CN110096748B - A Modeling Method of Human-Vehicle-Road Model Based on Vehicle Kinematics Model - Google Patents

Abstract

The invention relates to a human-vehicle-road model modeling method based on a vehicle kinematics model. The established vehicle kinematics model considers the action of a human driver in driving, and represents the human driver with different driving behaviors by using parameters such as neuromuscular delay time, preview time, steering gain and the like in the steering characteristic of the driver, so that a vehicle controller designed based on the model can drive like a person, and the riding comfort of passengers is improved. The human-vehicle-road model based on the vehicle kinematics model provided by the invention reduces the design difficulty of the controller, has good real-time performance, considers the riding comfort of human drivers, has practicability and also has wide commercial application prospect.

Description

A Human-Vehicle-Road Model Modeling Method Based on Vehicle Kinematics Model

technical field

The invention relates to path planning and path tracking, and belongs to the technical field of intelligent transportation, in particular to a human-vehicle-road model modeling method based on a vehicle kinematics model.

Background technique

"Intelligence, networking, electrification, and sharing" is the current development trend of the automobile industry. Autonomous vehicles and semi-autonomous vehicles represented by intelligence are currently the research hotspots of many technology companies and automobile manufacturers at home and abroad. Intelligent driving vehicles must be able to exchange and share intelligent information with X (vehicles, roads, people, etc.), and have functions such as complex environment perception and intelligent decision-making collaborative control. This improves driving safety and efficiency. However, the fully automated unmanned driving technology that liberates the driver from boring driving is not achieved overnight, but a gradual process. Compared with automatic driving, semi-autonomous driving is currently more likely to be realized. At present, most of the common vehicle models are vehicle dynamic models. The dynamic model has high requirements on vehicle parameters, many of which have nonlinear characteristics, such as side slip angle, steering stiffness, etc., which will increase the calculation burden of the controller, and the complex Calculation will reduce the real-time performance of vehicle motion control, so in practical applications, the model based on vehicle dynamics has defects.

Contents of the invention

In order to solve the above problems, the present invention discloses a human-vehicle-road model modeling method based on the vehicle kinematics model, which reduces the computational burden of the controller through vehicle kinematics modeling, and at the same time utilizes the delay time in the driver's steering characteristics , preview time, steering gain and other parameters to take into account the driver's driving characteristics, so that the vehicle controller can drive like a human and improve the ride comfort of the occupants.

To achieve the above object, the technical scheme of the present invention is as follows:

A human-vehicle-road model modeling method based on a vehicle kinematics model, which uses a vehicle-road kinematics model to replace the current more commonly used vehicle dynamics model. As shown in Figure 1, when ψ is a small angle, the trigonometric function value of ψ can be approximated, that is, cosψ=1, sinψ=ψ, tanψ=ψ, and the vehicle-road kinematics model can be expressed as

Further, l _f and l _r in Fig. 1 are the distances from the front and rear axles of the vehicle to the center of mass of the vehicle, V _x represents the longitudinal velocity of the vehicle, y is the current lateral position of the center of mass of the vehicle, and y _des represents the center of mass of the vehicle at The target position corresponding to the current position on the reference trajectory, e _y =yy _des represents the error between the current position of the center of mass of the vehicle and the reference trajectory. δ _f is the steering angle of the front wheels of the car, which is used as the output of the driver model to control the vehicle to perform steering operations, ψ represents the heading angle of the current position of the vehicle, and ψ _des is the tangent formed by the corresponding position of the center of mass of the vehicle on the reference path and the X axis The cutting angle of , e _ψ =ψ-ψ _des represents the heading deviation between the vehicle and the corresponding position on the reference path.

Assuming that the vehicle longitudinal velocity _Vx is constant and the front wheel steering angle _δf is small, then the lateral movement of the vehicle can be approximately expressed as:

Furthermore, in order to consider the driver's steering characteristics when designing the vehicle controller, the model introduces the driver model into the vehicle-road model.

Among them, T _d , T _p , and G _h represent the driver's delay time, preview time, and steering proportional gain, respectively. a ₀ is a constant, R _g is the transmission ratio of the steering system, that is, the ratio of the steering wheel angle to the front wheel angle of the car. In the model, when the driver applies steering control behavior to the vehicle, the delay time is divided into two parts, that is, T _d =τ _d1 +τ _d2 , where τ _d1 is a reaction that the driver realizes that he needs to turn to perform the steering action. At the same time, when the driver performs the steering action, there will be a neuromuscular delay τ _d2 on the arm; the driver always outputs a signal proportional to the input signal with the minimum workload, that is, the steering proportional gain, Y _p represents the lateral position of the preview point on the reference path. The preview point is the intersection point of the vertical line at L in front of the current position of the vehicle and the reference path. T _p represents the driver’s preview time. When the vehicle’s longitudinal speed is V _x When it remains unchanged, the driver’s forward-looking distance is L=V _x *T _p , Y _p represents the lateral position of the preview point on the reference path corresponding to L in front of the vehicle, and y+T _p V _x ψ represents the driver’s The vehicle position after the preview time of the current vehicle motion state estimation.

Further, in order to facilitate modeling in Matlab/Simulink, the above continuous model is transformed into a discrete time state model. According to the human-vehicle-road model established above, the discrete-time state variables of the driver-vehicle-road system model are defined as

x(k)=[x ₁ (k), x ₂ (k), x ₃ (k), x ₄ (k), x ₅ (k), x ₆ (k), x ₇ (k)] ^T , in

所述人-车-路模型的状态空间模型可以被描述为The input of the pedestrian-vehicle road model is

The state space model of the human-vehicle-road model can be described as

x(k+1)=Ax(k)+B _u u(k)

in

Furthermore, in the human-vehicle-road model proposed by the present invention, the driver's steering characteristics are taken into account, and when the human-vehicle-road model is used for path planning and path tracking, the driver's manipulation behavior and preferences can be considered. On the other hand, due to the adoption of a kinematic vehicle model, the proposed human-vehicle-road model is linear in nature, and the use of this model in complex path planning or path tracking can reduce computational costs.

