WO2004066047A1

WO2004066047A1 - System and method for generating steering angle and speed control data for automatic navigation vehicles via compensation of variation in driving

Info

Publication number: WO2004066047A1
Application number: PCT/KR2003/002933
Authority: WO
Inventors: Yong-Kwon Choi; Yong-Jun Choi; In-Cheol Choi
Original assignee: Yong-Kwon Choi; Yong-Jun Choi; In-Cheol Choi
Priority date: 2003-01-24
Filing date: 2003-12-31
Publication date: 2004-08-05
Also published as: KR100540743B1; KR20040067584A

Abstract

Disclosed is a system and method for generating steering angle data via deviation compensation in driving which can analyze image data obtained by image pickup means (10) mounted on a front portion of the vehicle in order to adjust the steering angle of the vehicle and control the speed thereof thereby enabling unmanned driving of the vehicle. The present invention can obtain a lateral deviation ¥Äx between a road and an optical axis of image pickup means (10), calculate the steering angle of the vehicle based upon the lateral deviation, shaft-to-shaft distances L of the vehicle and a lateral distance d from the image pickup means (10) to adjust the degree of steering, and analyze a longitudinal deviation ¥Äy and a position of a leading vehicle of image data in front of the automatic navigation vehicle to actuate a brake (58a) and an accelerator (57a) so as to automatically control the vehicle.

Description

SYSTEM AND METHOD FOR GENERATING STEERING ANGLE AND SPEED CONTROL DATA FOR AUTOMATIC NAVIGATION VEHICLES VIA COMPENSATION OF VARIATION IN DRIVING

Technical Field

The present invention relates to generation of driving data in an automatic navigation system mounted on a vehicle capable of allowing for unmanned driving of the vehicle. More particularly, the present invention relates to a system and method for generating steering angle data via deviation compensation in driving, which can analyze image data obtained by image pickup means mounted on a front portion of the vehicle in order to adjust the steering angle of the vehicle and control the speed thereof . In particular, the present invention relates to a method and system for generating speed control data, which can obtain a lateral deviation between a road and an optical axis of image pickup means, calculate the steering angle of the vehicle to provide steering data, and drive a brake and an accelerator.

Background Art

There is a conventional automatic navigation system, disclosed in Korean Patent Publication Serial No .10-0257592 , which can detect the curvature of lanes at both sides in a road with a lane detector so that a vehicle can run along the detected curvature .

The Korean Patent Laid-Open publication (Korean Patent Application Serial No .10-2000-0009711) discloses "Method of Geometrically Mounting Image Pickup Means and Calculating Steering Angles for Self -Controlled Steering. " However, this document fails to represent a detailed method for calculating steering angles but uses experimental estimates to calculate radii. Further, the disclosed method of calculating steering angles is already known from general introductions for vehicles. Furthermore, there have been published several methods for expressing steering angle based upon unknown radius of gyration such as " Supporting System and Method for Vehicle to

Maintain Lane" , disclosed in Korean Patent Laid-Open publication (Korean Patent Application Serial No.10-2001-0032824) , which warns lane departure at the failure of calculating radius of gyration. However, a specific method expressed in variation of an image has not been yet published. Furthermore, an apparatus for preventing lane departure was filed as can be seen from the Korean Patent Laid-Open Publication (Korean Patent Application Serial No.10-2001-0032815) entitled "System and Method for Preventing Lane Departure of Vehicle . "

Disclosure of Invention It is an object of the present invention to provide an automatic navigation system and method for a vehicle, which can analyze road image data obtained from image pickup means mounted on a front portion of a running vehicle to calculate a target steering angle and generate data for controlling the speed of the vehicle thereby to compensate deviation in driving .

It is another object of the present invention to provide an automatic navigation system and method for a vehicle via deviation compensation in driving without necessity for an accelerator sensor such as a gyroscope, an attitude indicator, a positioning device such as a GPS, an active mark on a road and the like .

The present invention can realize the above objects by providing a system and method which can obtain a lateral deviation between a road and an optical axis of image pickup means, calculate the steering angle of the vehicle based upon the lateral deviation, shaft-to-shaft distances of the vehicle and a lateral distance from the image pickup means to adjust the degree of steering, and analyze a longitudinal deviation and a position of a leading vehicle of image data in front of the automatic navigation vehicle to actuate a brake and an accelerator.

