KR20180138324A - Apparatus and method for lane Keeping control - Google Patents

Abstract

The present invention relates to an apparatus and a method for lane keeping control, capable of controlling the lateral motion of a vehicle by using a dynamic lateral motion model and a kinematic lateral motion model. The lane keeping control device includes: a creation unit for creating a dynamic lateral motion model (hereinafter referred to as a dynamic model) and a kinematic lateral motion model (hereinafter referred to as a kinematic model) in consideration of a forward viewing distance of a vehicle; a measurement unit for measuring information on a vehicle lane and a vehicle motion using at least one of a camera sensor and an inertial measurement sensor; and a control unit which calculates the probability representing the adaptability according to the driving condition for each of the dynamic model and the kinematic model using the information of the vehicle lane and the vehicle motion, calculates steering torque of the vehicle by using an interaction multi-model filter (IMM filter) that applies the weight according to the calculated probability to each model and a fusion model to which the weight is applied, and performs lateral motion control for allowing the vehicle to maintain the vehicle lane by using the calculated steering torque.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001]

The present invention relates to a lane keeping control device and method.

The main difference between the dynamic lateral kinematic model and the kinematic lateral motion model is whether the tire slip angle is considered. The tire slip angle is defined as the direction of the tire direction and the velocity vector of the wheel. The dynamic transverse motion model includes the tire slip angle, while the kinematic transverse motion model does not include the tire slip angle. The tire slip angle occurs when the vehicle is traveling at high speed with a large steer angle or a sudden change in yaw rate. Thus, in the automotive industry, dynamic lateral motion models have been widely used for lane keeping systems (LKS), lane change control (LXC), and collision avoidance. However, the dynamic transverse motion model is highly dependent on some unknown vehicle parameters. Literature review of the dynamic lateral motion model has been focused on unknown vehicle parameters, such as cornering stiffness and tire / road friction conditions associated with tire slip angles.

On the other hand, kinematic lateral motion models have been used for limited low-speed driving conditions, such as parking assist systems. Recently, however, research has been conducted to show that a kinematic lateral motion model for a lane-keeping or lane-keeping system during highway driving can be robust against a parameter uncertainty and can be used for a lane keeping system.

Both dynamic and kinematic transverse vehicle motion control are essential to know exactly the vehicle condition. However, due to sensor cost and / or physical limitations, not all vehicle conditions can be taken. Thus, for vehicle manufacturers, it may be more appropriate to develop an on-line estimation method than to install additional sensors on available sensors in current luxury vehicles, such as inertial measurement sensors (IMU) and vision sensors.

There are many approaches to estimating the state of the lateral motion model. Most of these approaches are based on dynamic models, and performance varies with parameter uncertainty.

On the other hand, the Interaction Multiple Model (IMM) has been widely used in various industrial applications such as target tracking and target estimation / detection, because it requires low computational power and high performance. In particular, interaction multi-models have been developed in the aircraft and vehicle industries to accurately track multiple targets. Also, in the vehicle industry, interaction multi-models have been used in other autonomous vehicle applications, such as road boundary detection and GPS accuracy improvements. The interaction multiple model is based on the criterion of the model probability that represents the model reliability.

The present invention uses a dynamic lateral motion model and a kinematic lateral motion model of a vehicle for vehicle maintenance to determine a suitable model for driving conditions, And to provide a lane keeping control device and method for controlling the lateral motion of the vehicle.

According to an aspect of the present invention, a lane keeping controller is disclosed.

A lane keeping control device according to an embodiment of the present invention includes a dynamic lateral motion model and a kinematic lateral motion model that take into account the forward viewing distance of the vehicle, A camera sensor, and an inertial measurement sensor, and a controller for measuring the information of the vehicle and the motion of the vehicle using the at least one of the camera sensor and the inertial measurement sensor, An interaction multiple model filter (IMM filter) for calculating a probability indicating an application fitness according to a driving condition for each of the kinematic models and applying the weight according to the calculated probability to each model, The steering torque of the vehicle is calculated using the fusion model to which the weight is applied, Using the exported steering torque and a controller for performing control lateral motion (lateral motion control) for maintaining drive of the vehicle.

