JPH03237505A - Steering controller for free-running vehicle - Google Patents

Abstract

PURPOSE:To irradiate a reference point with a light beam more surely by reducing at last one of the scanning speed of the light beam and the running speed of a free-running speed at the time when the frequency of loss signal exceeds a prescribed value. CONSTITUTION:The result of identification whether the reference point is detected or lost is supplied from a azimuth angle discriminating part 24 to a loss data storage part 39. identification results for all reference points obtained during a prescribed rotational frequency of the past rotation of an optical scanner are stored in the storage part 39. The number of loss data out of data stored in the storage part 39 is added by a loss number counting part 40, and the result is transferred to a loss comparing part 41. The comparing part 41 compares the number of supplied loss data and a set number in a predeterminate number setting part 42 with each other; and when the number of loss data is larger than the predeterminate number, signals to reduce the rotational frequency of the optical scanner and the running speed of the free-running vehicle are sent to scanning control part 5a of a motor and a driving control part 18 of an engine 19 respectively.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a steering control device for self-propelled vehicles, and particularly to a steering control device for self-propelled vehicles such as unmanned mobile conveyance devices in factories, agricultural and civil engineering machinery, etc. Regarding a control device.

(Prior Art) Conventionally, as a device for detecting the current position of a moving object such as the above-mentioned self-propelled vehicle, a device for scanning a destination beam generated by the moving object in a circumferential direction centering on the moving object, and A device has been proposed that is fixed at at least three locations away from the body and is equipped with a light reflecting means that reflects light in the direction of incidence, and a light receiving means that receives the reflected light from the light reflecting means (Japanese Patent Application Laid-Open No. 59-
67476).

The device detects the aperture angle between the three light reflecting means as seen from the moving object based on the light receiving output of the light receiving means, and compares the detected aperture angle with each preset light reflecting means. The mobile object position is calculated based on the position information of the mobile object.

In the above system, the light beam may not be able to be irradiated to the light reflecting means due to the tilt or vibration of the self-propelled vehicle, or the light receiving means may receive reflected light from an object other than the light reflecting means. Ta. If the reflected light from the scheduled light reflecting means is not received, the position of the self-propelled vehicle cannot be calculated, and as a result, the self-propelled vehicle may not be able to travel along the scheduled course.

On the other hand, for example, in Japanese Patent Publication No. 63-41489 (Japanese Patent Publication No. 59-104503), a light beam is scanned horizontally and vertically, and
A method of detecting the position of a moving object has been proposed in which the horizontal scanning speed and the vertical scanning angle are changed so that the light beam can be reliably irradiated onto the light reflecting means.

In this position detection method, when the direction of the light reflecting means is detected by the first scanning of the light beam, the vertical scanning angle of the light beam at that time is memorized, and in the second scanning, the direction of the light beam is memorized. Decrease the scan angle based on the angle,
The scanning speed in the horizontal direction is set to be higher than that during the first scanning.

In this way, the vertical scanning angle and the horizontal scanning speed are changed for each horizontal scanning to ensure that the reflected light from the light reflecting means is received.

(Problem to be solved by the invention) The above position detection method has the following problems.

In this method, the swing width of the optical mirror is changed to change the vertical scanning angle, and the speed of the scanner rotation motor is changed to change the horizontal scanning speed. In order to perform such an operation every rotation, that is, frequently, a scanner having very high control follow-up performance is required in consideration of durability and accuracy.

Additionally, in order to detect the position of a moving object based on the principle of triangulation, it is necessary to install at least three light reflecting means (two reflecting means if the moving object has a distance measurement function). There are also the following problems.

First, when the moving body is tilted, the vertical scanning angle ranges for detecting the light reflecting means located at opposite positions across the moving body are not the same. In other words, if the angular range in which one light reflecting means is detected is biased upward, the angular range in which the other light reflecting means is detected is conversely biased downward. Therefore, unless the light reflecting means installed in plurality and the moving body are on the same plane under favorable conditions, even if the vertical scanning angle is the same and one rotation is performed, the light reflecting means will not be accurate. It is not always possible to irradiate a light beam to

Furthermore, even if the moving object travels under the above-mentioned favorable conditions, the distances between the moving object and each light reflecting means are not the same, so the elevation angle of each light reflecting means as seen from the moving object is different from the above at any given time. The angle varies depending on the distance. Therefore, the vertical detection angle ranges of the respective light reflecting means centering on the elevation angle also differ.

As described above, in practice, it is necessary to frequently and variably control the fluctuation amplitude of the optical mirror, and as a result, it is necessary to use an expensive optical scanner or the like that can follow mechanical components at high speed. There is a problem.

An object of the present invention is to solve the above-mentioned problems of the prior art, and in particular, to increase the probability of irradiating a light beam to a light reflecting means in an environment where loss of sight occurs frequently without using a scanner with advanced performance. An object of the present invention is to provide a steering control device for a self-propelled vehicle configured to receive the reflected light.