According to one aspect of the invention for realizing the above objects, there is provided an automatic navigation system of the invention generally comprises a steering angle control unit, a speed control unit and a drive unit. The automatic navigation system of the invention further comprises image pickup means mounted on the vehicle for photographing front road images, image data storage means for storing photographed image data and lane image-extracting means for extracting lane data from the image data of the road stored in the image data storage means. The steering angle control unit includes first calculating means for designating horizontal distances and shaft-to-shaft distances and calculating a lateral deviation based upon the lane data extracted by the image-extracting means, second calculating means for calculating a target steering angle based upon the lateral deviation, the horizontal distances and the shaft-to-shaft distances, steering angle-comparing means for comparing the calculated target steering angle with an actual steering angle measured by a steering angle sensor and steering information-generating means for generating steering information based upon differences between the calculated and measured steering angles. The speed control unit includes deviation-calculating means for calculating y according to procedures in FIG. 11 and a longitudinal deviation, that is, a difference from in a previous period based upon vehicle data extracted from the road image data by the vehicle image-extracting means and lane image data extracted from the road image data by the lane image-extracting means; status value-generating means for determining a status value according to a status value expression, speed-comparing means for comparing speed data inputted by a speed sensor with a set speed to find a difference therebetween, controller-selecting means for selecting a speed controller based upon the status value and a current speed and speed control information-generating means for comparing values of a brake sensor and an accelerator sensor to generate speed control information.

According to another aspect of the invention for realizing the above objects, The drive unit includes steering information storage means for storing information from the steering information-generating means, steering actuation means for independently adjusting steering operation based upon the steering information stored in the steering information storage means, speed control information storage means for storing speed control information generated by the speed control information-generating means, brake-driving means for driving a brake based upon the stored speed control information and accelerator-driving means for driving an accelerator based upon the stored speed control information.

Brief Description of the Drawings FIG. 1A illustrates the concept of initializing image pickup means in accordance with the present invention;

FIG. IB illustrates the concept of image average technique in accordance with the present invention; FIG. 2 illustrates definition of a lateral distance in accordance with the present invention;

FIG.3A illustrates a road image with a halfway line (main lane) in accordance with the present invention;

FIG. 3B illustrates a road image with both lanes in accordance with the present invention;

FIG. 4 illustrates the concept of a lateral deviation in accordance with the present invention;

FIG. 5 illustrates the concept of radius of gyration known in the art; FIG. 6 illustrates the concept of steering based upon outer wheels in accordance with the present invention;

FIG. 7 illustrates the concept of steering based upon image pickup means in accordance with the present invention; FIG. 8 is a block diagram of an automatic navigation system in accordance with the present invention;

FIG. 9 illustrates a hardware structure in accordance with the present invention;

FIG. 10 is a flowchart for illustrating image data initialization in accordance with the present invention; FIG. 11 is a flowchart for illustrating a steering angle-calculating method in accordance with the present invention;

FIG. 12 is a flowchart for illustrating calculation of a longitudinal deviation Δy in accordance with the present invention; and

FIG. 13 is a flowchart for illustrating generation of a status value in accordance with the present invention.

Best Mode for Carrying Out the Invention

Hereinafter a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings .

While there are various schemes for extracting lane data from a road image taken by image pickup means, it will be described only about a lane image average method which is conceptually shown in FIG. IB.

According to the invention, image data obtained from the image pickup means is processed into black and white images to calculate average position values of a lane according to respective projection lines. In order to calculate the average position values, projection lines are selected from the bottom to the top of the image data, each of the selected projections lines being divided by n, i.e., the number of pixels, so that average positions are measured for some of the pixels of large gray scales. Describing this with reference to FIG. IB, numbers from 0 to 250 are designated to 250 pixels, and most of the 250 pixels are eliminated except for those (white) having gray scales larger than average values. Than, the remaining pixels are averaged in their positions. If the average position of an adjacent region is in the vicinity of a boundary of a next region and the average position of the next region is also in the vicinity of a boundary of a previous region, the average positions are incorporated into one lane. In Korea, left lane is main lane, and right lane is sub-lane. Of course, the main lane and sub-lane will be designated to the contrary in Japan or United Kingdom.

FIG. 1A illustrates the concept of initializing image pickup means in accordance with the invention. A vehicle is parked in a moving direction on a central region of a road with straight lanes . The steering angle of the vehicle is adjusted so that a steering angle sensor is set "0". The optical axis of the image pickup means is disposed in the center of the vehicle, parallel with the center line of the vehicle, before the image pickup means is initialized.