The interaction multi-model filter performs four steps of interaction, filtering, updating, and combination.

The interaction multi-model filter calculates a maintenance and change probability indicating the fitness of each model at a previous sampling time in the interaction, and estimates an initial state and an initial covariance for each model using the calculated maintenance and change probability do.

The maintenance and change probability of the dynamic model and the kinematic model is calculated by the following equation.

는 모델 i에서 j로 전환되는 마르코프 확률(Markov probability)을 나타내고,

는 시간 k에서 모델 i의 모델 확률을 나타내고,

이고, 인덱스 숫자 i, j = 1은 상기 다이나믹 모델을 나타내고, 인덱스 숫자 i, j = 2는 키네마틱 모델을 나타냄here,

Denotes a Markov probability which is converted from model i to j,

Represents the model probability of model i at time k,

, Index number i, j = 1 represents the dynamic model, and index number i, j = 2 represents a kinematic model

Wherein the interaction multi-model filter is a filter for estimating a current sampling time for a previous sampling time using the coefficient of each model calculated from the lane information and the vehicle motion information measured at the current sampling time and the estimated initial state and the initial covariance And calculates a predicted value of the current state and the current covariance.

The interaction multi-model filter calculates the innovation and covariance of the Kalman filter of each model in the filtering, corrects the predicted value using the Kalman filter of each model, and outputs the current state and the current covariance .

Wherein the interaction multi-model filter calculates the likelihood function of each model according to the Gaussian assumption using the innovation and covariance of the Kalman filter of each model in the update, The current maintenance and change probability for each model is calculated.

The current maintenance and change probability is calculated by the following equation.

이고,

및

는 각각 칼만 필터의 이노베이션(innovation) 및 공분산이고, c는 정규화 계수임here,

ego,

And

Are the innovations and covariances of the Kalman filter, respectively, and c is the normalization coefficient

Wherein the interaction multi-model filter calculates the combined state and covariance of the dynamic model and the kinematic model using the current maintenance and change probability and the estimated current state and the current covariance, Apply to the model.

, 횡방향 속도 오차

, 차량 무게중심에서의 헤딩 각도 오차(heading angle error)

및 요레이트(yaw rate)

를 포함하는 상태 벡터를 포함한다.The system state functions of the dynamic model and the kinematic model include a lateral lane center offset at the look-ahead point,

, Lateral velocity error

Heading angle error at the vehicle center of gravity,

And a yaw rate

And a state vector.

및 상기

를 포함하고,The lane information

And

Lt; / RTI >

를 포함한다.The vehicle motion information

.

According to another aspect of the present invention, a lane keeping control method is disclosed.

The present invention provides a lane keeping control method for a vehicle, including a dynamic lateral motion model (hereinafter referred to as a dynamic model) and a kinematic lateral motion model (hereinafter referred to as a dynamic lateral motion model) (Hereinafter referred to as " kinematic model "), measuring vehicle information and vehicle motion information using at least one of a camera sensor and an inertial measurement sensor, using the lane information and the vehicle motion information, Calculating a probability that each of the kinematic models represents a fitness that corresponds to a driving condition, applying a weight based on the calculated probability to each model, and calculating a steering torque of the vehicle using the weighted fusion model And using the calculated steering torque to maintain the lane keeping of the vehicle And a step of performing a lateral control motion (lateral motion control).

The apparatus and method for a lane keeping control according to an embodiment of the present invention are characterized by using a dynamic lateral motion model and a kinematic lateral motion model of a vehicle for lane keeping, And determine the appropriate model for the vehicle, and control the lateral motion of the vehicle using the determined model.

Brief Description of the Drawings Fig. 1 is a diagram illustrating a schematic configuration of a lane keeping control device according to an embodiment of the present invention. Fig.
2 is a flow chart of a lane keeping control method according to an embodiment of the present invention;
3 is a diagram showing the lateral motion of the vehicle by the lane keeping control device according to the embodiment of the present invention.
4 shows a single track bicycle model;
5 to 10 show experimental results of a lane keeping control apparatus according to an embodiment of the present invention.