(Means and Effects for Solving the Problems) In order to solve the above-mentioned problems and achieve the objects, the present invention provides a light beam that swings in the vertical direction around the self-propelled vehicle while moving the light beam in the circumferential direction. A beam scanning means for scanning, and an orientation of the light reflecting means as seen from the self-propelled vehicle by receiving the reflected light of the light beam from the light reflecting means installed at at least three reference points apart from the self-propelled vehicle. A means for measuring the angle and calculating the position of the self-propelled vehicle based on the result, and a criterion for determining whether or not the light reflecting means installed at the scheduled reference point has been detected based on the presence or absence of the output of the light receiving means. a point loss determination means, and scans the light beam by reducing at least one of the circumferential scanning speed of the light beam and the traveling speed of the self-propelled vehicle when the frequency of not being able to detect the light reflecting means is higher than expected. It is distinctive in that it is structured as follows.

In the present invention having the above configuration, the horizontal scanning density of the light beam is increased by lowering the scanning speed, thereby making it possible to reliably detect the light reflecting means.

In addition, the speed of the self-propelled vehicle itself is reduced to minimize the shaking of the vehicle body caused by the unevenness of the vehicle's running surface.
I strained it.

Both operations are performed only when the reference point is frequently lost, so in other normal conditions, the light beam is scanned at high speed and the traveling speed of the self-propelled vehicle itself is not reduced. efficiency will not decrease.

In addition, since the scanning speed and the traveling speed of the self-propelled vehicle need only be changed when the reference point is lost, the tracking performance of mechanical components such as the drive motor of the scanning means and the self-propelled vehicle drive section is particularly improved. It is not necessary.

(Example) An embodiment of the present invention will be described below with reference to the drawings.

FIG. 15 is a perspective view showing the arrangement of a self-propelled vehicle equipped with the control device of the present invention and a light reflector disposed in the travel area of the self-propelled vehicle. In the figure, a self-propelled vehicle 1 is, for example, a self-propelled vehicle for agricultural work such as a lawn mower.

The self-propelled vehicle 1 is equipped with an optical scanner 2. The optical scanner 2 consists of a fixed part 2a and a rotating part 2b, and the rotating part 2b is connected to a motor 5 by a belt.
It is driven by the motor 5 and rotates, for example, counterclockwise.

The light beam 2E generated by the optical scanner 2 is scanned in the direction of an arrow 29 as the rotating portion 2b rotates. Further, the rotating section 2b includes a swinging mechanism that swings the generated light beam in the vertical direction.

A plurality of reference points are set around the work area, and reflectors 6a to 6C made of well-known light reflecting means such as a corner cube prism having a reflective surface that reflects incident light in the direction of incidence are installed at the reference points. will be installed. The light beam 2E scanned in the direction of the arrow 29 is sequentially reflected by these reflectors 6a to 6C, and the reflected light 2R is sequentially received by the optical scanner 2.

The rotating part 2b is provided with a slit plate 7a, and the rotating part 2b is provided with a slit plate 7a.
rotates with. The slit of the slit plate 7a is detected by an angle sensor 7b, which will be described later, and the sensor 7b generates a pulse signal every time it detects a slit. By counting the pulses, the rotation angle of the rotating portion 2b can be detected.

For the purpose of explanation, the optical scanner 2 is shown in an exposed state, but it is of course equipped with a cover for dustproofing and drip-proofing.

Next, a swing mechanism that is built into the optical scanner 2 and scans the light beam in the vertical direction will be described.

FIG. 9 is a sectional view of the optical scanner, and the same reference numerals as in FIG. 15 indicate the same or equivalent parts.

In the figure, light generated by a light emitter (light emitting diode) 3a passes through lenses 4a and 4b and is reflected by a mirror 30. The direction of reflection is determined according to the inclination of the mirror 30. Of the light beam 2E reflected by the mirror 30 and emitted to the outside, reflected light 2R from the reflectors 6a to 6C returns to the optical scanner 2.

The reflected light 2R from the reflectors 6a to 6c is reflected by a mirror 30, passes through lenses 4b and 4a, hits a reflecting prism 4C, is refracted, and enters a light receiver (photodiode) 3b. The light received by the light receiver 3b is converted into an electrical signal, and based on the electrical signal, it is detected that the reflected light from the reflectors 6a to 6c has been received.

The mirror 30 is swung in the direction of an arrow 33 about a shaft 32 by rotation of a pin 31 fitted into a notch provided at one end thereof. That is, the pin 31 is attached eccentrically to the shaft of the swing motor 34, and the mirror 30 swings in accordance with the amount of eccentricity. As the mirror 30 swings, the light beam 2E is scanned in the vertical direction (in the direction of the arrow 35). The angle sensor 7b detects the slit of the slit plate 7a, and based on the detection signal, detects the rotation angle of the slit plate 7a, that is, the rotation angle of the rotating part 2b.