The image pickup means may be deviated from the center line of the vehicle to a predetermined distance X₀. If the image pickup means is deviated from the center line to the distance X₀ as shown in FIG. 1A, the deviated distance is reflected to the initial value. The image pickup means is initialized by inputting all of projection line numbers Pn, horizontal distances of an actual image, distances Xm of a main lane m from the center of the optical axis and distances Xs to a sub-lane s appearing in the projection lines and the image pickup means as reported in Table 1 below.

Table 1: Initialization of Image Pickup Device

Distances d to respective projection lines are inputted in meter unit by measuring actual distances of the road displayed on a screen according to the respective lines, distances Xm and Xs from the center of the optical axis are expressed by the number of pixels on the screen.

In the invention, the image pickup means is mounted on the vehicle with its optical axis aligned with the moving direction of the vehicle because there is an advantage that the actual distances d and lateral deviations Δx of the image pickup means and the projection lines are expressed.without being mixed together.

An initialization table of the image pickup means as shown in Table 1 can be obtained by setting the origin of orthogonal coordinates in the center of the optical axis with respect to the image from the image pickup means. As shown in FIG. 2, a horizontal distance d indicates a distance from a vehicle running on a current point A to a front point B on which the vehicle will run after a time period Δt . A lateral deviation Δx indicates the degree of winding of a road, that is, the lateral distance from the optical axis to a point on which the vehicle will run after a time period Δt with respect to the current time of the vehicle . The lateral deviation Δx is a continuously measured value, and reflects a remaining quantity which is not reached or overshot in a previous period. The lane appearing in the image can be recognized via edge extraction, in which lane marks can be simply extracted from the photographed image.

The lateral deviations Δx are calculated based upon image position information of the extracted lane marks . Table

2 indicates positions of the lane obtained from the image pickup means in running of the vehicle. The lateral distance is calculated with reference to values in Table according to

[Expressions] 1 to 3 : Calculating the lateral distance respective projection lines with respect to the main lane (as shown in FIG. 3A) , Δx = Road Width X (Xnm - Xm) / (Xm - Xs) .... Expression

Calculating the lateral distance with respect to the sub-lane (as shown in FIG. 3B) ,

Δx = Road Width X (Xns - Xs) / (Xm - Xs) .... Expression 2 , and

Calculating the lateral distance with respect to both of the main lane and the sub-lane (as shown in FIG. 3B) ,

Δx = Road Width (Xnm + Xns)/{ (Xm - Xs)X2} - X₀ .... Expression 3, wherein Xnm and Xnx indicate coordinate-based deviations of a road on which the vehicle is running, obtained by the image pickup means as reported in Table 2 below.

Table 2 : Operation of Image Pickup Device

In other words, the lateral deviation Δx is the distance between the center of the optical axis of the image pickup means and the lane in the vehicle-moving direction as indicated in FIGS. 6 and 7, in which X₀ is the distance of the image pickup means from the center of the vehicle expressed in meter unit.

At the present of any lane mark, lateral deviations Δx are calculated from the bottom to the top of the image.

Calculated values are stored in a table, and when projection line numbers for calculation of steering angle are determined with respect to a time used by a system and a vehicle speed, the lateral deviations ^'Δx are automatically selected.

It is to be understood that a point of reflection may appear in a middle of the road and thus control is unstable if distances d from the image pickup means to the projection lines are prolonged. It is to be understood also that there is a problem of confused lanes if a value d is enlarged in a curved course .

A process of calculating a steering angle based upon the above lateral deviations Δx will be described as follows : FIG. 5 illustrates the concept of a steering system of a vehicle well known in the art. Referring to FIG. 5, F indicates a forward direction of the vehicle, T indicates a steering direction of a front outer wheel . A steering angle Δ is defined by the forward direction F of the vehicle and the front outer steering wheel of the vehicle, in which the steering angle of the front outer wheel is oriented to a tangential direction of a circle of rotation and extended perpendicular to a line connecting with the center of rotation. L indicates the shaft-to-shaft distances between front and rear wheels, and the radius of rotation is calculated according to

[Expression 4] below:

R = L / sinφ + r .... [Expression 4] , wherein Δ is expressed in rad, and L, R and r are in meter unit . Since sinΔ = L/R on the assumption that R is sufficiently larger than r, the radius of rotation is expressed according to [Expression 5] :

R = L / sinφ .... [Expression 5] .