As used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. In this specification, the terms " comprising ", or " comprising " and the like should not be construed as necessarily including the various elements or steps described in the specification, Or may be further comprised of additional components or steps. Also, the terms " part, " " module, " and the like described in the specification mean units for processing at least one function or operation, which may be implemented in hardware or software or a combination of hardware and software .

Before describing embodiments of the present invention, various terms such as parameters and subscripts described in the present specification will be defined as follows.

: 차량의 무게중심으로부터 턴(turn) 중심까지의 거리

: Distance from the center of gravity of the vehicle to the center of the turn

: 차로 중심으로부터 턴 중심까지의 거리

: Distance from the center of the car to the center of the turn

: 차량 무게중심에서의 횡방향 차로 중심 오프셋(lateral lane center offset)

: Lateral lane center offset at the vehicle center of gravity

: 전방 주시 지점(look-ahead point)에서의 횡방향 차로 중심 오프셋

: The lateral lane center offset at the look-ahead point

: 전방 주시 거리(look-ahead distance)

: Look-ahead distance

: 차로 중심의 요 각도 기울기(yaw angle slope)

: Yaw angle slope at the center of the car

: 전방 주시 지점에서의 차로 중심의 요 각도 기울기

: The center yaw angle slope at the front sight point

: 차량 무게중심에서의 헤딩 각도 오차(heading angle error)

: Heading angle error at the vehicle center of gravity

: 전방 주시 지점에서의 헤딩 각도 오차

: Heading angle error at forward viewing point

: 요레이트(yaw rate)

: Yaw rate

: 차량 무게중심에서의 차량 속도

: Vehicle speed at the center of gravity of the vehicle

: 타이어 슬립 각도(tire slip angle)

: Tire slip angle

: 차량 무게중심에서의 사이드 슬립 각도(side slip angle)

: Side slip angle at the vehicle center of gravity

: 타이어의 코너링 강성(cornering stiffness)

: Cornering stiffness of a tire

: 횡방향 타이어 힘

: Lateral tire force

: 차량의 요 관성(yaw inertia)

: Yaw inertia of vehicle

: 차량의 총 중량

: Total weight of vehicle

: 차량의 무게중심과 타이어간 거리

: Distance between the center of gravity of the vehicle and the tire

: 제어 입력(=

: 스티어링 각도)

: Control input (=

: Steering angle)

: 전방(front)

: Front

: 후방(rear)

: Rear

: 종방향(longitudinal)

: Longitudinal

: 횡방향(lateral)

: Lateral

: desired

: desired

Hereinafter, various embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a diagram illustrating a schematic configuration of a lane keeping control apparatus according to an embodiment of the present invention. FIG. 2 is a flowchart of a lane keeping control method according to an embodiment of the present invention. Fig. 4 is a view showing a single track bicycle model; Fig. 4 is a view showing a single track bicycle model; Fig.

1, a lane keeping control apparatus 100 according to an embodiment of the present invention includes a model generating unit 110, a measuring unit 120, an interaction multiple model filter (IMM) 130) and a control unit 140. The control unit 140 includes a control unit 140 and a control unit 140. [

Hereinafter, the functions of each component and the steps performed in each step will be described in detail with reference to FIGS. 1 to 4. FIG.

First, in step S210, the model generating unit 110 generates a dynamic lateral motion model and a kinematic lateral motion model that take into account the forward viewing distance of the vehicle for lane keeping. do.

The dynamic and kinematic lateral motion models according to embodiments of the present invention are generated in a quadratic discrete-time state-space model.

Referring to Fig. 3, a vehicle controlled at the center of gravity may cause yaw rate vibration. Conversely, if the forward viewing distance L is used, the damping ratio of the dominant pole is affected by influencing the zero position. Therefore, each model will be derived with respect to the lateral position and the velocity error at the forward viewing point.

First, generation of the dynamic lateral motion model will be described.

