the convenience of description, this embodiment is described in various units according to their functions. Of course, the functions of each unit can be implemented in the same or multiple software and/or hardware during implementation. For example, the device may include integrated or separate functional modules or functional units to execute the corresponding steps in each method of embodiment 1. Since the device of this embodiment is basically similar to the method embodiment, the description process of this embodiment is relatively simple. For relevant matters, please refer to the partial description of embodiment 1. The embodiment of the dynamic safety filtering control system for extreme driving functions provided by the present invention is only schematic.

Specifically, the present invention also provides a dynamic safety filtering control system for extreme driving functions, the system comprising:

An information acquisition module, configured to obtain environmental status information and functional status information;

A physical constraint boundary judgment module determines whether the vehicle is close to the real-time physical constraint boundary;

a dynamic safety boundary determination module, configured to determine whether the vehicle exceeds the dynamic safety boundary;

The output module is configured to fully trust the extreme driving output if the vehicle is not close to the real-time physical constraint boundary;

A target quantity calculation module is configured to calculate a target yaw angle and a target vehicle speed of the vehicle;

The target quantity intervention module is configured to intervene in the target quantity for a limited time in combination with the boundary information, wherein the target The yaw angle is intervened to ensure that the vehicle is at a safe distance from the physical constraints, and the vehicle speed is intervened to ensure the dynamic stability of the vehicle. If the vehicle state has exceeded the dynamic control safety boundary, the reverse steering and speed reduction are performed simultaneously to stabilize the vehicle.

The control signal output module is configured to calculate a safety control amount according to a target vehicle speed and a target yaw angle.

In a preferred embodiment of the present invention, for cars that are not equipped with high-level autonomous driving, the dynamic safety filtering control system proposed by the present invention is also suitable for analyzing, monitoring and intervening in the driver's aggressive extreme operations.

Embodiment 3: As shown in FIG2 , the intelligent vehicle domain control architecture for extreme driving functions provided in this embodiment includes an automatic driving system 1 , a chassis domain controller 2 (intelligent chassis system) and a cloud control system 3 .

The autonomous driving system 1 is provided with an extreme driving comprehensive decision-making module 11, a fusion perception module 12 and an autonomous driving system decision planning and trajectory tracking module 13; the chassis domain controller 2 is provided with an extreme driving control module 21, a dynamic safety filtering control system 22 and a conventional chassis dynamics controller 23; the cloud control system 3 is provided with a V2X collaborative environment perception module 31 and a cloud computing platform 32; wherein,

The extreme driving comprehensive decision module 11 is configured to monitor in real time during normal driving whether the vehicle's operating state and dynamic state are close to the safety boundary, determine whether the current vehicle function involves the adhesion limit, and trigger the extreme driving control module 21 if it involves the adhesion limit, and coordinate the cloud computing platform 32, the vehicle computing unit and the chassis domain controller 2 to allocate computing power according to the computing power required for the extreme driving function. During the execution of the extreme driving function, the extreme driving comprehensive decision module 11 shields the original stability control function of the vehicle, and analyzes whether there are unexpected changes in the environment that are not considered by the executed function based on the current and historical information obtained by the fusion perception module 12, including the environment and itself, the behavior and state of the extreme driving control module, and feeds back the function state (the adaptation of the function to the environment and itself, such as sudden changes in the environment, or the function requires a normal braking system but the braking system is currently partially/severely failed) to the dynamic safety filtering control system 22 in the chassis domain controller 2.

The extreme driving control module 21 is triggered by the extreme driving comprehensive decision module 11 and inputs the control amount into the dynamic The dynamic safety filter control system 22 performs safety intervention and takes over the conventional chassis dynamics controller 23 in a linear or inertial transition manner;

The dynamic safety filtering control system 22 evaluates the behavior and status of the extreme driving function and determines the adaptability of the function to the current environment. For example, if the vehicle is not close to the physical constraint boundary, it will be directly output through the extreme driving control module 212; otherwise, it will use dynamic adaptive filtering to filter out operations that may cause vehicle instability and violate physical constraints.

