Active steering and braking integrated control system based on roll angle feedback

Abstract

The invention discloses an active steering and braking integrated control system based on roll angle feedback, and belongs to the technical field of vehicle dynamics control. In order to solve the problems of slow control response and multi-actuator action conflict of the vehicle under the limit working condition, the method quantifies the comprehensive instability risk of the whole vehicle by constructing a dynamic virtual potential field. The potential field is based on key state variables such as a roll angle and a side slip angle, and the topological structure (namely the weight of each variable) of the potential field is remodeled in real time by a dynamic risk factor reflecting the intention urgency degree of a driver and the environmental danger level. The system evaluates the instantaneous control efficiency by calculating the contribution degree of each actuator to the reduction of the potential field gradient in a distributed manner. According to the method, the complex control problem is converted into the process of finding the minimum potential energy point, prospective suppression of the instability trend and conflict-free efficient cooperation between the actuators are achieved, and the limit stability and safety of the vehicle are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle dynamics control, in particular to an active steering and braking integrated control system based on roll angle feedback. BACKGROUND

[0002] Modern vehicles are generally equipped with electronic stability control systems (ESC), which apply braking force to individual or multiple wheels to correct the oversteer or understeer that deviates from the driver's intention during driving, greatly improving driving safety. The control logic of such systems is usually based on the driver's steering wheel angle input to calculate a desired yaw rate, which is compared with the actual yaw rate measured by sensors. When the error between the two exceeds a preset threshold, the system initiates braking intervention.

[0003] However, with the development of vehicle chassis control technology, this traditional error tracking-based control paradigm gradually reveals its inherent limitations. First, the system's intervention is essentially lagging, as it must respond only after the instability trend has become apparent, i.e., a large enough tracking error has occurred. In some transient, high-risk emergency obstacle avoidance conditions, this response delay reduces valuable control opportunities and limits the stability control limits that the system can achieve. In addition, the control target mainly focuses on the motion in the yaw plane, and the vehicle roll motion, which is a key factor in rollover, is not given enough proactive attention.

[0004] Furthermore, to pursue higher handling performance and safety, more active control actuators such as active steering systems are introduced into vehicles. This makes the vehicle's stability control evolve from single braking system intervention to a complex coordination problem involving steering and braking and other subsystems. In the prior art, a hierarchical control architecture is usually used to address this challenge. That is, an upper-level arbitration controller coordinates and allocates control tasks for lower-level actuators according to a set of preset, relatively fixed priority rules or complex allocation algorithms. This centralized arbitration mechanism not only increases the complexity and calibration difficulty of the control algorithm, but also introduces additional computational time delay in the decision-making process. In dynamic and variable extreme conditions, the fixed priority allocation logic is difficult to make optimal decisions at all times, and even in certain scenarios, it can cause action conflicts or effect cancellation between different actuators, thereby weakening the overall stability control effect of the vehicle.

[0005] Therefore, a new vehicle stability control method is needed that can break free from the dependence on error lag response, proactively estimate and suppress comprehensive instability risks, and achieve intrinsic, self-organizing, conflict-free, and efficient coordination among multiple actuators, thereby breaking through the bottleneck of existing technology. SUMMARY

[0006] The present application aims to provide a side tilt angle feedback-based active steering and braking integrated control system and method, aiming to solve the problems of control response delay, action conflict between active steering and braking and lack of vehicle instability prediction in the prior art.

[0007] To achieve the above-mentioned purpose, the first aspect of the present application provides a side tilt angle feedback-based active steering and braking integrated control method, which realizes self-organizing and cooperative control of each actuator based on a dynamic virtual potential field reflecting the comprehensive stability risk of the vehicle, so as to achieve rapid, forward-looking and coordinated control of the vehicle attitude, especially the roll stability.