[Expression 5] expresses a steering angle of an outer wheel of a vehicle in the absence of slippage. The steering angle of the vehicle is set on the basis of the outer wheel. A steering angle calculation method according to a lateral deviation Δx will be described as follows:

On the assumption that the forward road photographed by the image pickup means mounted on the vehicle is curved to the right as shown in FIG. 6, there is established [Expressions 6 to 8] below. When it is supposed to adjust steering so that the vehicle can move on a target point B as a result of analysis to the image taken from the vehicle moving toward a point A, the center C of a circle is set the origin of an xy coordinate system as expressed in [Expression 6] in order to obtain R based upon geometric conditions and coordinates to calculate the steering angle φ: x² + y² = R² .... [Expression 6] Since the vehicle is supposed to necessarily move on the point B as shown in FIG. 6, [Expression 6] will be modified as follows: Since the point B has a coordinate (R-Δx, d) , (R - Δx)² + d² = R², and R² - 2RΔx + (Δx)² + d² = R². Eliminating R² from both sides and then re-arranging above expressions,

2RΔx = (Δx)² + d².

Accordingly, the steering angle according to the lateral deviation Δx will be calculated as in [Expression 7] below: R = {(Δx) + d }/2Δx .... [Expression 7], wherein Δx is in meter unit. Substituting R to [Expression 4] above, sinφ = L/R = 2LΔx/{(Δx)² + d²} . Therefore, the target steering angle φ can be simply calculated based upon trigonometric function as expressed in [Expression 8] below:

Target steering angle φ = sin^_1{2LΔx/ ( (Δx²) + d² ) } .... [Expression 8]

In [Expression 8] above, if d is sufficiently larger than Δx, the target steering angle Δ = sin^"1 (2LΔx/d²) .

Therefore, the target steering angle can be calculated according to [Expression 8] above.

In the case of longitudinal deviations φy, lateral deviations are calculated based upon lane images extracted by lane image-extracting means. The lateral deviations Δx are calculated for all of projection lines in which the lateral deviations exist repeatedly. In this way, lateral deviation values are stored in Table 2. y indicates the maximum projection line number in which a lateral deviation exists. Then, it is judged by the vehicle image-extracting means for extracting the color of a vehicle with a color filter whether any vehicle image exists in a position having a lateral deviation. This operation is simply performed by inspecting whether any color value of the vehicle exists in a position where a lateral deviation exists. Then, presence of any vehicle between the stored y value and the automatic vehicle can be simply sensed. In this way, the position of any leading vehicle running in front of the automatic navigation vehicle can be simply detected. As shown in FIG. 4, according to the present invention, lane data detected by the image pickup means is processed according to respective projection lines to determine the maximum value a-a' of a projection line in which a lateral deviation Δx of the lane is calculated and thus adjust the speed of the vehicle.

According to Table 2, it can be understood that the road is linear in which the position of the lane is same as that of the lane on the screen set in initialization. It can be understood that the lateral deviation cannot be calculated further since there is an obstacle at a forward position of 150m from the vehicle.

Herein a longitudinal deviation y is a projection line of an a-a¹ line, and it is calculated that Δy = y (current period) - y (previous period) . Δy is difference in projection line numbers indicating the actual distance between proj ection lines, and determined as the balance to a distance which is determined in initialization according to Table 1. Regarding that the image pickup is operated at a predetermined time interval, its variation Δy/Δt can be calculated. That is, the vehicle speed is adjusted based upon a value which is varied according to relative velocity.

If there is no obstacle in front of the vehicle, the vehicle will proceed at a fixed speed (based upon speed limits on respective road sections) .

Absence of an obstacle is determined by judging whether the color value extracted by the vehicle image-extracting means exists in a lateral deviation position. In the absence of a leading vehicle between a-a' and the automatic navigation vehicle, the vehicle maintains the current speed. In the presence of a leading vehicle between a-a' and the automatic navigation vehicle, the speed is adjusted according to following status value expressions:

If 0 > Δy/Δt, an accelerator is stopped and a brake is operated to decelerate the vehicle speed;

If 0 = Δy/Δt, the current speed is maintained; and If 0 < Δy/Δt, the accelerator is operated only if the vehicle speed under a predetermined speed. According to above conditions, operation data of the brake and the accelerator can be generated.