, 제어 입력

및 외부 신호

에 관하여 하기 수학식으로 생성될 수 있다.The systematic state function of the basic dynamic transverse motion model relates to transverse position and yaw angle error with respect to the road and can be formulated with respect to transverse position and velocity error at the forward viewing point. The forward viewing distance affects the position of the zero and affects the damping ratio of the dominant pole. Thus, the system state function of the dynamic transverse motion model is expressed as a state vector

, Control input

And external signals

Can be generated by the following equation.

here,

로부터

로 나타내어진다.As such, the dynamic transverse motion model can be obtained with a discrete-time equivalent model, such as Equation 1, using a zero-order-hold method. Given a sampling time T _c of an electronic control unit (ECU), the discrete time matrix is

from

Lt; / RTI >

Next, the generation of the kinematic lateral motion model will be described.

The kinematic lateral motion model is derived on the assumption that there is no tire slip angle alpha. Thus, the kinematic lateral motion model does not include vehicle parameters such as cornering stiffness or tire / road conditions. For this reason, the kinematic lateral motion model is more suitable for describing vehicle behavior than the dynamic lateral motion model when the vehicle is traveling at low speed and there are unknown vehicle parameters. However, the kinematic lateral motion model is not valid for the lateral control system when there is a large steering angle that causes the tire slip angle. For example, changing from high speed to sudden lane or obstacle avoidance can result in a tire slip angle that can no longer be ignored.

를 이용하여 도 2로부터 키네마틱 횡방향 모션 모델이 생성될 수 있다. 이 상태 벡터는 전술한 다이나믹 횡방향 모션 모델에 사용된 상태 벡터와 동일하다. 하기의 도출 과정을 이용하여, 키네마틱 횡방향 모션 모델이 생성된다.State vector

A kinematic lateral motion model can be generated from Fig. This state vector is the same as the state vector used in the above-described dynamic transverse motion model. Using the following derivation process, a kinematic lateral motion model is created.

here,

를 이용하여, 수학식 2는 다시 하기 수학식으로 나타낼 수 있다.

(2) can be expressed by the following equation. &Quot; (2) "

이며, 다음과 같이 나타낸다.Another factor to be considered here is the lateral acceleration of the vehicle

And is expressed as follows.

으로부터, 수학식 4는 하기 수학식과 같이 횡방향 오프셋 오차에 관하여 나타낼 수 있다.

, Equation (4) can be expressed in terms of the lateral offset error as shown in the following equation.

, 제어 입력

및 외부 신호

를 이용하여, 하기 수학식으로 이산시간 상태공간 모델이 생성될 수 있다.State vector

, Control input

And external signals

The discrete time state space model can be generated by the following equation.

here,

Next, in step S220, the measuring unit 120 measures the lane information and the vehicle motion information.

과 차량 무게중심에서의 헤딩 각도 오차

는 카메라 센서를 이용하여 측정될 수 있으며, 요레이트

는 관성 측정 센서를 이용하여 측정될 수 있다.For example, the measuring unit 120 may include a camera sensor and an inertial measurement sensor. The camera 120 may measure information using a camera sensor, and may measure vehicle motion information using an inertial measurement sensor. 3 and / or 4, the lateral lane center offset at the forward viewing point

And the heading angle error at the vehicle center of gravity

Can be measured using a camera sensor,

Can be measured using an inertia measurement sensor.

Next, in step S230, the IMM filter 130 calculates a probability indicating the adaptability according to the driving condition for each of the dynamic lateral motion model and the kinematic lateral motion model using the measured lane information and vehicle motion information And weights according to the calculated probabilities are applied to each model.

The IMM filter 130 is a hybrid filter for lanes, which is one of the most cost effective hybrid state estimation techniques. The lane keeping controller 100 according to the embodiment of the present invention can perform lateral motion control for maintaining the vehicle by a lane using both a dynamic lateral motion model and a kinematic lateral motion model. For example, in a highway driving situation, the motion of the vehicle can always be switched between the dynamic lateral motion model and the kinematic lateral motion model.

Based on the above equations, the IMM filter 130 is designed for vehicle lateral motion state estimation and reliability of each model under various driving conditions.

In the IMM filter 130 according to the embodiment of the present invention, although the model noise covariances of the respective models are different, since the information of the lane and the vehicle motion are measured using the same camera sensor and inertial motion sensor, The sensor noise covariance is exactly the same.

The IMM filter 130 may perform the four steps of interaction, filtering, updating, and combination, as described below.