[0008] In a specific embodiment, the method comprises the following steps:

[0009] Step one: obtaining a plurality of vehicle state variables including the vehicle roll angle. The system collects a series of key state variables representing the current motion attitude of the vehicle in real time through the vehicle-mounted sensor network, mainly including: roll angle φ, roll angle speed center side slip angle β and yaw rate γ. These variables together constitute a multi-dimensional state space for evaluating the stability of the vehicle.

[0010] Step two: based on the vehicle state variables, and according to the dynamic risk factor reflecting the driver's intention or external environment, a vehicle stability virtual potential field is constructed, wherein the dynamic risk factor is used to dynamically adjust the weight coefficient corresponding to each vehicle state variable in the vehicle stability virtual potential field.

[0011] This step is one of the core innovations of the present application. The system constructs a scalar form of vehicle stability virtual potential field V, the value of which is used to quantify the instantaneous instability risk of the vehicle. The potential field is defined by the following formula:

[0012]

[0013] wherein, is the real-time vehicle state variable obtained in step one; γ d is the expected yaw rate calculated based on the driver's steering wheel input and the current vehicle speed, representing the driver's path following intention; is the weight coefficient corresponding to each state variable.

[0014] Key, the weight coefficient is not a fixed value, but a function adjusted in real time by one or more dynamic risk factors. The dynamic risk factor can include:

[0015] Driver intent urgency factor: quantified by monitoring steering wheel angular velocity or the degree of brake pedal operation, reflecting whether the driver is in an emergency obstacle avoidance or other intense driving state.

[0016] Environmental potential risk factor: quantified by information such as road curvature, road adhesion coefficient estimate obtained by the vehicle-mounted perception system, reflecting the challenge of the external environment to the vehicle stability.

[0017] When the risk represented by any dynamic risk factor increases (for example, the driver hits the steering wheel or the system identifies a slippery road), the corresponding weight coefficient (for example, ) will increase nonlinearly. This mechanism enables the "terrain" of the vehicle stability virtual potential field to be dynamically reshaped, and when facing the risk of rolling or sliding, the potential field will become unusually "steep" in the corresponding dimension, thus providing more sensitive and accurate risk criteria for subsequent control decisions. Step three: calculate the contribution of the active steering system and the braking system to reducing the potential field gradient of the vehicle stability virtual potential field. This step realizes distributed calculation of control decisions. The system first calculates the gradient vector of the vehicle stability virtual potential field V at the current state point This vector points to the fastest direction of risk increase. Then, for each independent actuator (including the active steering system and each brake unit in the braking system that can independently adjust the braking force of a single wheel), the system will independently calculate its contribution to reducing risk. The potential field gradient descent contribution G i of the i-th actuator is calculated by the following formula:

[0018]

[0019] Where b i is a vector representing the impact on the vehicle state variable when the i-th actuator applies a unit control input. G i is a scalar value that intuitively reflects the efficiency and ability of the actuator to suppress the vehicle instability trend at the current time.

[0020] Step four: generate and execute the coordinated control instructions of the active steering system and the braking system based on the potential field gradient descent contribution.

[0021] This step is based on the results of the aforementioned distributed calculation, realizing the self-organizing coordination of each actuator. The generation of control instructions does not depend on traditional central allocation logic, but follows a unified principle: the size of the control instruction of each actuator is proportional to its potential field gradient descent contribution G iThe proportion of all positive contribution degrees is proportional. This means that the stronger the executor's contribution ability, the more control tasks it will automatically undertake. In addition, the overall control strength of the cooperative control instruction is proportional to the current value V of the vehicle stability virtual potential field, realizing the on-demand allocation of control strength: the higher the risk, the stronger the intervention; when the risk is low, unnecessary control actions are not generated.