FIG. 8 is a block diagram of an automatic navigation system of a vehicle via compensation of variation in driving in accordance with the present invention. The automatic navigation system of the invention will be described in detail with reference to FIG.8. The invention is related to a driving system of a hybrid vehicle available for both manual and automatic operation. The automatic navigation system of the invention generally comprises a steering angle control unit 60, a speed control unit 50 and a drive unit 70. The automatic navigation system of the invention further comprises image pickup means 10 mounted on the vehicle for photographing front road images, image data storage means 20 for storing photographed image data and lane image-extracting means 30 for extracting lane data from the image data of the road stored in the image data storage means 20. The steering angle control unit 60 includes first calculating means 61 for designating horizontal distances d and shaft-to-shaft distances L and calculating a lateral deviation Δx based upon the lane data extracted by the image-extracting means 30, second calculating means 62 for calculating a target steering angle based upon the lateral deviation Δx, the horizontal distances d and the shaft-to-shaft distances L, steering angle-comparing means 63 for comparing the calculated target steering angle with an actual steering angle measured by a steering angle sensor 60a and steering information-generating means 64 for generating steering information based upon differences between the calculated and measured steering angles. The speed control unit 50 includes deviation-calculating means 51 for calculating y according to procedures in FIG. 11 and a longitudinal deviation Δy, that is, a difference from y in a previous period based upon vehicle image data extracted from the road image data by the vehicle image-extracting means 31 and lane image data extracted from the vehicle image data by the lane image-extracting means 30; status value-generating means 52 for determining a status value according to a status value expression, speed-comparing means 53 for comparing speed data inputted by a speed sensor 50a with a set speed to find a difference therebetween, controller-selecting means 54 for selecting a speed controller based upon the status value and a current speed and speed control information-generating means 55 for comparing values of a brake sensor 58b and an accelerator sensor 57b to generate speed control information.

The drive unit 70 includes steering information storage means 65 for storing information from the steering information-generating means 64, steering actuation means 66 for independently adjusting steering operation 67 based upon the steering information stored in the steering information storage means 65, speed control information storage means 56 for storing speed control information generated by the speed control information-generating means 55, brake-driving means 58 for driving a brake 58a based upon the stored speed control information and accelerator-driving means 57 for driving an accelerator based upon the stored speed control information. Hereinafter the invention will be described in conjunction with its construction and operation with reference to FIG. 9. According to the invention, an image pickup module 11 photographs images of a road which are in turn stored in an image data storage module 22, and an Electronic Control Unit (ECU) 40 (or any other processor for executing corresponding operations) controls the lane image-extracting means 30 to extract the lane data from the road image data stored in an image data storage module 21. While edge extraction is generally used to extract the lane image data, an average method of section images per respective projection lines is also available.

The ECU 40 controls the first calculation means 61 for designating the horizontal distances d, the shaft-to-shaft distances L and the lateral deviation Δx based upon the lane data extracted from the lane image-extracting means to generate the steering information. The second calculating means- 62 for calculating a target steering angle based upon the lateral deviation Δx, the horizontal distances d and the shaft-to-shaft distances L is also controlled by the ECU 40. The ECU 40 also controls the steering angle-comparing means 63 to compare the calculated target steering angle with the actual steering angle measured by the steering angle sensor 60a and the steering information-generating means 64 to generate the steering information based upon the differences between the calculated and measured steering angles. In order to generate speed control information, the ECU 40 receives road image data extracted by the vehicle image-extracting means 31 from the vehicle image data and lane image data extracted by the lane image-extracting means 30 from the vehicle image data, and controls the deviation-calculating means 51 to calculate y according to procedures in FIG. 11 and a longitudinal deviation Δy, that is, a difference from y in a previous period. The ECU 40 controls the status value-generating means 52 to determine a status value according to a status value expression and the speed-comparing means 53 to compare speed data inputted by the speed sensor 50a with a set speed to find a difference therebetween. Then, the ECU 40 controls the controller-selecting means 54 to select a speed controller based upon the status value and the current speed and the speed control information-generating means 55 to compare the values of the brake sensor 58b and the accelerator sensor 57b to generate the speed control information.

In the drive unit 70, the information stored in the steering information-generating means 64 is stored in the steering information storage means 65 by the ECU 40, and a steering adjustment module 65a independently adjusts the steering operation 67 based upon the steering information stored in the steering information storage means 65. The speed control information storage means 56 stores the speed control information generated by the speed control information-generating means 55 under the control of the ECU

40, and the brake-driving means 58 and the accelerator-driving means 57 drive the brake 58a and the accelerator 57a based upon the stored speed control information to regulate the speed.