The pulley 37 of the rotating part 2b is connected to the fixed part 2a via the bearing 36.
The motor 5 is connected to the motor 5 by a belt 38.
rotation is transmitted to the pulley 37.

Note that the optical scanner shown in FIG. 9 is an example, and a well-known optical scanner equivalent to this scanner can be used. Similarly, the detection of the rotation angle of the rotating part 2b is not limited to the combination of the slit plate 7a and the angle sensor 7b, but may be performed using a known angle detection means such as a rotary encoder directly connected to the motor 5. It can also be used.

With the above configuration, the azimuth angle (hereinafter simply referred to as azimuth angle) of each reflector 6a to 6C as seen from the self-propelled vehicle is calculated based on the reflected light that is sequentially detected.

Then, based on the results, the self-position of the self-propelled vehicle 1 with respect to the reference point where each of the reflectors 6a to 6C is installed is detected, and steering control is performed.

By the way, there is a possibility that there is a reflective object or a light-emitting object other than the reflector in or near the driving area of the self-propelled vehicle 1, and the light receiver 3 may detect the light from this reflective object. It is possible that the reflected light from the reflector cannot be detected.

Therefore, in this embodiment, whether or not the detected light is from a scheduled reflector is determined by the following process.

FIG. 11 is an explanatory diagram of the reference point identification process. In the figure, the reflectors 6a to 6C are arranged at reference points A-C around the work area 22, respectively.

An arrow 29 indicates the scanning direction of the light beam emitted from the self-propelled vehicle 1.

In the illustrated arrangement, the azimuth of each reference point is calculated based on the detection signal of the light receiver 3b, and the same reference point is detected in the next scan based on the azimuth detected up to the present time. The correct azimuth is predicted. The predicted azimuth angle (predicted azimuth angle) of each reference point is the angle θp.
It is shown as a~θpc.

Light beam scanning direction 29 from each predicted azimuth θpa to θpc
The reference point identification direction pa is set in the direction in which the scanning has progressed by the angle θh.
-pc is provided. Of the reference point identification directions pa-pc, among the lights detected between two adjacent reference point identification directions, the incident light from the direction closest to the predicted azimuth is selected from the direction set at the scheduled reference point. It is determined that the light is from a reflector.

For example, the reference point identification direction pc and the reference point identification direction pa
Between the noise sources N1. Assume that the light from N2 and the reflected light from the reflector 6a installed at the reference point A are detected. In this case, by extracting the light from the direction closest to the predicted azimuth θpa from among these lights, the reflected light from the reference point A can be distinguished from the light from the other light sources Nl and N2 and detected.

Furthermore, the following process can be added to improve the accuracy of identifying reference points. In other words, if a predetermined range (an angle equal to or smaller than the angle θh) is set before and after the predicted azimuth, and even light from the direction closest to the predicted azimuth θpa to θpc deviates from the range. It is determined that the planned reference point has been lost.

If it is determined that the scheduled reference point has been lost because no light is detected between the two reference point identification azimuths or within the scheduled range set to improve identification accuracy, the processing cycle is started using the predicted azimuth angle. The position of the self-propelled vehicle 1 is calculated, and the predicted azimuth is further stored as the next predicted azimuth.

Next, the functional configuration of the control device of this embodiment will be explained according to the block diagrams shown in FIGS. 1 and 2. In the figure, the part surrounded by chain lines can be constructed by a microcomputer.

Fig. 1 shows the steering control function and reference point identification function of a self-propelled vehicle, and Fig. 2 shows a system for detecting frequent loss of reference points from the identification results of the reference point identification function and taking measures to deal with this. Indicates the function for

In FIG. 1, a light beam 2 emitted from a light emitter 3a
E is scanned in the rotating direction of the rotating section 2b, and the reflector 6
(6a-6c). The reflectors 6a-6
The reflected light 2R of c is received by the light receiver 3b.

The counter 9 counts pulses output from the angle sensor 7b as the slit plate 7a of the rotating section 2b rotates. The count value of the pulse is transferred to the azimuth detecting section 11 every time the light receiver 3b detects light. The azimuth detecting section 11 detects the reflector 6a based on the number of supplied pulses.
An azimuth angle of ~6C is calculated.

The azimuth detected by the azimuth detection unit 11 is stored in the azimuth storage unit 1.
The data stored in the azimuth storage unit 12 is transferred to the azimuth identification unit 24 in response to the identification timing signal supplied from the identification timing generation unit 23.

The identification timing signal is transmitted to the reference point identification azimuth pa-pc at the time when the scanning progresses to an azimuth that passes through the azimuth indicated by the predicted azimuth calculated by the azimuth angle prediction calculation unit 27 by a predetermined angle θh. It is output when the beam scan progresses. For this purpose, the identification timing generator 23
Then, when the number of output pulses from the angle sensor 7b corresponding to the angle obtained by adding the angle θh to the predicted azimuth calculated by the azimuth angle prediction calculating section 27 is taken in, an identification timing signal is output.