1. Interaction

는 모델 i에서 j로 전환되는 마르코프 확률(Markov probability)을 나타내고,

는 시간 k에서 모델 i의 모델 확률을 나타낸다. 그리고, 혼합 확률

은 하기 수학식과 같이 나타낼 수 있다.

Denotes a Markov probability which is converted from model i to j,

Represents the model probability of model i at time k. Then, the mixing probability

Can be expressed by the following equation.

Equation 8 is a normalized factor. Here, the index number i, j = 1 represents the dynamic transverse motion model, and the index number i, j = 2 represents the kinematic transverse motion model. For example, the following Markov probability matrix can be used.

및 혼합된 초기 공분산

은 하기 수학식에 의하여 주어진다.Mixed initial state of model j

And mixed initial covariance

Is given by the following equation.

및

는 각각 이전 샘플링 시간에서 모델 j의 추정된 상태 및 공분산이다.here,

And

Is the estimated state and covariance of model j at each previous sampling time.

That is, at this stage, the IMM filter 130 calculates the maintenance and change probability indicating the applicable fitness of each model at the previous sampling time, and calculates the initial state and the initial covariance for each model using the calculated maintenance and change probability .

2. Filtering

및

이 이용된다. 여기서, 요레이트

는 관성 센서(inertia sensor)에 의하여 측정되고, 전방 주시 지점에서의 횡방향 차로 중심 오프셋

과 무게중심에서의 헤딩 각도 오차

는 카메라 센서에 의하여 획득된다. 그래서, 하기의 측정 행렬 H는 키네마틱 횡방향 모션 모델 및 다이나믹 횡방향 모션 모델 둘 다에서 사용된다.For filtering, the dynamic lateral motion model and the kinematic lateral motion model, respectively,

And

. Here,

Is measured by an inertia sensor, and the lateral lane center offset at the forward viewing point

And the heading angle error at the center of gravity

Is obtained by a camera sensor. Hence, the following measurement matrix H is used in both the kinematic lateral motion model and the dynamic lateral motion model.

The camera sensor uses a third order polynomial such as: < RTI ID = 0.0 >

이라는 정의에 의하여 계산된다.Where c ₀ represents the lateral lane center offset at the center of gravity of the vehicle, c ₁ represents the heading angle error that is an in-lane heading slope at the center of gravity of the vehicle, and c ₂ represents s = 0 Curvature / 2, and c ₃ represents curvature ratio / 6. s is the arc length, and L is the forward viewing distance of the vehicle. The state measured by the camera sensor is

Is calculated by definition.

및 공분산

은 각 싱글 칼만 필터(single Kalman filter)에 의하여 계산된다.Based on both the dynamic lateral motion model of equation (1) and the kinematic lateral motion model of equation (6), the state

And covariance

Is calculated by a single Kalman filter.

1) Prediction

는 다이나믹 횡방향 모션 모델을 나타내고,

는 키네마틱 횡방향 모션 모델을 나타낸다.here,

Represents a dynamic lateral motion model,

Represents a kinematic lateral motion model.

및

)와 각 모델에 대하여 추정된 초기 상태 및 초기 공분산을 이용하여 이전 샘플링 시간에 대한 현재 샘플링 시간의 현재 상태 및 현재 공분산의 예측값을 산출한다.That is, at this stage, the IMM filter 130 calculates the coefficient of each model calculated from the lane information and the vehicle motion information measured at the current sampling time

And

) And the estimated initial state and initial covariance for each model to calculate the current state of the current sampling time for the previous sampling time and the predicted value of the current covariance.

2) Correction

및

는 각각 j 조건 필터의 이노베이션(innovation) 및 공분산이다.here,

And

Is the innovation and covariance of the j-conditional filter, respectively.

That is, at this stage, the IMM filter 130 calculates the innovation and the covariance of the Kalman filter of each model, and then calibrates the predicted value using the Kalman filter of each model to estimate the current state and the current covariance.

3. Updating

는 가우시안 가정에 따라 하기 수학식으로 주어진다.The likelihood function of each model j at time k

Is given by the following equation according to the Gaussian assumption.

The model probability of each model is calculated by the following equation.

