JP2676737B2 - Vehicle steering control device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle steering control device, and more particularly, to a device for automatically steering control of an automatic guided vehicle by fuzzy reasoning. <Prior Art> The applicant of the present invention has previously proposed in Japanese Patent Application No. 62-110100 a steering control device for a vehicle that applies fuzzy inference to the steering control of an automatic guided vehicle. The proposed steering system is equipped with a camera on an automated guided vehicle (referred to as a "vehicle") to capture an image of a white line on a travel route, image-processes the image, and forms an angle between the direction of the white line and the traveling direction of the vehicle. The steering angle is estimated and calculated by fuzzy inference from the vehicle attitude and the distance from the white line to the vehicle, that is, the vehicle distance and the distance to the target point to be moved based on the map information. It is like this. The control based on the above fuzzy inference controls the operation amount, that is, the steering angle using a control rule based on a language.
If is then, then it is controlled using a control rule, that is, a set of ambiguous concepts. In the above proposal, X, Y whose origin is the movement target point of the vehicle
Assuming distance coordinates, the X and Y axes are F (x), g
It is divided by the membership function of (y). This membership function uses the trapezoidal function parameters. The steering output angle is set for each space divided by the above-mentioned membership function, and the output angle is the output parameter H (m, n) determined so as to draw a smooth steering curve from the starting point to the target point. It is set based on. That is, the output parameter H (m, n) is obtained by the following control law, for example. if x = dx y = dy then H (m, n) Further, a membership function force steering output of output parameter H (m, n) and x = dx, y = dy is obtained. The parameter H (m, n) is such that when the distance y to the target point is long, the steering is gentler than when the distance y is short, and when the deviation x to the target point (the distance that the vehicle has to approach) is large, When the steering angle is small, a steeper steering is given, and the distance y to the target point is determined by weighting more than the displacement x to the target point. The steering control based on such fuzzy reasoning has a feature that the vehicle can move while drawing a smooth running locus. The steering angle output parameter H (m, n) based on the above fuzzy inference is changed by the operator so as to obtain an optimum trajectory while observing the simulation of the steering trajectory. <Problems to be Solved by the Invention> However, in the above proposal, the setting of the steering output parameter H (m, n) of the vehicle is approximated to a curve manually imagined by the operator or a target steering curve. As shown in FIG. 9, when the vehicle actually travels, the vehicle does not follow the initial running path of the curved line indicated by the chain line but may meander as shown by the solid line. In the case of meandering running, as shown in the figure, it was often not close to the initial curve or the target steering curve. Further, since this parameter setting is manually input by the operator, there is a problem that the parameter is different depending on each operator. <Object of the Invention> The present invention has been made to solve the above problems, and an object of the present invention is to provide a vehicle steering system capable of automatically correcting a parameter for a steering angle output. In the provision of the control device. << Means for Solving the Problems >> In order to achieve the above object, the present invention is configured as shown in FIG. 1 and calculates the steering angle by the set steering output parameter to set the vehicle from the starting point to the target point. In a steering control device for a vehicle that steers up to, a target steering curve setting means a for setting a target steering curve from a starting point to a target point, and a target steering curve set by the target steering curve setting means a are calculated as described above. In order to evaluate the displacement amount calculation means b for calculating the displacement amount with respect to the traveling locus based on the steering angle and the displacement amount calculated by the displacement amount calculation means b, the steering output parameter is corrected as the vehicle approaches the target point. Evaluation function setting means c for setting an evaluation function in which the allowable area for the calculation is smaller, and an allowable area in which the displacement amount calculated by the displacement amount calculation means b is set by the evaluation function setting means c. When in, and having a steering output parameter modifying means d for correcting the steering output parameter according to the calculated displacement. << Operation >> Since the present invention has the above-described configuration, the operation output parameter is corrected by the difference in displacement between the calculated steering locus and the target steering curve, and steering control is performed. << Description of Examples >> Hereinafter, the present invention will be described based on illustrated examples. FIG. 2 is a plan view showing a traveling state of the vehicle 1. The vehicle 1 has a front wheel 1a that is steered and a rear wheel 1b that is drive-controlled, and a camera 2 that can image the front of traveling of the vehicle 1.