Hereinafter, the invention of the above construction will be described in conjunction with its operation referring to FIGS. 10 to 12. According to the invention, the road image data stored in the image data storage means are initialized according to initializing steps as can be seen in FIG. 10. In these steps, variables including projection line numbers Pn, lateral distances d, main lane distances Xm and sub-lane distances inputted from the image pickup means are initialized in Sll. The lateral distances d are measured for the respective main lane numbers and stored, and the main lane distances and the sub-lane distances are calculated and stored in Table 1 in S12 and S13. In these steps, initial values are set before start of automatic navigation. The main lane number is successively increased to the last number while corresponding data are stored in these steps. The initialization steps are completed by storing data of the last main lane in S14 and S15. The shaft-to-shaft distances L are separately measured and stored according to the type of the vehicle.

An actual subsequent automatic navigation process of the vehicle on a road w^'ill be described as follows. The image pickup means 10 mounted on the vehicle for photographing front road images take image data which are in turn stored in the image data storage means 20, and the lane image-extracting means 30 selectively extracts lane information from the road image data stored in the image data storage means 20. The lane information can be extracted via edge extraction or an average method of section images. Alternatively, other schemes may be also available. Deviations of front lanes are stored based upon the extracted lane image data as reported in Table 2. The first calculating means 61 of the steering angle control means 60 calculate a lateral deviation Δx according to one of [Expressions 1 to 3] above, and the lateral deviation Δx is calculated as set forth above. As a specific projection line is selected previously or determined according to judgment of the system, the lateral distance and the shaft-to-shaft distances L are determined. When the lateral deviation Δx is calculated, the steering angle for deciding the proceeding direction of the vehicle is calculated by the steering angle-calculating means 62.

As can be seen in FIG. 11, the steering angle is calculated based upon the data inputted by the image pickup means. A memory (L, d and Δx) is initialized in S21, values of the main lane numbers Pn are selected in S22, and the lateral deviation Δx is calculated as reported in Table 2 and stored in new variables in S23.

Then, the steering angle is calculated according to [Expression 8] above in S24 so that the steering angle of the vehicle can be adjusted.

The steering angle-comparing means 63 compares the calculated target steering angle with actual steering angle information from the steering angle sensor 60a, and the steering information-generating means 64 generates a target steering angle to which the vehicle will be steered.

The generated steering information is stored in the steering information storage means 65. The steering actuation means 66 is operated based upon the stored steering information. A time period from the operation of the image pickup module 11 to that of the steering actuation means 66 is referred to a control period Δt . At a long control period, overshooting occurs or control is impossible at the failure of reaching the target values . Accordingly, it is essential to reduce the time necessary for calculating the steering angle in determination of a response rate.

Further, the speed control unit 50 calculates the lateral deviation Δx of the road based upon the road lanes which are extracted for the respective projection lines from the bottom to the top by the lane image-extracting means 30. Calculation is repeated from the bottom to the top to the last lateral deviation. Upon completing calculation of the lateral deviation Δx, it is judged whether the vehicle image extracted by the vehicle image-extracting means 31 is positioned within the range of the lateral deviation. Then, a value y is established by the longitudinal deviation-calculating means 51 according to the afore-described method, and a longitudinal deviation Δy is obtained.

FIG. 12 illustrates a process of obtaining Δy. Referring to FIG. 12, this process initializes a main lane distance Xm, a sub-lane distance Xs , a distance X₀ from the center line of the vehicle and a lateral deviation Δx in S31. In S32 to S34 , the main lane number Pn is counted and the lateral deviation Δx is calculated and stored until the main lane number Pn corresponds to the last line based upon the values in Table 2. If the main lane number Pn is the last line, # y = the maximum main lane number Pn without a color value is stored in S35, and Δy = y (current period) -y (previous period) is obtained based upon the stored information.

If the main lane number Pn is not the last line in S34, Pn value is stored as a coordinate (Pn, Δx, Cn) , and then a next lateral deviation is repeatedly calculated for a next main lane number in S36 and S37. Then, according to the invention, the projection line position of the vehicle is compared with the maximum lateral deviation projection line to determine a small value as y. When the actual distance between y in the current period and y in the previous period is determined as a longitudinal deviation Δy, a status value is determined by the status value-generating means 52 according to the Expression as mentioned above. As shown in FIG. 13, the status value-generating means

52 retrieves a table of the coordinate (Pn, Δx, Cn) prepared by the longitudinal deviation (Δy) -calculating means in FIG. 12 to inspect whether there is a color value of a leading vehicle . For the color value of a leading value, data extracted by the vehicle image-extracting means is used in S41 and S42. If there is no color for a leading vehicle, a C code for maintaining a fixed speed is generated, and the process is returned. If there is a color for a leading vehicle, the calculated longitudinal deviation Δy and the operation period Δt are read and a relative velocity Δy/Δt is calculated in

S43 and S44, and then it is judged whether the value is "0".