Here, c is a normalization coefficient.

That is, at this stage, the IMM filter 130 calculates the likelihood function of each model according to the Gaussian assumption by using the innovation and covariance of the Kalman filter of each model, Lt; RTI ID = 0.0 > probability of change. ≪ / RTI >

4. Combination

및 이의 공분산

는 하기 수학식과 같이 계산된다.Combined state

And its covariance

Is calculated by the following equation.

That is, at this stage, the IMM filter 130 calculates a state and a covariance by using the current maintenance and change probability for each model, the estimated current state, and the current covariance, The weights according to the change probability can be applied to each model.

Next, in step S240, the controller 140 calculates the steering torque of the vehicle using the fused model to which weighting is applied to each of the dynamic lateral motion model and the kinematic lateral motion model that considers the forward viewing distance of the vehicle, And carries out lateral motion control for maintaining the vehicle by using the calculated steering torque.

5 to 10 are diagrams showing experimental results of a lane keeping control apparatus according to an embodiment of the present invention.

The performance of each model-based filter and IMM filter 130 was verified through experiments using a small SUV (sports utility vehicle). Camera sensors, yaw rate sensors and dSPACE's AutoBox were installed in the test vehicle. And the experiment was carried out at ITS high speed circuit of KIAPI. The high-speed circuit specifications are as follows.

Total length: 3,681m (4th lane)

Minimum curve radius: 100 m

Maximum bank angle: 30 degrees

Maximum speed: 204km / h

The reliability of the IMM filter 130 and the dynamic and kinematic models according to the embodiment of the present invention was evaluated in the following three driving conditions.

Maintain by car on straight road

Maintain by car on the curve road (R = 100m)

Change from straight road to car

Because there was no significant noise in the measurement data, the great sensitivity of switching from noise to noise was not taken into account. Hereinafter, the experimental results will be described only for the change case with one difference.

Figs. 5 to 7 show experimental results under a change from a straight road to a lane. Figs. 5 to 7 show camera data, steering wheel angle, yaw rate, and probability of each model, respectively. The tire slip angle can not be measured, but indirectly, the tire slip angle due to the side slip angle can be observed. The side slip angle of the vehicle was measured using a real-time kinematic GPS as shown in FIG.

As shown in Figs. 5 to 8, when changing to a car, the side slip increases and the probability of the dynamic model increases.

9 shows the estimation performance of the IMM filter 130 and each single filter when changing from a straight road to a car. A reference signal is obtained from the camera sensor and the inertial measurement sensor. Each single model based estimation has its drawbacks.

을 추정할 수 있음을 보여준다. 그러나, 다이나믹 모델 기반 필터는

에서 나쁜 성능을 가진다. 이에 반해, 키네마틱 모델 기반 필터는

에서 나쁜 성능을 가진다.Figure 10 shows that both dynamic and kinematic model based filters are effective

Can be estimated. However, the dynamic model-based filter

Which has a bad performance. In contrast, the kinematic model based filter

Which has a bad performance.

에 대하여, 차량은 스티어링 각도

를 이용하여 횡방향 힘을 만들 수 있다. 즉, 횡방향 차량 키네마틱 모델은 횡방향 모션을 표현하기에 충분한다. 반면에, 큰

에 대하여, 스티어링 각도

는 횡방향 힘을 만들기에 충분치 않다. 이 상황에서, 슬립 각도

는 도로를 추적하기 위하여 충분한 횡방향 힘을 만드는 것을 필요로 하며, 횡방향 차량 키네마틱 모델은 차량 횡방향 모션을 표현하기에 더 이상 적합하지 않다.small

The steering angle

To create a lateral force. That is, the lateral vehicle kinematic model is sufficient to represent the lateral motion. On the other hand,

The steering angle

Is not sufficient to make lateral forces. In this situation,

Requires a sufficient lateral force to track the road, and the lateral vehicle kinematic model is no longer suitable for expressing vehicle lateral motion.

Thus, the IMM filter 130 can reflect the characteristics of each model and contribute to the slip angle estimation, which is very difficult to estimate accurately due to sensor cost and complexity.