[0022] The application provides an active steering and braking integrated control system based on roll angle feedback. The application has the following beneficial effects:

[0023] 1. The instability risk is suppressed in advance, and the stability margin is improved. The application quantifies the comprehensive stability of the vehicle as potential energy by constructing a virtual potential field that dynamically remodels with the driver's intention and environmental risk. When the potential risk increases, the "terrain" of the potential field will become steep in advance, thus generating strong "restoring force" when the instability trend is still in the embryonic stage, guiding the vehicle state to return to the stable domain. Compared with the traditional error tracking-based lag response, this method can intervene earlier and significantly improve the stability margin of the vehicle in transient operating conditions.

[0024] 2. The control conflicts between executors are eliminated, and efficient cooperation is achieved. The application calculates the contribution degree of each executor to reducing the system potential energy in a distributed manner, and allocates control instructions in proportion based on this as the only basis. This mechanism allows the executor with the highest control efficiency to naturally obtain the maximum control weight at any time, and all executors work together for the unified goal of "reducing total potential energy". This self-organizing feature eliminates the executor action conflicts that may be caused by fixed priorities or complex arbitration logic in traditional hierarchical control, ensuring the optimization of the overall control effect.

[0025] 3. The vehicle's rollover resistance is significantly enhanced. The application directly controls the roll angle and roll angular velocity, which are directly related to the vehicle's rollover risk, as the core state variables for constructing the virtual potential field. This means that the objective function of the control system fundamentally includes the maintenance of roll stability. Therefore, compared with traditional electronic stability control systems that mainly focus on yaw motion, the application can more directly and effectively suppress excessive vehicle roll, thereby significantly enhancing the driving safety of high-center-of-gravity vehicles such as SUVs.

[0026] 4. The control algorithm architecture is simplified, and the system response speed is improved. The control method of the application eliminates the traditional complex upper arbitration module and fixed priority allocation rules, and the overall architecture is more flat. The self-organizing cooperative control law based on potential field gradient contribution degree has a parallel and efficient calculation process. This not only simplifies the design and calibration work of the control algorithm, but also significantly reduces the computational delay in the decision-making process, making the response of the entire closed-loop system more rapid and real-time.

[0027] 5、Improved adaptability of the system to complex working conditions. By introducing dynamic risk factors quantifying the driver's intention and the level of environmental danger, the application enables the intervention timing and intensity of the control system to match the current actual driving scenario in real time. Whether in regular cruising, emergency obstacle avoidance or driving on low adhesion road surface, the system can adaptively adjust the "sensitivity" of its control strategy, maximally preserving the driver's control intention under the premise of safety, and showing excellent environmental and working condition adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0028] Fig. 1 is a functional block diagram of the control system of the application;

[0029] Fig. 2 is a flowchart of the method of the application;

[0030] Fig. 3 is a schematic diagram of the potential field gradient contribution calculation principle of the application. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the application clearer, the application will be further described in detail below with reference to the drawings and specific examples. It should be noted that the specific examples here are only used to explain the application, and do not constitute any form of limitation on the protection scope of the application.

[0032] Reference Figs. 1 to 3 The application provides an active steering and braking integrated control system, which realizes the prospective and self-organizing collaborative control of active steering actuators and multi-channel braking actuators by taking the roll stability of the vehicle as the core control target through an innovative control architecture.

[0033] The integrated control system mainly consists of an integrated control unit, a sensor system, an actuator system and a vehicle-mounted communication bus at the hardware level.

[0034] The integrated control unit is the core of the operation and decision of the control system, and its physical carrier is a vehicle-mounted high-performance microprocessor, which internally integrates random access memory (RAM) and non-volatile memory (such as flash memory). The non-volatile memory has computer program instructions for implementing the control method of the application solidified therein. When the integrated control unit is powered on and runs, the processor will load and execute these instructions, thereby instantiating a series of software function modules that work collaboratively. These software function modules include: information acquisition and processing module, dynamic risk assessment module, virtual potential field construction module, gradient contribution calculation module and collaborative control instruction generation module.