If the relative velocity Δy/Δt is "0" in S45, the C code for maintaining a fixed speed is generated (S49) and the process is returned. If the relative velocity Δy/Δt is not "0", it is judged whether the relative velocity Δy/Δt is larger than "0". If the relative velocity Δy/Δt is larger than 0", an accelerator operation code A is generated in S48, and the process is returned. If the relative velocity Δy/Δt is smaller than "0", a BA code for simultaneously operating the brake and accelerator is generated in S47 and the process is returned.

As set forth above, the status code generated by the status value-generating means 52 is compared with the speed from the speed sensor 50a by the speed-comparing means 53 before operation of the brake 58a and the accelerator 57a is determined by the speed controller-selecting means 54. In the case of the C code according to the previous conditions by the speed control information-generating means 55, the current speed is maintained. In the case of the A code, the speed is accelerated. In the case of the BA code, the speed control information for stopping the accelerator while actuating the brake is generated and stored in the speed control information storage means 56. The brake and the accelerator are independently controlled according to the stored speed control information. In addition to the ECU 40, a hardware structure is provided with a command input module 40a and a monitor module

80 installed inside the vehicle for displaying driving conditions and the forward image. The monitor module 80 displays the vehicle speed, the distance to a leading vehicle, the steering angle, the brake status and the accelerator actuating status from the speed sensor 50a, the steering angle sensor 60a, the accelerator sensor 57b and the brake sensor

58b. Further, a GPS module, a communication module and a map module can be additionally provided. A positioning system of the GPS module is used to obtain general positional information of the vehicle and substantial automatic navigation is performed according to the invention.

Industrial Applicability

As set forth above, the invention provides a steering and speed control system for an automatic navigation vehicle. Based upon lane information from the image pickup means, a method and system for processing steering angle can be provided . The invention further provides a method and system for controlling speed based upon visible lane information.

Claims

What Is Claimed Is:

1. A system for generating target steering angle data for vehicles comprising: image pickup means (10) mounted on the vehicle for photographing a front road image; image data storage means (20) for storing photographed image data; lane image-extracting means (30) for extracting lane data from the image data of the road stored in the image data storage means (20) ; first calculating means (61) for designating horizontal distances d and shaft-to-shaft distances L and calculating a lateral deviation Δx based upon the lane data extracted by the lane image-extracting means (30) ; second calculating means (62) for calculating a target steering angle based upon the lateral deviation Δx, the horizontal distances d and the shaft-to-shaft distances L; steering angle-comparing means (63) for comparing the calculated target steering angle with an actual steering angle measured by a steering angle sensor (60a) ; and steering information-generating means (64) for generating steering information based upon differences between the calculated and measured steering angles.

2. The system for generating target steering angle data as set forth in claim 1, wherein the first calculating means calculates a lateral deviation according to any of [Expressions 1 to 3] below: Δx = Road Width X (Xnm - Xm) / (Xm - Xs) .... [Expression 1], Δx = Road Width X (Xns - Xs ) / ( Xm - Xs) .... [Expression 2] , and

Δx = Road Width X (Xnm + Xns) / ( ( Xm - Xs) X2 ) - X₀ .... [Expression 3] , wherein [Expression 1] is used .if the lane data corresponds to a main lane, [Expression 2] is used if the lane data corresponds to a sub-lane, and [Expression 3] is used if the lane data corresponds to both the main lane and the sub-lane .

3. The system for generating target steering angle data as set forth in claim 1, wherein the steering angle-calculating means calculates a lateral deviation Δx and a horizontal distance d of a target point based upon a lane image in the road image data obtained by the image pickup means, and then sets a target steering angle φ based upon the shaft-to-shaft distances L of the vehicle according to [Expression 4] below, Φ = sin^_1{2LΔx/ ( (Δx)²+ d²) } .... [Expression 4].

4. A system for generating speed control data for vehicles comprising: image pickup means (10) mounted on the vehicle for photographing a front road image; image data storage means (20) for storing photographed image data; lane image-extracting means (30) for extracting lane data from the image data of the road stored in the image data storage means (20) ; deviation-calculating means (51) for calculating a longitudinal deviation Δy based upon vehicle image data extracted from the road image data by vehicle image-extracting means (31) and lane image data extracted from the road image data by the lane image-extracting means (30) ; status value-generating means (52) for determining a status value according to a status value expression; speed-comparing means (53) for comparing speed data inputted by a speed sensor (50a) with a set speed to find a difference therebetween; controller-selecting means (54) for selecting a speed controller based upon the status value and a current speed; and speed control information-generating means (55) for comparing values of a brake sensor (58b) and an accelerator sensor (57b) to generate speed control information.