have. Further, 3 is a white line indicating the traveling route of the vehicle. Further, FIG. 3 is a block diagram showing an electrical hardware configuration, in which 10 is an image processing unit, which processes the screen imaged by the camera 2 provided in the vehicle 1 to display the white line 3
To the vehicle distance dxi and the angle between the direction of the white line 3 and the traveling direction of the vehicle, that is, the vehicle attitude θsi. Reference numeral 11 denotes a map information section, which stores map information from the starting point to the destination point of the vehicle. Reference numeral 12 denotes a steering control unit, which is a vehicle posture and vehicle position obtained from the image processing unit 10 and a distance of a certain section from a starting point (current position) obtained from the map information unit 11 to a target point separated by a predetermined distance. Therefore, the steering angle from the starting point to the target point is calculated by fuzzy inference, and the traveling distance for one steering angle output is set based on the distance from the target point to the starting point. Reference numeral 13 denotes a steering output parameter correction unit which corrects the steering output parameter using the steering output parameter used in the fuzzy inference of the steering control unit 12 as a correction amount based on the displacement amount between the calculated traveling locus and the target steering curve. It is like this. Further, 14 is a traveling drive unit, which is configured to drive the vehicle 1 for a predetermined distance by the output from the steering control unit 12. Further, 15 is a steering drive unit, which is the steering control unit 12
The steering angle of the vehicle is driven to a predetermined steering angle by the output from. Explaining the operation of this embodiment having the above-mentioned configuration, first, as shown in FIG. 4, x, y coordinates having the starting point O of the vehicle 1 as the origin are created, and the target point T is set on these coordinates, From the starting point O to the target point T, one ideal steering trajectory, that is, a target steering curve is connected by a spline curve,
The curve is shown by the chain line in FIG. 4 and is represented by the function of x = f (y). That is, the curve indicated by the chain line indicates that when the vehicle 1 having the initial attitude θs travels from the starting point O to the target point T, the vehicle attitude θt is 0, that is, the vehicle 1 travels side by side so as to be parallel to the white line 3. Based on the premise, it has become a smooth and ideal running path. If the number of steering outputs by fuzzy inference from the starting point O to the target point T is i times (8 times in the example shown in the figure) and the position of each output is indicated by a black circle, the position deviated from the above running locus. Is displayed in. FIG. 5 is a graph showing the deviation between each output point and the traveling locus, that is, the relationship of the displacement amount (f (yi) -xi). In the figure, the displacement amount 0 means the target steering curve position, and the displacement amount does not become 0 because it is an approximation to the traveling locus, but it is not the final output point, that is, the target point T (i = 8 output point). If the displacement amount at 0 is not close to 0, it means that the vehicle attitude at this point and the distance from the white line 3 are not close to the instructed value. Therefore, as an evaluation function of this displacement amount, an evaluation function such that γ (i)> 0 in which the range of variation becomes smaller as the target point is approached is set, and depending on whether the displacement amount is larger or smaller than this function, the displacement amount The steering output parameter is corrected with a correction amount commensurate with the amount. The function g of the correction amount is expressed by the following equation. g (f (yi) −xi) = a · (f (yi) −xi) + b (a ≠ 0) Hereinafter, correction of the steering output parameter will be described. For example, when the displacement amount (f (yi) -xi) of the i-th output is larger than the evaluation function γ (i), that is, γ (i)-(f
When (yi) -xi) <0, the control law concerning the output i-th, that is, the rule having the weight concerning the i-th output is called. According to this rule, the i-th output position is “if the x direction is small, the y direction is small, and the vehicle attitude θs is small, the output angle H is small” (hereinafter, if x = “S” and y = “. S "andθ
It is described as s = S then H = H (S, S, S). Incidentally, S
Means small, M means medium and B means large. ) Is represented by the control rule in the language. Output parameter H of each rule related to i-th output
From the correction amount obtained from (m, n) and the value corresponding to the i-th output of the simulation, the weight of the rule is obtained by weighted averaging, and this is subtracted from the original output parameter to obtain the modified parameter. Can be obtained. To explain this correction method more specifically, for example, i
It is assumed that the following three rules (R1 to R3) are involved in the second output. R1: if x ＝ “S” and Y ＝ “S” and θs ＝ “S” then H ＝ H
(S, S, S) R2: if x = “S” and Y = “M” and θs = “S” then H = H
(S, M, S) R3: if x = "M" and Y = "S" and θs = "S" then H = H
(M, S, S) The weight μ of each rule described above is obtained from the membership functions of x, y and θs as follows. μR1 = μXS (xi) ∧ μYS (yi) ∧ μθs S (θsi) μR2 = μXS (xi) ∧ μYM (yi) ∧ μθs S (θsi) μR3 = μXM (xi) ∧ μYS (yi) ∧ μθs S (θsi Therefore, each modified output parameter is obtained as follows by subtracting the value obtained by multiplying the correction amount by the weighted average of μ from the original output parameter. It should be noted that, after the fifth output in FIG. 5, all parameters are in the parameter correction range, but since the next steering output is determined based on the vehicle position and vehicle attitude after steering output, relative correction cannot be performed. 5th output, that is, correction is repeated until γ (i)-(f (yi) -xi)> 0 at the first time in the parameter correction area, that is, until it enters the evaluation function area. . FIG. 7 shows the membership function when the above-mentioned parameter modification is performed. In this membership function, the target point is the origin, and nine points S1 to S9 with a membership value of 1 in both the x and y regions are extracted. FIG. 6 shows the displacement angle (H (i) -H (i-