The azimuth identification unit 24 selects the light detected in the direction closest to the predicted azimuth calculated by the azimuth prediction calculation unit 27 from among the supplied azimuths from a reflector placed at a predetermined reference point. It is determined that the light is reflected light.

The azimuth angle data of the reflector determined by this judgment is
It is used when the azimuth prediction calculation unit 27 predicts the azimuth of the reflector to be detected in the next scan.

That is, the predicted azimuth is determined by a function of the azimuth determined by the azimuth angle identification unit 24 that is expected to be obtained experimentally. The predicted azimuth is not limited to the method of finding it based on a predetermined function, but may be found by adding the difference between the current and previous azimuths obtained by the azimuth identification unit 24 to the current azimuth.

The azimuth detected by the azimuth angle identification section 24 is determined by the opening angle calculation section 1.
0, and the opening angle between the reflectors 6a to 6C as seen from the self-propelled vehicle 1 is calculated.

The position/progressing direction calculating section 13 calculates the current position coordinates of the self-propelled vehicle 1 based on the aperture angle, and calculates the traveling direction of the self-propelled vehicle 1 based on the azimuth angle. This calculation result is input to the position/direction comparison section 25. The position/direction comparison unit 25 compares the data representing the travel course set in the travel course setting unit 16 with the coordinates and travel direction of the self-propelled vehicle 1 obtained by the position/direction calculation unit 13. be done.

This comparison result is manually input to the steering section 14, and based on the comparison result, a steering motor 28 connected to the front wheels 17 of the self-propelled vehicle is driven. The steering angle of the front wheels 17 by the steering motor 28 is detected by a steering angle sensor 15 provided on the front wheel of the self-propelled vehicle 1 and fed back to the steering section 14 . Drive control section 1
8 starts and stops the engine 19, and the engine 19
The operation of the clutch 20 that transmits the power to the rear wheels 21 is controlled.

Furthermore, the following functions are added to improve the identification accuracy of reference points. That is, a value indicating the angular range set in the range setting section 26 is supplied to the azimuth angle identification section 24, and it is determined whether the light detected in the direction closest to the predicted azimuth is detected within the planned angular range. A function is added to identify whether or not the

As a result of the identification by the identification function, if the light detected in the direction closest to the predicted azimuth is within the planned range, the aperture angle is calculated using the azimuth, and if it is outside the planned range. , the aperture angle is calculated using the predicted azimuth calculated by the azimuth angle prediction calculation unit 27.

Either the light from the angular range set in the range setting unit 26 is determined to be reflected light from the scheduled reference point and the aperture angle is calculated using the azimuth of the light, or the azimuth of the incident light is within the range. Whether or not the opening angle is calculated using the azimuth determined without determining whether it is within the range set in the setting section 26 is necessary depending on the type and type of work performed by the self-propelled vehicle 1. It may be selected arbitrarily depending on the degree of the system to be implemented.

In this embodiment, the azimuth angle detection section 11, the azimuth angle storage section 12, the identification timing generation section 23, the azimuth angle identification section 24,
Although the range setting section 26, the azimuth angle prediction calculation section 27, etc. constitute the lost sight determining means, the present invention is not limited to this configuration. A simple means for determining loss of sight can also be applied to the present invention (see Japanese Patent Application No. 63-262192).

Next, the functional configuration for processing when reference points are frequently lost will be explained according to the block diagram shown in FIG.

In the figure, the same reference numerals as in FIG. 1 indicate the same or equivalent parts.

The identification result as to whether the reference point has been detected or lost is supplied from the azimuth angle identification unit 24 to the lost data storage unit 39. The lost sight data storage unit 39 stores the identification results of detection or lost sight for all reference points identified while the optical scanner 2 rotated at a predetermined number of revolutions in the past. When the latest data is supplied to the lost data storage section 39, the oldest stored data is erased, and the planned number of data is always stored.

Of the data stored in the lost data storage section 39, only the number of lost data items is added up by a lost number counting section 40, and the result is transferred to a lost number comparison section 41.

The lost data number comparison unit 41 compares the supplied number of lost data and the number set by the scheduled number setting unit 42, and if the number of lost data is greater than the scheduled number, the rotation speed and automatic rotation of the optical scanner 2 are changed. The scanning control unit 5a of the motor 5 sends a signal for setting the traveling speed of the traveling vehicle 1 to a low value.

and a selector switch 43.4 for the drive control unit 18.
4.

On the other hand, if the number of lost data is smaller than the planned number, a signal is sent to each control unit to set the rotational speed of the optical scanner 2 and the traveling speed of the self-propelled vehicle 1 to high values (normal values). The signals are sent out via changeover switches 43 and 44, respectively.

Note that the changeover switches 43 and 44 are for the optical scanner 2.
This is used when only one of the rotational speed of the vehicle 1 and the traveling speed of the self-propelled vehicle 1 is changed, and may be arbitrarily selected according to the priority depending on the work type.