In a preferred embodiment of the present invention, the extreme driving control module 21 uses a data-driven reinforcement learning algorithm, and the output control variables include the front wheel steering angle δ _f , the throttle opening and the brake pedal opening τ, τ>0 represents the throttle opening, and when τ>0, |τ| is taken as the brake pedal opening. The vehicle state includes the horizontal and vertical coordinates x _t ,l of the vehicle in the Frenet coordinate system, the vehicle speed _vs ,v _l , and the yaw angle

Furthermore, the extreme driving control module 21 includes sequential decision-making and optimal control problems for optimizing preset extreme driving performance, optimizing preset performance indicators in the form of reward functions or cost functions, and solving control quantities such as front wheel steering angle, driving and braking targets, etc. driven by data or models.

Furthermore, the extreme driving control module 21 includes but is not limited to performance-driven optimal control such as linear/nonlinear model predictive control algorithms, and selects different reinforcement learning algorithms and their corresponding improved algorithms according to actual conditions, which will not be elaborated here.

In a preferred embodiment of the present invention, the exit mechanism of the extreme driving function is as follows: when the fusion perception module 11 feedbacks that the vehicle begins to move away from the newly added physical constraint boundary (when moving away from the newly added physical constraint boundary, for example, the distance between the vehicle and the physical constraint boundary becomes farther, the relative motion speed is negative, etc.), the extreme driving comprehensive decision module 11 transmits the normal environment and functional state to the dynamic safety filtering control system 22, and directly trusts the output of the extreme driving control module 21. Until the extreme driving control module 21 completes the extreme driving task with the predetermined performance, the extreme driving comprehensive decision module 11 returns the control authority to the automatic driving system decision planning and trajectory tracking module 13, and the conventional chassis dynamics controller 23 controls the steering system, drive system and braking system to perform the target operation.

In a preferred embodiment of the present invention, the fusion perception module 11 obtains current and historical information through the sensor combination 4 supporting the extreme driving function, including lidar signals, combined inertial navigation signals, wheel speed signals, and the vehicle dynamics control signals include feedback signals of the drive system, braking system, steering system, and suspension system.

The following describes in detail the workflow of the intelligent vehicle domain control architecture for extreme driving functions of this embodiment through specific embodiments.

This embodiment assumes that the environment changes suddenly and a pedestrian approaches the vehicle that is at the adhesion limit. However, the original extreme driving function cannot adapt to the change of the physical constraint boundary due to limited generalization.

a _input = [δ _f , τ]

Where x,l are the coordinates of the vehicle along the tangent and normal directions of the road; _δf is the front wheel steering, τ is the throttle opening, _vs , _vl are the velocity components of the vehicle along the tangent and normal directions of the road respectively.

When no probing pedestrian is detected, the extreme driving function operates normally, the extreme driving comprehensive decision module 11 selects the extreme driving function, and the chassis domain controller 2 takes over the conventional chassis dynamics controller 23 to execute the data-driven extreme driving control strategy. The dynamic safety filtering control system 22 does not detect that the vehicle is approaching the physical constraint boundary, nor does it detect that the vehicle has the risk of instability beyond the dynamic safety boundary, so it fully trusts the extreme driving control output.

When the sensor combination 4 supporting extreme driving and the V2X cooperative environment perception module 31 integrate the perception calculation to obtain the pedestrian approach information, the obstacle information, road adhesion coefficient, vehicle position and posture are transmitted to the extreme driving comprehensive decision module 11. The extreme driving comprehensive decision module 11 transmits the environmental sudden change and functional abnormality information to the dynamic safety filtering control system 22. The dynamic safety filtering control system 22 determines that the vehicle is close to the real-time physical constraint boundary, and then calculates the target speed, yaw angle and lateral displacement of the vehicle, combines the boundary information to intervene in the target quantity for a limited time, and tracks the target quantity to obtain a safety control signal, which is specifically:

First, the current vehicle state s _t and _{the input} before filtering a input are substituted into the vehicle dynamics model to calculate the next time At the moment state s _t+1 , since the state selected in this embodiment is in the Frenet coordinate system, it should be converted to the Cartesian coordinate system for dynamic model calculation, and then converted into the Frenet coordinate system, and the vehicle speed, yaw angle and lateral deviation are selected as the tracking target quantity.

Secondly, the target quantity is intervened in combination with the newly added physical constraints.

Since the present embodiment mainly involves the newly added physical constraints (s _n , l _n ), the target yaw angle To prevent the newly added physical constraint from being less than the safety distance s _m , the following control obstacle function is designed in this embodiment:

Δs ² =(xx _n ) ² +(ll _n ) ²

Where k _r is a positive parameter for adjusting the degree of intervention of the dynamic safety filter control system on the extreme driving control module, and (x _n , l _n ) is the real-time coordinate of the probe pedestrian.

