ADAS auxiliary driving steering control method

Abstract

The invention discloses an ADAS auxiliary driving steering control method. The method comprises the steps that S1, a vehicle dynamics model conforming to steering control is constructed; s2, constructing a state equation of stable turning of the vehicle; s3, solving and analyzing the constructed state equation to obtain a corner request; and S3, vehicle turning is controlled based on the turning angle request. The method has the advantages that the cost is low, no hardware cost exists, the method can be integrated in a controller of a vehicle, the system can pass a curve at different vehicle speeds, response is more timely and accurate, the turning capacity of the system is faster and smoother, and the system capacity is improved; and moreover, feedforward parameters and the requirement for a feedback controller in bending are reduced, and the calibration pressure of engineers is relieved.

Description

Technical Field

[0001] The present invention relates to the field of ADAS, and in particular to an ADAS assisted driving steering control method. Background Art

[0002] The ADAS assisted driving system can calculate a reasonable steering wheel angle request based on the road information and vehicle information detected by the current sensor, such as curvature, vehicle speed, vehicle posture and other signals, and control the vehicle to pass through the curve stably and comfortably. For example, the patent application number is 201911338790.3, a method, system and vehicle for selecting targets when the ADAS vehicle passes through a curve, which discloses: detecting the lane line in front of the vehicle to obtain lane line information; judging whether the lane line information meets the lane line judgment condition; if it meets, calculating the curvature of the lane line in front of the vehicle in real time; if the lane line curvature exceeds the curvature threshold, it is judged that the vehicle is about to enter the curve; judging whether the historical driving trajectory of the vehicle is similar to the historical lane line information; if it is similar, selecting the vehicle closest to the vehicle in the lane line as the target vehicle. By detecting the lane line in front of the vehicle, it is judged whether the historical driving trajectory of the vehicle is similar to the historical lane line information; if it is similar, selecting the vehicle closest to the vehicle in the lane line as the target vehicle. The ADAS auxiliary system can reduce the probability of incorrectly selecting the target vehicle when the vehicle passes through a curve, and improve the safety of the vehicle during driving.

[0003] In the entire curve control process of ADAS assisted driving, feedforward parameters play an important role, and their quality will determine the speed and smoothness of the system's curve control.

[0004] The prior art simply refers to the curvature of the road, the vehicle speed and other information to adjust the feedforward parameters of the lane keeping function when cornering. Feedback control is added to the feedforward to keep the vehicle cornering. During ADAS development, the data needs to be calibrated to form a corresponding MAP. However, during calibration, there are many feedforward parameters and high feedback requirements. More parameters will cause the system to have higher requirements on controller performance. At the same time, more parameters will also cause calibration pressure on engineers during calibration, making the development of the ADAS system's assisted cornering function more complicated. Moreover, the MAP formed by the unchanging calibration cannot meet the requirements of driving safety. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an ADAS assisted driving steering control method. Starting from vehicle dynamics analysis, the kinematic model and dynamic parameters of the vehicle are fully utilized to automatically calculate and obtain the control angle and perform control. Compared with the prior art, the feedforward controller adjustment parameters are reduced, the system's requirements for the feedback controller are lowered, and the system's cornering ability is improved.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: an ADAS assisted driving steering control method, comprising:

[0007] S1. Construct a vehicle dynamics model that meets the steering control requirements;

[0008] S2, constructing the state equation of the vehicle in stable cornering;

[0009] S3, solving and analyzing the constructed state equation to obtain a rotation angle request;

[0010] S3. Controlling the vehicle to turn based on the turning angle request.

[0011] S1 constructs a vehicle dynamics model based on the vehicle lateral dynamics, where the model is expressed as x is the state space variable, y is the lateral error, Ψ is the heading error, A is the system matrix; B is the control matrix; δ is the front wheel steering angle.

[0012] The expected heading rate of the vehicle in stable cornering in S2 is Ψ des , the output equation can be described as δ = -Kx, and the state equation for constructing a stable vehicle cornering can be described as

[0013] The expected stability of the vehicle in turning in step S2 des Using the theoretical formula:

[0014]

[0015] in The yawrate corresponding to stable cornering, R road radius, v x is the vehicle longitudinal velocity.

[0016] The feedforward term δ of the front wheel angle is introduced into the state equation in step S2 ff To compensate for the road curvature, let δ = -Kx + δ ff , introduced into the state equation in step S2 to obtain the corrected state description equation.

[0017] The feedforward term of the front wheel angle is solved using the following formula:

[0018]

[0019] Where R is the road radius, l f Distance from vehicle center of gravity to front axle, l r The distance from the center of gravity of the vehicle to the rear axle, m is the vehicle mass, vx is the longitudinal velocity of the vehicle, C αr Wheel cornering stiffness.