The basic principle for detecting the position and traveling direction of the self-propelled vehicle 1 in this embodiment with the above configuration will be explained. 13 and 14 show the positions of the self-propelled vehicle 1 and the reflector 6 in a coordinate system for indicating the working range of the self-propelled vehicle 1. FIG.

In FIGS. 13 and 14, reference points A, B, and C are where reflectors 6a to 6c are placed, respectively.

The position of the self-propelled vehicle 1 is represented by an xy coordinate system with reference point B as the origin and a straight line connecting reference points B and C as the X axis.

As can be seen from the figure, the position T of the self-propelled vehicle 1 is located at the triangle A
It exists on the circumcircle of TB and at the same time on the circumcircle of triangle BTC. Therefore, the position of the self-propelled vehicle 1 is determined by calculating the two intersections of the circumscribed circles Q and P of the triangle ATB and the triangle BTC, respectively.

As shown in the figure, the position of the self-propelled vehicle 1 can be determined by setting the reference point B, which is the intersection of one of the circumscribed circles Q and P, as the origin, and calculating the other intersection T of the circumscribed circles Q and P according to the following procedure. .

The calculation formula for determining the position of the self-propelled vehicle 1 according to the basic principle is disclosed in Japanese Patent Application No. 116689/1989 and Japanese Patent Application No. 14/1983.
Since the details are shown in No. 9619, the details will be omitted.

Further, the traveling direction of the self-propelled vehicle 1 is calculated using the following formula. In FIG. 14, the angle between the traveling direction of the self-propelled vehicle 1 and the X-axis is θf, and a reference point A based on the traveling direction,
The azimuth angles of B and C are θa.

When θb, θC, θf- 360°-tan' (y/ (xc-x)1-θC・
...(1) becomes.

The position and traveling direction of the self-propelled vehicle 1 are calculated by the position and traveling direction calculating section 13 using the above-mentioned calculation formula and the above-mentioned calculation formula (1).

Next, steering control for controlling the traveling direction of the self-propelled vehicle 1 based on the position information of the self-propelled vehicle 1 calculated by the above procedure will be described. FIG. 12 is a diagram showing the positional relationship between the traveling course of the self-propelled vehicle 1 and a reference point, and FIG. 3 is a flowchart of steering control.

In Fig. 12, reference point B is taken as the origin, and reference points B and C are shown.
The position of the self-propelled vehicle 1 and the work area 22 by the self-propelled vehicle 1 are shown in a coordinate system whose X axis is a straight line passing through the .

Point R (Xret, Yret) indicates the return position of the self-propelled vehicle 1, and has coordinates (Xs t, Ys t), (Xst.

The area connected by the points indicated by Ye), (Xe, Yst), and (Xe, Ye) is the work area 22.

Here, the position T of the self-propelled vehicle 1 is indicated by (Xp, Yp).

Although FIG. 12 shows an example in which the work area 22 is parallel to the X-axis or the y-axis to simplify the explanation, there are reference points A around the work area 22.

As long as B and C are arranged, the shape of the work area 22 and the orientation of the four sides of the work area 22 are arbitrary.

The control procedure will be explained according to the flowchart in FIG.

First, in step S1, the self-propelled vehicle 1 is moved from point R to the work start position by radio control.

In step S2, Xs is set as the X coordinate Xn of the driving course.
Set t and determine the driving course.

In step S3, the self-propelled vehicle 1 starts running.

In step S4, it is determined whether the light receiver 3b has received light from the reference point or another light source. If light is detected, the process proceeds to step S5, where a light reception process to be described later is performed, and if no light is detected, the process proceeds to step S6.

In step S6, it is determined whether or not it is time to perform a reference point identification process to determine which of the received incident lights is from a scheduled reference point. This determination is made based on whether the scanning has progressed to any of the reference point identification azimuths pa to pd that are advanced by a predetermined angle θh from the predicted azimuths θpa to θpd.

Steps S4 to S until the judgment in step S6 becomes affirmative
Step 6 is repeated, and if the determination is affirmative, the process proceeds to step S7, where a reference point identification process, which will be described later, is executed. Once the azimuth of the scheduled reference point is determined by the reference point identification process, the process proceeds to step S8.

In step S8, arithmetic processing for determining a frequent loss of sight, which will be described later, is performed.

In step S9, it is determined based on the arithmetic processing as to whether or not the reference point is being lost more frequently than a predetermined value.

If the determination in step S9 is affirmative, the process proceeds to step S10, where the rotational speed of the optical scanner 2 and the traveling speed of the self-propelled vehicle 1 are set to low values.

If the determination in step S9 is negative, the process proceeds to step Sll, where the rotational speed of the optical scanner 2 and the traveling speed of the self-propelled vehicle 1 are set to normal high values.

Note that step S10. In Sll, step S
Even if the value is not set to change both the rotational speed and the running speed according to the judgment in step 9, a value can be set to change at least one of the rotational speed and the running speed as described above. Good too.