[0035] The sensor system is responsible for providing all necessary information for the integrated control unit to make control decisions. The system includes:

[0036] an inertial measurement unit for measuring the roll angle φ, roll rate and yaw rate γ of the vehicle in real time;

[0037] wheel speed sensors distributed on each wheel for collecting the rotational speed of each wheel, whose data are used by the information acquisition and processing module to calculate the longitudinal vehicle speed v and estimate the side slip angle β of the center of mass;

[0038] a steering wheel angle sensor for detecting the steering wheel angle δ sw and its angular velocity to analyze the driver's path tracking intention and operation urgency;

[0039] brake system sensors for monitoring the brake master cylinder pressure or pedal displacement as a basis for judging the driver's braking intention;

[0040] and a forward-looking perception unit, which can be composed of millimeter wave radar and / or camera, for obtaining the geometric information of the road ahead, such as road curvature ρ curve and estimating the road adhesion condition μ est .

[0041] The actuator system is a physical unit that executes the control instructions and directly acts on the motion state of the vehicle. This system includes: an active steering system, which contains a steering motor superimposed on the steering transmission mechanism in its structure, and the motor can provide an additional front wheel steering angle δ afs outside the driver's input angle according to the instructions from the integrated control unit; an electronic brake system, which has the ability to independently and quickly adjust the braking force of the four wheels, such as electronic hydraulic brake system (EHB), which can accurately generate a target braking force F x,ij on any wheel (where ij represents the front left, front right, rear left, and rear right wheels).

[0042] The vehicle-mounted communication bus, such as controller area network (CAN-FD) or vehicle Ethernet, is responsible for establishing a high-speed and reliable data communication link between the integrated control unit, the sensor system and the actuator system. All the information collected by the sensor system is sent to the information acquisition and processing module in the integrated control unit through the bus. At the same time, the control instructions finally generated by the cooperative control instruction generation module are also issued to the active steering system and the electronic brake system through the bus, thus forming a complete closed-loop control loop. The cooperative work of the whole system aims to maintain the motion state of the vehicle in a dynamically optimized stable domain at all times.

[0043] In a specific embodiment, the control method performed by the integrated control unit has the following detailed process. The process integrates information perception, risk assessment, dynamic decision-making and collaborative control into one, forming a continuous, high-frequency closed-loop system.

[0044] The first step of the method is performed by the information acquisition and processing module. The core task of this module is to integrate the heterogeneous data from the sensor system and construct a real-time state vector that can comprehensively represent the current stable state of the vehicle. Specifically, this module directly obtains accurate measurement values of roll angle φ, roll angular velocity and yaw rate γ from the inertial measurement unit. At the same time, it processes signals from the wheel speed sensor to calculate a reliable vehicle longitudinal speed v through weighted averaging algorithm. For the mass side slip angle β which is not easy to measure directly, the module fuses roll angle, yaw rate, vehicle speed, and driver steering input δ sw provided by the steering wheel angle sensor based on a pre-set vehicle dynamics observer model to estimate it in real time. Finally, these key state variables are combined into a unified four-dimensional state vector as the basis for all subsequent calculations.

[0045] The second step of the method is completed by the dynamic risk assessment module and the virtual potential field construction module. The goal of this step is to construct a virtual potential field V that can sensitively reflect the instantaneous instability risk of the vehicle, and make its "terrain" dynamically changeable according to the driving scene. First, the dynamic risk assessment module is responsible for quantifying the driver's immediate operation and the potential danger of the external environment into two key dynamic risk factors. The driver's intention urgency factor λ int is calculated by monitoring the absolute value of the steering wheel angular velocity , and its specific function form can be:

[0046]

[0047] where k int is the gain coefficient, τ int is the trigger threshold value representing the boundary between regular driving and emergency operation. This formula makes the value of λ int jump quickly from the baseline state close to zero to the high position close to one when the driver performs sharp steering operations such as emergency obstacle avoidance.