1)), which means that the displacement angle 0 coincides with the direction locus shown by the chain line in FIG. Also with this displacement angle, the output parameters are corrected in the same manner as described above. Since the evaluation function of the displacement angle may be determined only by the displacement angle amount, the function γ ′ that makes the domain constant is set. Since the method of correction is the same as the above-mentioned displacement amount, detailed description thereof will be omitted. However, the first time the parameter correction range is entered, that is, γ′− | H (i) −H (i−1) | <0. The correction is repeated until γ′− | H (i) −H (i−1) |> 0 at the output times. The above-mentioned parameter correction processing is shown in the flowchart of FIG. 8. First, in step 100, initial values of parameters are set. It should be noted that this initial value can be randomly set by manual input, and in some cases, even if all are set to 0, the initial value will be corrected later and can be roughly set. Next, after the correction flag is set to 0 in step 102, the distance to the target point T and the initial vehicle attitude are set in the image processing (step 104). Further, in step 106, both the traveling locus spline curve of the vehicle up to the target point T and the fuzzy inference based on the output parameters arbitrarily set are simulated, and the deviation amount for each output point is calculated (step 108). It is judged whether or not the calculated deviation amount is within the correction area determined by the evaluation function (step 110), and if it is judged as the correction area in this judgment (step 110 affirmative), the process proceeds to step 112 and step 100 The parameter set in step 1 is corrected and the process proceeds to step 106. If it is determined in step 110 that the area is outside the correction area (NO in step 110), the flag is set to 0 (step 114), and the deviation angle is calculated (step 116). Then, it is determined whether or not the calculated deviation angle is in the correction area (step 118), and if it is determined to be within the correction area (step 118 affirmative), after setting a flag (step 120),
Proceed to the correction step 112 described above. On the other hand, if it is determined in step 118 that the area is outside the correction area (step 118, No), it is determined whether or not flag 1 is set (step 122). If flag 1 is set, the procedure returns to step 106 and stands. If not, it ends. In the above embodiment, the displacement amount and the displacement angle are obtained for each output from the starting point of the vehicle to the target point, and the steering output parameter is corrected by the correction amount corresponding to this, so this correction By outputting the steering angle based on the output parameter thus generated, the vehicle can be driven while drawing an ideal smooth steering locus. << Effect >> As described above, according to the present invention, when the displacement amount calculated by the displacement amount calculating means is outside the allowable range set by the evaluation function setting means, the steering output parameter is calculated according to the calculated displacement amount. The steering angle is calculated based on the corrected steering output parameter, and the vehicle can be steered by approximating the target steering curve.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram for responding to a claim, FIG. 2 is a plan view showing a traveling state of a vehicle, FIG. 3 is a block diagram showing an electrical configuration of the present invention, and FIG. 4 is a target steering. A graph showing a curve and the position of the calculated output point, FIG. 5 is a graph showing the relationship between the evaluation function and the displacement amount, FIG. 6 is a graph showing the relationship between the evaluation function and the displacement angle, and FIG. 7 is a member. Explanatory drawing showing the ship function, 8th
FIG. 9 is a flow chart showing a processing procedure of parameter correction, and FIG. 9 is an explanatory view showing a traveling locus of a conventional vehicle. a ... Target steering curve setting means b ... Displacement amount calculation means c ... Evaluation function setting means d ... Steering output parameter correction means

Claims (1)

(57) [Claims] In a vehicle steering control device that calculates a steering angle from a set steering output parameter and steers the vehicle from a starting point to a target point, a target steering curve setting means that sets a target steering curve from the starting point to the target point, Displacement amount calculation means for calculating the displacement amount between the target steering curve set by the target steering curve setting means and the traveling locus based on the calculated steering angle, and the displacement amount calculated by the displacement amount calculation means are evaluated. In order to
Evaluation function setting means for setting an evaluation function in which the allowable area for steering output parameter correction becomes smaller as the vehicle approaches the target point, and the displacement amount calculated by the displacement amount calculation means is set by the evaluation function setting means. And a steering output parameter correcting means for correcting the steering output parameter in accordance with the calculated displacement amount when the steering control device is outside the allowable range.