In step S12, the self-propelled vehicle weight position T (Xp.

Yp) and the traveling direction θf are calculated.

In step S13, the amount of deviation from the driving course (ΔX=
Xp-Xn, Δθf) is calculated, and in step S14, steering angle control is performed in the steering section 14 according to the calculated deviation amount.

In step S15, it is determined whether the self-propelled vehicle 1 is traveling in a direction away from the origin (going direction) or in a direction approaching the origin (returning direction) in the y-axis direction.

If it is the forward direction, it is determined in step S16 whether the -stroke has ended (Yp>Ye); if it is the return direction, it is determined in step s17 whether the -stroke has ended (Yp>Ye).
<Yst) is determined. If it is determined in step S16 or S17 that the -stroke has not been completed, the process returns to step S4.

If it is determined in step 816 or S17 that the -stroke has been completed, then in step s18 it is determined whether all the strokes have been completed (Xn>Xe-L).

If the entire stroke has not been completed, the process moves from step S18 to step S19, and U-turn control of the self-propelled vehicle 1 is performed. The U-turn control is different from the straight-line steering control performed by the processing of steps 512 to S14 in which the position information of the self-propelled vehicle 1 calculated by the position/direction calculation unit 13 is fed back to the steering unit 14. It is done by method.

That is, during the turning stroke, the self-propelled vehicle 1 travels with the steering angle fixed at a preset angle.

Then, each reference point A, B, C for the self-propelled vehicle 1
When at least one of the azimuths matches a predetermined angle or falls within a predetermined angle range, the turning is stopped and the steering control returns to straight-line travel.

In step S20, Xn is set to Xn+L, and the driving course for the next stroke is set. Once the driving course is set, the process returns to step S4 and the above process is repeated.

When the entire stroke is completed, the return position R (Xret.

Yret) (step 521), and the traveling is stopped (step 522).

Next, the light reception process and reference point identification process in steps S5 and S7 will be explained.

A flowchart of the light reception process is shown in FIG. In the figure, in step 55 (l), the light reception flag is set to "1°" in order to memorize that light has been detected.

In step 351, the azimuth of the detected light source is stored in the azimuth storage section 12.

A flowchart of the reference point identification process is shown in FIG. This flowchart shows an example of a procedure for further narrowing down the detected azimuth angle closest to the predicted azimuth angle using a scheduled angle set in the range setting section 26 to identify a reference point.

In the figure, in step S70, a value n of a pole counter (hereinafter simply referred to as a pole counter) for distinguishing a reference point to be identified is incremented. The pole counter is associated with each reference point. That is, the pole counter "l" is at the reference point A, and the pole counter "2" is at the reference point A.
corresponds to reference point B, and pole counter "3" corresponds to reference point C, respectively.

If the initial value of the poll counter is “0”, step 5
Through the process 70, the poll counter becomes "1", and the reference point corresponding to this becomes "A". In this embodiment, the initial value is set to "0".

In step 571, the light reception flag is determined. If the light reception flag is "1", the process proceeds to step S72, and if the light reception flag is "0", the process jumps to step S79.

In step S72, the predicted azimuth θpn is selected from among the azimuths of the light source stored in the azimuth storage unit 12 (since the pole counter is set to “1”, the predicted azimuth θp
Assume that the one closest to a) is the azimuth angle of the scheduled reference point, and store that value as the angle θS.

In step S73, the presence or absence of noise is determined by determining whether the number of received lights is "2" or more, that is, whether a plurality of azimuths are stored in the azimuth storage section 12.

If the determination in step S73 is affirmative, it is assumed that one or more noises have been detected, and the process moves to step S74, in which the fact that noise has been detected is stored as noise processing. This stored data will later provide clues about the state of the work environment, making it easier to take measures such as eliminating noise sources.

If the determination in step S73 is negative, the process proceeds to step S75, where the temporarily determined azimuth θS and the predicted azimuth θp are
It is determined whether the difference with n (predicted azimuth θpa) is smaller than the angle θh. If the error is larger than the angle θh, it is determined that the temporarily determined azimuth θS is not the azimuth of the planned reference point but the azimuth of the noise source, and the process proceeds to step 378, where the same noise as in step S74 is detected. Perform processing.

After the noise processing, the process proceeds to step S79, where the predicted azimuth θpn (predicted azimuth θpa
) is set as the azimuth angle θn (θa) of the scheduled reference point.

In step S80, data indicating that the reference point has been lost is set in the lost data storage section 39. Details of the lost data set will be described later.

On the other hand, if the difference (θS - θpn) is smaller than the angle θh, the assumed azimuth θS is determined to indicate the azimuth of the scheduled reference point, and the process proceeds to step S76.

In step S76, the azimuth θ determined in the previous process is
Based on n and the azimuth θS determined in the current process, the predicted azimuth at which the same reference point should be detected in the next process is calculated using the calculation formula (θs+(θS−θnN).