In order to limit the impact of the new constraint on the extreme driving control module within the time T, this embodiment combines the incremental time scale function Φ _T to obtain the target yaw angle after intervention: Due to the existence of the increasing time scaling function Φ _T , the obstacle control function The impact time on the target tracking volume is limited to T.

Where, ∈,b is a positive parameter, _sp is the maximum distance for controlling the intervention of safety filtering, and β is the sideslip angle of the vehicle center of mass.

Since there are no factors that cause vehicle instability in this embodiment, the extreme driving comprehensive decision module 11 determines that the vehicle is within the dynamic safety boundary. The dynamic safety filtering control system 22 does not intervene in the target speed, that is, v _t =v _d , and obtains the target tracking amount after intervention:

Then, the safety control amount a _safe =[δ _fs ,τ _s ] is calculated through feedback control according to the target amount and the current state. The method selected in this embodiment is model-based sliding mode control.

Finally, the control amount a _safe is converted into a control signal, which is transmitted to the steering system, drive system and brake system through the vehicle wire control system to perform the control operation. At the same time, the intelligent vehicle in this embodiment can be equipped with an active suspension system, which will adjust the suspension state according to the calculated control signal and the vehicle body state to improve the handling stability during extreme driving.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In the description of this specification, the description of the reference terms "a preferred embodiment", "further", "specifically", "in the present embodiment", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of this specification. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradiction.

The present application is described with reference to the flowchart and/or block diagram of the method, device (apparatus), and computer program product according to the embodiment of the present application. It should be understood that each flow and/or box in the flow chart and/or block diagram, and the combination of the flow chart and/or box in the flow chart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one flow chart or multiple flows and/or one box or multiple boxes of the block chart.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

A dynamic safety filtering control method for extreme driving functions, characterized by comprising: S10, obtaining environmental status information and vehicle extreme driving function status information; S20, determining whether the vehicle is close to the real-time physical constraint boundary, if yes, proceeding to step S50; if no, proceeding to step S30; S30, determining whether the vehicle exceeds the dynamic safety boundary, if yes, proceeding to step S50; if no, proceeding to step S40; S40, full trust in the extreme driving control output; S50, calculating a target yaw angle and a target vehicle speed of the vehicle; S60, intervening in a target yaw angle and/or a target vehicle speed in combination with a boundary limited time, wherein intervening in the target yaw angle ensures that the vehicle is at a safe distance from the physical constraint boundary, and intervening in the target vehicle speed ensures dynamic stability of the vehicle; S70: Calculate a safety control amount according to the target vehicle speed and the target yaw angle. The dynamic safety filtering control method for extreme driving functions according to claim 1 is characterized in that the step of determining whether the vehicle is close to a real-time physical constraint boundary comprises: The proximity of the vehicle is determined based on the distance between the vehicle and the physical constraint boundary and the relative movement speed. The real-time physical constraint boundary refers to the sudden road boundary or sudden obstacle in the road to which the current function cannot adapt. The dynamic safety filtering control method for extreme driving functions according to claim 1 is characterized in that the step of determining whether the vehicle exceeds the dynamic safety boundary comprises: The dynamic safety boundary refers to the preset boundary of the vehicle's yaw rate and center of mass sideslip angle in the phase diagram. If the point of the vehicle's current state is not within the boundary, the vehicle is considered to have exceeded the dynamic safety boundary. The dynamic safety filtering control method for extreme driving functions according to claim 1 is characterized in that the calculation of the target yaw angle and target vehicle speed of the vehicle comprises: Establish a vehicle dynamics model to predict the next vehicle state based on the current state of the vehicle and the action of the extreme driving control module; The vehicle speed and yaw angle are selected from the next vehicle state as the target tracking quantity, and the lateral deviation of the Frenet coordinate system is selected as the tracking reference quantity:
In the formula, are the expected vehicle speed, expected yaw rate, and expected normal speed along the road, s _t is the current vehicle state, a _input is the input before filtering, and s _t+1 is the state at the next moment. The dynamic safety filtering control method for extreme driving functions according to claim 1 is characterized in that the limited time intervention of the target yaw angle and/or target vehicle speed in combination with the real-time physical constraint boundary comprises: Combined control barrier function The target yaw angle is intervened by the increasing time scaling function Φ _T , which limits the influence of the approaching physical constraint boundary on the extreme driving control module to a finite time T:

Determine the risk of vehicle instability in real time and reduce the target speed v _d to obtain v _safe ; In the formula, is the estimated yaw rate, l _d is the estimated normal velocity along the road, l is the actual normal velocity along the road, v is the total vehicle speed, ∈, b is a positive parameter, s _p is the maximum distance for controlling the intervention of the safety filter, β is the sideslip angle of the vehicle's center of mass, T is the preset finite time, t is the actual time after the filter is triggered, and n is the order. A dynamic safety filtering control system for extreme driving functions, characterized in that the system comprises: An information acquisition module, configured to obtain environmental status information and functional status information; A physical constraint boundary judgment module determines whether the vehicle is close to the real-time physical constraint boundary; a dynamic safety boundary determination module, configured to determine whether the vehicle exceeds the dynamic safety boundary; The output module is configured to fully trust the extreme driving output if the vehicle is not close to the real-time physical constraint boundary; A target quantity calculation module is configured to calculate a target yaw angle and a target vehicle speed of the vehicle; A target quantity intervention module is configured to intervene in the target quantity for a limited time in combination with the boundary information, wherein the target yaw angle is intervened to ensure that the vehicle has a safe distance from the physical constraint, and the vehicle speed is intervened to ensure the dynamic stability of the vehicle. If the vehicle state has exceeded the dynamic control safety boundary, the reverse steering and deceleration are simultaneously performed to stabilize the vehicle; The control signal output module is configured to calculate a safety control amount according to a target vehicle speed and a target yaw angle. An intelligent vehicle domain control structure for extreme driving function, characterized in that the domain control structure includes an automatic driving system and an intelligent chassis system; the automatic driving system is provided with an extreme driving comprehensive decision module and a fusion perception module, and the intelligent chassis system is provided with a dynamic safety filtering control system as claimed in claim 6, an extreme driving control module and a conventional chassis dynamics controller; wherein, The extreme driving comprehensive decision module is configured to shield the original stability control function of the vehicle, analyze whether there are unexpected changes in the environment that are not considered by the executed function based on the current and historical information obtained by the fusion perception module, and feed back the function status to the dynamic safety filtering control system; The extreme driving control module, after being triggered by the extreme driving comprehensive decision module, inputs the control amount into the dynamic safety filter control system for intervention, and takes over the conventional chassis dynamics controller in a linear or inertial transition manner; The dynamic safety filtering control system evaluates the behavior and status of the extreme driving function and determines the adaptability of the function to the current environment. If the vehicle is not close to the physical constraints, it will be directly output through the extreme driving control module; otherwise, it will filter out operations that may cause vehicle instability and violate physical constraints through dynamic adaptive filtering. According to the intelligent vehicle domain control structure of claim 7, it is characterized in that the automatic driving system also includes an automatic driving system decision planning and trajectory tracking module. When the fusion perception module feedbacks that the vehicle begins to move away from the newly added physical constraint boundary, the extreme driving comprehensive decision module transmits the normal environment and functional status to the dynamic safety filtering control system, and directly trusts the output of the extreme driving control module until the extreme driving control module completes the extreme driving task with predetermined performance. The extreme driving comprehensive decision module returns the control authority to the automatic driving system decision planning and trajectory tracking module, and the conventional chassis dynamics controller controls the steering system, drive system and braking system to perform the target operation, and the extreme driving function exits at this time. The intelligent vehicle domain control structure according to claim 7 or 8 is characterized in that the extreme driving control module includes a sequence decision and optimal control problem for optimizing preset extreme driving performance, optimizes preset performance indicators in the form of a reward function or a cost function, and solves the front wheel angle and driving and braking targets driven by data or models. According to the intelligent vehicle domain control structure of claim 7, it is characterized in that the condition that the extreme driving control module is triggered by the extreme driving comprehensive decision-making module is: the extreme driving comprehensive decision-making module monitors the vehicle's operating state and dynamic state in real time during conventional driving to see whether they are close to the safety boundary, and determines whether the function currently performed by the vehicle involves the adhesion limit, and triggers the extreme driving control module if it involves the adhesion limit.