[0020] The feedforward term δ of the front wheel steering angle is introduced ff The corrected state description equation is:

[0021]

[0022] The state description equation is solved by the ADAS controller, with the front wheel steering angle δ as input and the model state as output. The lateral error and yaw error are equal to 0 and the steering angle request for controlling the vehicle is obtained.

[0023] When ADAS controls the vehicle, it controls the preview distance of the lane line information and dynamically adjusts the preview distance according to the vehicle speed and the rate of change of lane curvature.

[0024] The advantages of the present invention are: low cost, no hardware cost, and the method can be integrated into the vehicle's controller, so that the system can turn at different vehicle speeds, respond more promptly and accurately, make the system's cornering ability faster and smoother, and improve the system's capabilities; and reduce the feedforward parameters and the requirements for the feedback controller in the corner, alleviating the calibration pressure on engineers. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The following is a brief description of the contents expressed in the drawings of the present invention and the symbols in the drawings:

[0026] Figure 1 It is a schematic diagram of the control principle of the present invention. DETAILED DESCRIPTION

[0027] The specific implementation of the present invention will be further explained in detail below by describing the optimal embodiment with reference to the accompanying drawings.

[0028] The present invention starts from the vehicle dynamics analysis, makes full use of the vehicle's kinematic model and the vehicle's dynamic parameters, reduces the feedforward controller adjustment parameters, reduces the system's requirements for the feedback controller, reduces the engineers' calibration pressure, and improves the system's cornering ability.

[0029] The technical solution of the present invention is:

[0030] 1) Construct a vehicle dynamics model based on the vehicle lateral dynamics x is the state space variable, y is the lateral error, Ψ is the heading error; A is the system matrix; B is the control matrix; δ is the front wheel steering angle;

[0031] 2) The expected heading rate for a stable vehicle cornering is Ψ des, which is the theoretical stable cornering lateral angular velocity, the output equation can be described as δ = -Kx, K is the output matrix; the state equation for constructing a stable cornering vehicle is described as δ=-Kx;

[0032] Where A is the system matrix; B is the control matrix; B1 is the direct transmission matrix; K is the output matrix, Ψ des The yaw rate of the vehicle in a stable cornering process is the theoretical stable cornering yaw rate. Stable cornering means that the yaw rate of the vehicle in a lane is equal to the theoretical value when the vehicle passes a fixed radius curve at a certain speed. The yaw rate of the stable cornering theory can be obtained through the theoretical formula. The yawrate corresponding to stable cornering, R road radius, v x is the longitudinal speed of the vehicle. The control goal is to control the yawrate of the vehicle to be equal to the yawrate of stable cornering, so as to achieve fast and stable cornering of the vehicle.

[0033] 3) Solving and analyzing the constructed system description equation, the turning angle request required for the vehicle to turn stably can be obtained;

[0034] Step 2) A closed-loop control model of the front wheel steering angle of the vehicle is established. The purpose of the control is to give the front wheel steering angle input δ so that the model state output The lateral error and yaw error can be equal to 0, but the control output δ = -Kx of this model actually only considers the state feedback control. In actual control, when encountering a curve, the control will be slow and not smooth and stable if it relies solely on feedback control. In fact, the control is feedback plus feedforward control. Here, the feedforward input term δ is introduced. ff To compensate for the road curvature and allow the controller to control the vehicle to turn better, we can get δ=-Kx+δ ff Then the state equation of the second step can be updated The main purpose of introducing feedforward is to make the vehicle better at cornering. The main purpose of the vehicle's stable cornering is to make the vehicle's yaw rate equal to the theoretical yaw rate. Therefore, the yaw rate error term of state x in the state equation is 0, which is the target of the controller. In this way, the theoretical feedforward angle requirement can be obtained by solving the equation. RRoad radius, l f Distance from vehicle center of gravity to front axle, l r The distance from the center of gravity of the vehicle to the rear axle, m is the vehicle mass, vx is the longitudinal velocity of the vehicle, C αr Wheel cornering stiffness.