In step S77, the azimuth angle θn is updated by the angle θS.

In step S81, the next reference point identification direction pn+1(
That is, the predicted azimuth θpn+1 calculated when the reference point B was detected during the previous scan as the reference point identification azimuth pb)
The angle obtained by adding the planned angle θh to is set.

In step S82, the light reception flag is reset.

In step S83, the data stored in the azimuth storage section 12 is erased.

In step S84, the lost data stored in the lost data storage section 39 is shifted.

Details of the lost data shift will be described later.

In step S85, it is determined whether the pole counter is "3#" or not. The numerical value "3" is the number of installed reference points, and the numerical value is set according to the number of installed reference points.

If the number of installed reference points matches the pole counter, the pole counter is set to "0" in step S85 and the process returns to the main routine (processing in FIG. 3).

When the pole counter is "1", in the next process, the pole counter is incremented to "2" in step S70, and the reference point B identification process is performed.

Thereafter, identification processing for the reference point C is also performed in the same manner.

Next, the lost data set will be explained.

FIG. 6 is a flowchart of a lost data set.

Before entering into the explanation of the flowchart, the storage format of the lost data storage section 39 will be explained.

FIG. 10 is an example of lost and stored data. In the figure, the identification result of M(p, i) is stored as stored data. The variable p indicates the pole counter, that is, the distinction between the reference points A and C, and the variable i indicates how many times the stored data M is lost data during the previous scan. gc! Memory data "1" indicates that the reference point has been lost, and memory data "0" indicates that the reference point has been detected.

In this embodiment, settings are made so that lost data up to the 20th previous scan can be stored. From the data example shown in Figure 10, reference point A (pole counter 1) is set 3 times before and 19
It can be recognized that the reference point C (pole counter 3) was lost during the previous scan, and that the reference point C (pole counter 3) was lost during the previous scan.

Now, in the flowchart of FIG. 6, step 58
At step 00, a pole counter is set as a variable p indicating which reference point is currently being performed the reference point identification process.

In step 5801, data "1" indicating that the reference point has been lost is stored as stored data M(p, 0).

That is, the stored data M(p, O) indicates the latest detection data from the zeroth previous scan of the reference point indicated by the pole counter p, and it is stored that the scheduled reference point was lost in the latest scan. be done.

Next, the operation of lost data shift will be explained. FIG. 7 is a flowchart of lost data shifting.

In the figure, in step 5840, a pole counter is set as a value p indicating which reference point is currently being performed the reference point identification process.

In step 5841, the stored data M (p.

Clear the value i of i).

In step 5842, the value i is incremented.

In step 5843, stored data M(p, i) is converted into stored data (p, 1-1). That is, the data from the i-th previous scan is converted into the data from the next scan (1-1).

In step 5844, it is determined whether the value i is "20". Steps 5842 and 8 until the value i becomes "20'"
The process of 843 is repeated to update the data by shifting all the lost data of one reference point forward one time at a time.

In step 5845, data "0" is stored as storage data M(p,0).

Through the process of step 5845, the stored data M(p, 0
) is set to the value "0°," so if the scheduled reference point is detected in the next process, this stored data M(p
, 0) is treated as the latest value. Then, only when the reference point is lost, the value of M(p, o) of this stored data is converted to "1" as shown in step 5801.

Next, the determination of frequent loss of sight based on the stored data will be explained. FIG. 8 is a flowchart for determining the frequency of losing sight.

In the figure, in step S90, stored data M(1
, 1) to M(1,20> to obtain data M1, and add the values of stored data M(2, 1) to M(2, 20) to obtain data M2. (3,1)~M
Data M3 is obtained by adding the values (3, 20).

In step S91, the total value of the stored data M1 to M3 is set to the value LK set in the planned number setting section 42.
Determine whether it is greater than or not. Memory data M from value LK
If the total value of 1 to M3 is large, the frequent loss of sight flag is set as “
If the total value of the stored data M1 to M3 is smaller than the value LK, the frequent loss flag is set to "0" (step 593).

As described above, in this embodiment, lost reference point data, that is, identification data as to whether the reference point was detected or lost, is stored for all reference points for the number of past scheduled scans. If it is determined based on the stored data that more points have been lost than expected, the rotational speed of the optical scanner 2 is reduced, and the speed of the self-propelled vehicle 1 is reduced to create a state in which it is easier to detect the reference point. I decided to do so.

(Effects of the Invention) As is clear from the above description, according to the present invention, the following effects can be obtained.

(1) If the reference point is frequently lost, the rotational speed of the optical scanner is automatically reduced, so the effective horizontal distance between the vertically swinging light beams can be reduced. The probability of irradiating a point with a light beam can be increased.

(2) If the reference point is frequently lost, the running speed of the self-propelled vehicle is automatically reduced, which increases the sway of the self-propelled vehicle on uneven driving surfaces where it is easy to lose sight of the reference point. The probability of irradiating the light beam to the reference point can be increased.