[0048] At the same time, based on the information provided by the forward-looking perception unit, the module also calculates the environmental potential risk factor λ env . This factor integrates the influences of road curvature ρ curve and road adhesion coefficient estimate μ est , reflecting the objective challenge of the driving environment to the vehicle stability.

[0049] Subsequently, the virtual potential field construction module uses the state vector X and the above-mentioned dynamic risk factors to construct the vehicle stability virtual potential field V. The potential field V is defined as follows:

[0050]

[0051] In this formula, w φ ,w φ ,w β ,w γ is the dynamic weight coefficient corresponding to each state variable. d is the desired yaw rate, which depends on the driver's steering input δ sw and vehicle speed v are calculated through a steady-state linear two-degree-of-freedom model, representing the driver's normal tracking intention. These weight coefficients are the core of the present invention to achieve dynamic adaptive control, and their values ​​are determined by the dynamic risk factor λ int and λ env For example, the weight w is directly related to the roll stability φ and Its value will change with the driver's intention urgency factor λ int The weight w related to the sideslip stability increases exponentially or in a high-order polynomial form. β , whose value will change with the environmental potential risk factor λ env Increased (e.g., μ est This design makes the topology of the potential field V no longer static, but a dynamic risk "topography" that can be reshaped in real time, forming an extremely sensitive potential energy barrier at the budding stage of impending danger.

[0052] The third step of the method is performed by the gradient contribution calculation module, which is to distribute

[0053] Parallel evaluation of each actuator's instantaneous ability to reduce system risk. This module first calculates the gradient vector of the potential field V at the current state point X This vector points to the direction where the risk increases fastest.

[0054] Then, for each independent actuator, such as any wheel brake unit in the active steering system or electronic braking system, the module calculates its potential field gradient descent contribution G i The calculation formula is:

[0055]

[0056] Among them, b i is the influence vector of the i-th actuator, which describes the effect of the actuator on each component of the state vector X when applying a unit of control input (such as one degree of additional angle or one Newton of braking force). The rate of change that can be generated. This vector b i Is pre-calibrated or real-time identified according to the multi-body dynamics model of the vehicle. G i The calculation result of is a scalar, the positive value of which directly represents the efficiency and effectiveness of the actuator in suppressing the instability trend at the current moment.

[0057] The fourth step of the method is executed by the cooperative control instruction generation module, which generates the final cooperative control instruction in a self-organizing manner based on the contribution degree G i Of each actuator. The module first sums up all positive contribution degrees to obtain the total contribution degree G total =∑ j max(0,G j ). Then, for the i-th actuator, the final control instruction quantity u i Is generated by the following control law:

[0058]

[0059] In this control law, K gain Is a global control gain used to adjust the overall response sensitivity of the system; multiplied by the potential field value V, it ensures that the overall control intensity matches the actual risk level of the vehicle; the core fractional term dynamically and real-time allocates the total control task according to the contribution degree of each actuator, and the actuator with greater contribution naturally obtains greater control weight; ∈ is a very small normal number used to prevent the denominator from being zero. The final generated logical control instruction u i (Such as the target additional steering angle or target braking force) will be converted into a bottom-level electrical signal and issued to the corresponding actuator through the vehicle-mounted communication bus, thereby completing a complete and efficient cooperative closed-loop control.

[0060] In summary, the control method and system disclosed in the embodiments of the present application are characterized by constructing a new control paradigm. The method no longer follows the traditional error tracking-based control idea, but cleverly abstracts and transforms the complex vehicle multi-actuator coordination and stability problem into a physical process of finding and maintaining the lowest potential energy point in a dynamic virtual potential field. In this way, the instability risk of the vehicle is intuitively quantified as potential energy, and the control target is clearly defined as the steepest descent direction of the system state point along the potential energy gradient.