[0035] The feedforward term δ ff =f(L, Curv, a y, k), where the coefficient k is related to the vehicle's own dynamic parameters, and the change of this parameter is added to the feedforward adjustment factor. In this way, for the feedforward controller, it is only necessary to adjust the adjustment factor under different working conditions;

[0036] 4) In actual vehicle control, due to the delay of actuators and sensors, if the corresponding feedforward input is executed completely according to the input curvature, the control will often be late and fail to achieve a good control effect. In order to allow the controller to respond more quickly to the curve to be passed, the preview distance of the lane line information will be adjusted according to the vehicle speed and the lane curvature change rate, so as to obtain the preview lane curvature information, predict the road curvature change earlier, and thus control the vehicle earlier. The curvature is equal to 1 / R, which represents the road radius; through different vehicle speeds and curvature changes, the predicted time can be obtained by a simple table lookup, and the time multiplied by the speed is the predicted distance.

[0037] 5) The feedforward adjustment factor is mainly used to correct the changes in the vehicle body dynamic parameters under different working conditions. It can be expressed as a nonlinear function f(v,ay...) with information such as vehicle speed. Due to its nonlinearity and complex functions, it is not suitable for direct engineering use. Here, the main input parameters of the function, such as vehicle speed and lateral acceleration, are used to construct a numerical description of this equation. Moreover, since the range of change of this function under normal working conditions is not particularly large, the size of the numerical description model can also be greatly reduced. As for the lost part of the description, it is repaired through feedback control. For example, one of the items in step 3 is the wheel cornering stiffness. This coefficient is closely related to the vehicle speed and other states, so a feedforward adjustment factor will be introduced. By setting a simple speed curvature map table, the default value is 1. However, due to the change of wheel cornering stiffness, the theoretically calculated feedforward value at different vehicle speeds may be inappropriate. At this time, it is necessary to dynamically adjust the adjustment factor to adapt to the specific vehicle under different working conditions. The feedforward term refers to the vehicle model, which means the feedforward part of the model; the feedforward parameter is the specific parameter in the feedforward term; the feedforward adjustment factor is an adjustment coefficient to adapt the vehicle under different working conditions.

[0038] 6) Although there are certain errors in the numerical description model, the establishment of the previous model fully considers the vehicle dynamics model and the description of the main working range, so it can greatly reduce the feedforward parameters and alleviate the pressure on the feedback controller.

[0039] Obviously, the specific implementation of the present invention is not limited to the above-mentioned methods. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, they are all within the protection scope of the present invention.

Claims

1. An ADAS assisted driving steering control method, characterized in that: include S1. Construct a vehicle dynamics model that meets the steering control requirements; S2, constructing the state equation of the vehicle in stable cornering; S3, solving and analyzing the constructed state equation to obtain a rotation angle request; S3. Controlling the vehicle to turn based on the turning angle request.

2. An ADAS assisted driving steering control method as claimed in claim 1, characterized in that: S1 constructs a vehicle dynamics model based on the vehicle lateral dynamics, where the model is expressed as x is the state space variable, y is the lateral error, Ψ is the heading error, A is the system matrix; B is the control matrix; δ is the front wheel steering angle.

3. An ADAS assisted driving steering control method as claimed in claim 2, characterized in that: The expected heading rate of the vehicle in stable cornering in S2 is Ψ des , the output equation can be described as δ = -Kx, and the state equation for constructing a stable vehicle cornering can be described as 4. An ADAS assisted driving steering control method as claimed in claim 3, characterized in that: The expected stability of the vehicle in turning in step S2 des Using the theoretical formula: in Yawrate corresponding to stable cornering, R road radius, v x is the vehicle longitudinal velocity.

5. An ADAS assisted driving steering control method as claimed in claim 4, characterized in that: The feedforward term δ of the front wheel angle is introduced into the state equation in step S2 ff To compensate for the road curvature, let δ = -Kx + δ ff , introduced into the state equation in step S2 to obtain the corrected state description equation.

6. An ADAS assisted driving steering control method as claimed in claim 5, characterized in that: The feedforward term of the front wheel angle is solved using the following formula: Where R is the road radius, l f Distance from vehicle center of gravity to front axle, l r The distance from the center of gravity of the vehicle to the rear axle, m is the vehicle mass, vx is the longitudinal velocity of the vehicle, C αr Wheel cornering stiffness.

7. An ADAS assisted driving steering control method as claimed in claim 5, characterized in that: Introducing the feedforward term δ of the front wheel steering angle ff The corrected state description equation is:

8. An ADAS assisted driving steering control method according to any one of claims 1 to 7, characterized in that: The state description equation is solved by the ADAS controller, with the front wheel steering angle δ as input and the model state as output. The lateral error and yaw error are equal to 0 and the steering angle request for controlling the vehicle is obtained.

9. An ADAS assisted driving steering control method according to any one of claims 1 to 7, characterized in that: When ADAS controls the vehicle, it controls the preview distance of the lane line information and dynamically adjusts the preview distance according to the vehicle speed and the rate of change of lane curvature.

Images