(3) As described above, by reducing the scanning speed or traveling speed only under conditions such as conditions of the traveling surface where loss of sight occurs frequently, it is possible to reduce loss of sight of the reference point without substantially reducing travel work efficiency. can.

(4) Since the rotational speed of the optical scanner or the traveling speed of the self-propelled vehicle is changed only in situations where the reference point is frequently lost, it is necessary to improve the mechanical tracking performance of the optical scanner, etc. Little complexity in the configuration is required.

[Brief explanation of drawings]

Figures 1 and 2 are block diagrams showing the functions of an embodiment of the present invention, Figure 3 is a flowchart of steering control, Figure 4 is a flowchart of reference point identification processing, and Figure 5 is a flowchart of light reception processing. , FIG. 6 is a flowchart of a lost data set, FIG. 7 is a flowchart of a lost data shift, FIG. 8 is a flowchart of determining a frequent loss of sight, and FIG. 9 is a flowchart of a lost data set.
The figure is a sectional view of an optical scanner, Figure 10 is a diagram showing an example of reference point identification data, Figure 11 is an explanatory diagram of reference point identification processing, and Figure 12 is a driving course of a self-propelled vehicle and the arrangement of reflectors. 13 is a diagram explaining the principle of detecting the position of a self-propelled vehicle, FIG. 14 is a diagram explaining the principle of detecting the traveling direction of a self-propelled vehicle, and FIG. 15 is a diagram showing the arrangement of the self-propelled vehicle and the reflector. FIG. DESCRIPTION OF SYMBOLS 1... Self-propelled vehicle, 2... Optical scanner, 2a... Fixed part, 2b... Rotating part, 3a... Light emitter, 3b...
・Light receiver, 6° 6a to 6C...Reflector, 7a...Slit plate, 7b...Angle sensor, 11...Azimuth angle detection section, 12...Azimuth angle storage section, 13...・Position/progressing direction calculation unit, 23...Identification timing generation unit, 24・
... Azimuth angle identification section, 26... Range setting section, 27...
Azimuth angle prediction calculation unit, 30...Mirror, 39...Loss of sight data storage unit, 40...Loss of sight counting unit, 41...
・Loss of sight comparison section, 42... Planned number setting section

Claims (6)

[Claims] (1) A light beam is scanned in the circumferential direction while swinging vertically around the self-propelled vehicle, and the light beam is reflected from light reflecting means arranged at at least three reference points distant from the self-propelled vehicle. A steering control device that receives the reflected light of the light beam, measures the azimuth of the light reflecting means as seen from the self-propelled vehicle, and causes the self-propelled vehicle to travel along a planned travel course based on the results. a reference point loss determining means for determining whether or not the light reflecting means has been detected based on the presence or absence of reflected light from a direction expected to be received by the self-propelled vehicle; and an output from the determining means. Steering of a self-propelled vehicle, comprising means for reducing at least one of the circumferential scanning speed of a light beam and the traveling speed of the self-propelled vehicle when the frequency of lost-of-track signals exceeds a predetermined frequency. Control device. (2) When the lost-of-place signal outputted based on the determination by the lost-of-place determination means returns to a frequency lower than the scheduled frequency, based on the reduced circumferential scanning speed of the light beam and the traveling speed of the self-propelled vehicle. The steering control device for a self-propelled vehicle according to claim 1, further comprising means for returning the vehicle. (3) The reference point loss determining means includes means for detecting the azimuth of each light reflecting means as seen from the self-propelled vehicle based on the reflected light, and based on the azimuth detected by the azimuth detecting means. azimuth prediction means for calculating the azimuth at which each light reflection means should be detected in the next scan; It is characterized by comprising a reference point identification means for identifying whether the light is reflected by the reflecting means or not, and a means for outputting a lost signal when the intended light reflected from the light reflecting means is not detected by the identification means. The steering control device for a self-propelled vehicle according to claim 1 or 2. (4) The reference point identification means determines the immediately preceding reference point identification direction and Among the incident light detected between the current reference point identification azimuth, the incident light from the angle closest to the predicted azimuth is assumed to be the reflected light from the light reflecting means placed at the planned reference point, and this 4. The steering control device for a self-propelled vehicle according to claim 3, further comprising means for treating the object as an object to be identified by said identifying means. (5) Only when the incident light from the angle closest to the predicted azimuth is light from an angular range established with the predicted azimuth as a reference, the incident light is placed at the predetermined reference point. 5. The steering control device for a self-propelled vehicle according to claim 4, further comprising means for identifying that the light is reflected from a reflective means. (6) comprising means for storing only the azimuth of light detected within an angular range set with the predicted azimuth as a reference, and the azimuth stored in the storage means is used as the reference point identification azimuth; 6. The steering control device for a self-propelled vehicle according to claim 5, wherein the steering control device is treated as an object to be identified.