On the other hand, the components of the above-described embodiment can be easily grasped from a process viewpoint. That is, each component can be identified as a respective process. Further, the process of the above-described embodiment can be easily grasped from the viewpoint of the components of the apparatus.

In addition, the above-described technical features may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

It will be apparent to those skilled in the art that various modifications, additions and substitutions are possible, without departing from the spirit and scope of the invention as defined by the appended claims. Should be regarded as belonging to the following claims.

100:
110:
120:
130: Interaction multi-model filter
140:

Claims (12)

A generating unit for generating a dynamic lateral motion model (hereinafter referred to as a dynamic model) and a kinematic lateral motion model (hereinafter referred to as a kinematic model) in consideration of a forward viewing distance of the vehicle;
A measurement unit for measuring information on a vehicle and motion information of the vehicle using at least one of a camera sensor and an inertial measurement sensor;
And calculating a probability that each of the dynamic model and the kinematic model represents an application fitness according to a driving condition using the lane information and the vehicle motion information and applying a weight based on the calculated probability to each model, IMM Filter (Interaction Multiple Model Filter); And
And a control unit for calculating a steering torque of the vehicle using the fusion model to which the weight is applied and performing lateral motion control for maintaining the vehicle by the vehicle using the calculated steering torque, controller.
The method according to claim 1,
Wherein the interaction multi-model filter performs a four-step procedure of interaction, filtering, updating, and combination.
3. The method of claim 2,
The interaction multi-model filter calculates a maintenance and change probability indicating the fitness of each model at a previous sampling time in the interaction, and estimates an initial state and an initial covariance for each model using the calculated maintenance and change probability And a control unit for controlling the vehicle.
The method of claim 3,
Wherein the maintenance and change probability of the dynamic model and the kinematic model is calculated by the following equation.

here,

Denotes a Markov probability which is converted from model i to j,

Represents the model probability of model i at time k,

, Index number i, j = 1 represents the dynamic model, and index number i, j = 2 represents a kinematic model
The method of claim 3,
Wherein the interaction multi-model filter is a filter for estimating a current sampling time for a previous sampling time using the coefficient of each model calculated from the lane information and the vehicle motion information measured at the current sampling time and the estimated initial state and the initial covariance And calculates a predicted value of the current state and the current covariance.
6. The method of claim 5,
The interaction multi-model filter calculates the innovation and covariance of the Kalman filter of each model in the filtering, corrects the predicted value using the Kalman filter of each model, and outputs the current state and the current covariance Is estimated based on the estimated vehicle speed.
The method according to claim 6,
Wherein the interaction multi-model filter calculates the likelihood function of each model according to the Gaussian assumption using the innovation and covariance of the Kalman filter of each model in the update, And calculates a current maintenance and change probability for each model.
8. The method of claim 7,
Wherein the current maintenance and change probability is calculated by the following equation.

here,

ego,

And

Are the innovations and covariances of the Kalman filter, respectively, and c is the normalization coefficient
8. The method of claim 7,
Wherein the interaction multi-model filter calculates the combined state and covariance of the dynamic model and the kinematic model using the current maintenance and change probability and the estimated current state and the current covariance, Model is applied to the vehicle.
The method according to claim 1,
The system state functions of the dynamic model and the kinematic model include a lateral lane center offset at the look-ahead point,

, Lateral velocity error

Heading angle error at the vehicle center of gravity,

And a yaw rate

And a state vector including the state vector.
11. The method of claim 10,
The lane information

And

Lt; / RTI >
The vehicle motion information

And a control unit for controlling the vehicle.
Generating a dynamic lateral motion model (hereinafter referred to as a dynamic model) and a kinematic lateral motion model (hereinafter referred to as a kinematic model) in consideration of the forward viewing distance of the vehicle;
Measuring vehicle information and vehicle motion information using at least one of a camera sensor and an inertial measurement sensor;
Calculating a probability that each of the dynamic model and the kinematic model represents an application fitness according to a driving condition using the lane information and the vehicle motion information, and applying a weight according to the calculated probability to each model; And
Calculating a steering torque of the vehicle using the fusion model to which the weight is applied and performing lateral motion control for maintaining the vehicle by the vehicle using the calculated steering torque; Control method.

Images