5. The system for generating speed control data as set forth in claim 4, wherein speed is adjusted according to the status value expression by stopping an accelerator and operating a brake to decelerate the vehicle speed at 0 > Δy/Δt, by maintaining current speed at 0 = Δy/Δt, and operating the accelerator at 0 < Δy/Δt.

6. A method for generating speed control data for vehicles, the method comprising the steps of: photographing a road image in front of a vehicle with image pickup means mounted on the vehicle and storing the the photographed road image data in a memory; extracting a lane image from the road image data stored in the memory; extracting a vehicle image from the road image data stored in the memory; calculating a steering angle of a traveling vehicle based upon the lane data extracted from the road image data and steering data inputted by a steering angle sensor; controlling vehicle speed based upon the vehicle image data extracted from the road image data and speed data inputted by a speed sensor; and controlling a steering drive unit for adjusting steering operation of the vehicle, a brake and an accelerator in response to control signals from steering angle control means and speed control means to control vehicle speed.

7. The method for generating speed control data as set forth in claim 6, wherein the step of calculating a steering angle comprises the steps of : calculating a lateral distance d and a lateral deviation Δx based upon the lane image data; calculating the steering angle based upon the calculated lateral distance d and lateral deviation Δx; comparing the calculated steering angle with the steering data inputted by the steering angle sensor to calculate a target steering value; and generating steering information based upon the target steering value.

8. The method for generating speed control data as set forth in claim 6, wherein the step of controlling vehicle speed comprises the steps of: calculating a lateral deviation Δx and a longitudinal deviation Δy of a vehicle based upon the lane image data and the vehicle image data; generating a vehicle status value based upon the calculated lateral deviation Δx and longitudinal deviation Δy; comparing the vehicle status value with the speed data inputted by the speed sensor to calculate a target speed value; and generating speed control information based upon the calculated target speed value and signals inputted by a brake sensor and an accelerator sensor.

9. The method for generating speed control data as set forth in claim 8, wherein the step of calculating a longitudinal deviation Δy comprises the steps of: initializing a main lane distance Xm, a distance Xs to a sub-lane, a distance X₀ from a center line of the vehicle and the lateral deviation Δx; counting a main lane number Pn, and until the main lane number Pn corresponds to the last line, calculating the lateral deviation Δx; if the main lane number Pn does not correspond to the last line, storing lateral deviation Δx as a coordinate (Pn, Δx, Cn) and calculating a next lateral deviation Δx for a next main lane number; and if the main lane number Pn corresponds to the last line, storing y = the maximum main lane number Pn without a color value and calculating Δy = y (current period) - y (previous period) based upon the stored information.

10. The method for generating speed control data as set forth in claim 6 , wherein the step of controlling vehicle speed comprises the steps of: analyzing the road image data for respective projection lines to obtain lane data and storing projection line numbers and lateral deviations; judging whether a color code for a leading vehicle exists in a stored coordinate position to confirm whether a leading vehicle exists within a predetermined distance; and after confirmation of a leading vehicle, comparing a projection line number corresponding to the leading vehicle with a projection line number obtained in a previous period to calculate a longitudinal deviation Δy, and dividing the longitudinal deviation Δy by [with] an operation period Δt to stop an accelerator and operate a brake to decelerate the vehicle speed if 0 < Δy/Δt , to maintain the vehicle speed if

0 = Δy/Δt, and to operate the accelerator if 0 > Δy/Δt.

11. The method for generating speed control data as set forth in claim 6, wherein the step of calculating a steering angle comprises the steps of: calculating a lateral deviation Δx and a horizontal distance of a target point based upon the lane image extracted from the road image data; and setting a target steering angle Φ based upon shaft-to-shaft distances L of the vehicle according to [Expression 4] below,

Φ = sin^_1{2LΔx/( (Δx)² + d²) } .... [Expression 4].

12. A system for controlling vehicle speed comprising: steering information storage means (65) for storing information generated by steering information-generating means (64) ; a steering control module (65b) for independently controlling steering operation (67) based upon the steering information stored in the steering information storage means 65; a brake control module (58) for actuating a brake (58a) based upon speed control information generated by speed control information-generating means (55) ; and an accelerator control module (57e) for actuating an accelerator (57e) based upon the speed control information generated by the speed control information-generating means (55) .