[0061] In the technical solutions of the application, by introducing a dynamic risk factor coupled with the driver's intention and the external environment in real time, the virtual potential field itself is endowed with the ability of dynamic self-adaptive remodeling. This makes the core judgment basis of the control system, the risk "terrain map", be able to adjust its sensitivity and constraint strength in advance according to the upcoming danger, thereby realizing the fundamental change from "passive response" to "active prevention". Further, based on the distributed calculation and self-organizing cooperative control law of the potential field gradient descent contribution, the complex central arbitration and command distribution logic is completely abandoned. Each actuator takes reducing the global system potential energy as the only and unified action criterion, and the allocation of its control authority and responsibility is dynamically and optimally achieved in each control cycle, thereby eliminating the potential action conflict among the actuators and achieving an inherent and efficient control cooperation.

[0062] It should be understood that the above description is only a specific embodiment of the present application, which is intended to help those skilled in the art to more fully understand the technical essence of the present application, and is not intended to constitute any improper limitation on the protection scope of the present application. For those skilled in the art, any simple modification, equivalent replacement or improvement of the above embodiments without departing from the spirit and principles disclosed by the present application, such as changing the selection of specific components, adjusting the non-core calibration parameters in the formula, or combining or splitting the software function modules in different ways, should be considered to fall within the scope of the present application.

[0063] Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made thereto without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. Active steering and braking integrated control method based on roll angle feedback, characterized in that: The steps include: Step 1: Obtain multiple vehicle state variables including vehicle roll angle; Step 2: Based on the vehicle state variables and the dynamic risk factors reflecting the driver's intention or the external environment, a vehicle stability virtual potential field is constructed. The dynamic risk factors are used to dynamically adjust the weight coefficients corresponding to each vehicle state variable in the vehicle stability virtual potential field. Step 3: Calculate the contribution of the active steering system and the braking system to reducing the potential field gradient of the vehicle stability virtual potential field; Step 4: Generate and execute coordinated control instructions for the active steering system and braking system based on the potential field gradient descent contribution.

2. The method according to claim 1, characterized in that The dynamic risk factors include: a driver intention urgency factor quantified according to a steering wheel angular velocity or a brake pedal operation, and / or an environmental potential risk factor quantified according to a road curvature or a road adhesion coefficient estimation value.

3. The method according to claim 1, characterized in that The step of constructing the vehicle stable virtual potential field is specifically as follows: calculating the potential field value V by the following formula: Wherein, V is the potential field value of the stable virtual potential field of the vehicle; φ is the roll angle, is the roll angular velocity, β is the center of mass slip angle, γ is the yaw rate; w φ , w β ,w γ is the corresponding weight coefficient; γ d is the desired yaw rate calculated based on the driver's steering wheel input.

4. The method according to claim 3, characterized in that The weight coefficient w φ , w β ,w γ The value of is a function of the dynamic risk factor. When the risk represented by the dynamic risk factor increases, the corresponding weight coefficient value increases nonlinearly.

5. The method according to claim 1, wherein The calculated potential field gradient descent contribution The steps of the degree include: First, calculate the gradient vector of the vehicle stable virtual potential field under the current vehicle state variable Then, for the i-th actuator, the formula Calculate its potential field gradient descent contribution G i ; Among them, b i is the influence vector of the unit control input of the i-th actuator on the vehicle state variables.

6. The method according to claim 1 or 5, characterized in that The step of generating the collaborative control instruction is specifically as follows: making the size of the control instruction of each actuator proportional to the proportion of the potential field gradient descent contribution of the actuator in all positive contributions.

7. The method according to claim 6, characterized in that The overall control strength of the cooperative control command is proportional to the current value of the vehicle stable virtual potential field.

8. The method according to claim 1, characterized in that The vehicle state variables also include: roll angular velocity, center of mass sideslip angle, and yaw rate.

9. The method according to claim 1, characterized in that The braking system is a braking system capable of independently adjusting the braking force of a single wheel.

10. An active steering and braking integrated control system based on roll angle feedback, characterized in that: The method comprises a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method according to any one of claims 1 to 9 is implemented.

Images