The beneficial effects of the present invention are:

Since the vehicle kinematics model is used to design the human-vehicle-road model, the computational burden of the controller is low when the lateral motion control is implemented in the model, and the real-time performance of the vehicle motion control is greatly improved. The invention considers the driver's preview function, neuromuscular delay, steering gain and other driving characteristics, so that the designed vehicle controller can drive like a human, thereby improving the ride comfort of the passengers.

Description of drawings

Fig. 1 is a schematic diagram of the vehicle-road kinematics model considering the forward-looking effect of the driver in the present invention.

Fig. 2 is a schematic block diagram of the human-vehicle-road model based on the kinematic vehicle model of the present invention.

Detailed ways

The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

The present invention proposes a human-vehicle-road model based on the vehicle kinematics model, which is used to reduce the calculation burden of the vehicle controller, improve the real-time performance of the controller, enable the controller to drive like a human, and improve the ride comfort of the occupants.

The present invention adopts two coordinate systems, that is, a global coordinate system and a vehicle body coordinate system. When ψ is a small angle, the trigonometric function value of ψ can be approximated, that is, cosψ=1, sinψ=ψ, tanψ=ψ, according to the analysis of vehicle kinematics, the vehicle-road kinematics model can be expressed as .

In the formula, l _f and l _r are the distances from the front axle and rear axle of the vehicle to the center of mass of the vehicle, V _x represents the longitudinal velocity of the vehicle, y is the current lateral position of the center of mass of the vehicle, and y _des represents the distance between the center of mass of the vehicle on the reference trajectory and the current The target position corresponding to the position, e _y =yy _des represents the error between the current position of the center of mass of the vehicle and the reference trajectory. δ _f is the steering angle of the front wheels of the car, ψ represents the heading angle of the current position of the vehicle, ψ _des is the tangent angle formed by the tangent of the corresponding position of the center of mass of the vehicle on the reference path and the X axis, e _ψ = ψ-ψ _des represents the tangent between the vehicle and the reference path The heading deviation for the corresponding position on the path.

Assuming that the vehicle’s longitudinal velocity _Vx is constant and the front wheel steering angle _δf is small, then the relational expression of the vehicle’s yaw motion can be approximately expressed as:

The present invention is in order to consider the driver's steering characteristic in the recovery of design vehicle controller, introduces driver's model in vehicle-road model, and driver's model is expressed as:

where T _d , T _p , and G _h represent the driver's delay time, preview time, and steering proportional gain, respectively.

is a constant, and R _g is the transmission ratio of the steering system, that is, the ratio of the steering wheel angle to the front wheel angle of the vehicle. When the driver applies the steering control behavior to the vehicle, the delay time is divided into two parts, that is, T _d =τ _d1 +τ _d2 , where τ _d1 is a reaction delay from when the driver realizes the need to turn to perform the steering action, and at the same time , when the driver performs the steering action, there will be a neuromuscular delay τ _d2 on the arm; the driver always outputs a signal proportional to the input signal with the minimum workload, that is, the steering proportional gain, T _p represents the driving The driver’s preview time, when the vehicle’s longitudinal speed V _x is constant, the driver’s forward-looking distance is L=V _x *T _p , and Y _p represents the lateral position of the preview point on the corresponding reference path at L in front of the vehicle , y+T _p V _x ψ represents the vehicle position after the preview time estimated by the driver based on the current vehicle motion state.

The present invention defines discrete-time state variables as x(k)=[x ₁ (k), x ₂ (k), x ₃ (k), x ₄ (k), x ₅ (k), x ₆ (k) , x ₇ (k)] ^T ,

则人车路模型的状态空间模型可以被描述为The input of the pedestrian-vehicle road model is

Then the state space model of the pedestrian-vehicle-road model can be described as

x(k+1)=Ax(k)+B _u u(k)

In the formula

The description of other symbols is the same as before.

Claims (1)

1. A human-vehicle-road model modeling method based on a vehicle kinematic model is characterized in that: the method comprises the following steps:

(a) Building a vehicle-road model based on a vehicle kinematics model

In the formula I _f And l _r Respectively the distance from the front and rear axle of the vehicle to the center of mass of the vehicle, V _x Representing the longitudinal speed of the vehicle, e _y ＝y-y _des Representing the error between the current position of the vehicle's center of mass and the reference trajectory, y being the current lateral position of the vehicle's center of mass, y _des Representing a target position of a center of mass of the vehicle on a reference trajectory corresponding to the current position; psi denotes the heading angle of the current position of the vehicle, psi _des Is the tangent angle formed by the tangent of the corresponding position of the center of mass of the vehicle on the reference path and the X axis, e _ψ Indicating a heading deviation of the vehicle from a corresponding position on the reference path; delta _f Is the steering angle of the front wheels of the automobile,

is the front wheel steering angle acceleration;

(b) In order to enable the vehicle controller to be driven like a person, the model design takes into account the driver model

Wherein, T _d 、T _p 、G _h Respectively representing the delay time, the preview time and the steering proportion gain of a driver; a is a ₀ Is a constant number, R _g Is the transmission ratio of the steering system, i.e. the ratio of the steering wheel angle to the front wheel angle of the vehicle, G _h Is a steering proportional gain; y is _p Representing the transverse position of the pre-aiming point on the corresponding reference path at the front L of the vehicle, y + T _p V _x ψ denotes the vehicle position after the estimated preview time by the driver from the current vehicle motion state.

Images