JPS63273916A - Driving course forming device - Google Patents

Abstract

PURPOSE:To enable an unmanned vehicle to reach a target point with no disturbance due to an obstacle by confirming the positions of the target and obstacles within a driving area, then forming a driving course including no obstacle between a starting point and the target point. CONSTITUTION:A recognizing means 1 recognizes a target and obstacles out of the photographed picture data based on the colors designated by the color information and the extraction of features. Then the position information on the target and obstacles are obtained and applied to a course forming means 2. The means 2 forms a driving course from said position information so that an unmanned vehicle can move up to a target point from a starting point with no disturbance of obstacles. As a result, the vehicle can travel up to the target point without stopping for recognition of obstacles since said course is free from obstacles. Thus the driving time of the vehicle can be shortened up to the target point and therefore the working efficiency of the vehicle is improved.

Description

[Detailed Description of the Invention] [Summary] Conventional unmanned vehicles travel according to fixed travel course data input by the user or fixed travel course data pre-installed in the device main body. Therefore, if there is an obstacle on the travel course while traveling along the travel course, the vehicle must stop and create a new route to avoid the obstacle, which increases the travel time. Was.

The present invention is designed to recognize obstacles within a travel area and create the most efficient travel route (shortest distance travel route, shortest time travel route, etc.) that allows travel to a target point while avoiding obstacles. It is highly practical as it can create the most efficient driving route according to the environment of the driving area.

[Industrial application field]

TECHNICAL FIELD The present invention relates to unmanned vehicles, and particularly to an apparatus for creating a driving route for unmanned vehicles.

[Traditional technique]

Transport robots are used for applications such as remote work in nuclear couplants and unmanned transport in factories and hospitals. II like this! The robot robot, ie, the unmanned vehicle, has a built-in driving course creation device that determines whether it is passable or not.

Here, the operation of a conventional travel course creation device and the operation of an unmanned vehicle incorporating the travel course creation device will be described with reference to the drawings.

FIG. 7 is a flowchart showing the flow of the operation of a conventional driving control creation device, and FIG. 8 is a diagram specifically explaining the above operation.

First, in the process SDI shown in FIG. 7, before traveling, the user inputs position information, etc. of the driving course C1 from the starting point S+ to the target point G1. In addition, if the data of the driving course C1 is stored in the creation device in advance, this process ISDI is not performed.

Next, the unmanned vehicle starts traveling based on the data of the traveling course C1. At this time, the unmanned vehicle uses the data obtained from the TV camera and the driving course C using the ultrasonic sensor.
Each time the vehicle travels a predetermined distance, the vehicle stops to determine whether there are any obstacles within the vehicle (processing 5D2). At this time, if the driving course creation device determines that there is no obstacle within the driving course C1, the unmanned vehicle can reach the target point G1 (processing 5D4).

On the other hand, if it is determined by the determination process SD2 that the obstacle D+ (see FIG. 8) is within the driving course C+, the driving course creation device creates an avoidance route C2 (processing 5D3).
. The unmanned vehicle can reach the target point G1 by temporarily traveling on the avoidance route C2 and traveling on the driving course C1 again (processing 5D4).

Figure 7 shows the operation of a driving course creation device with an avoidance route creation function that creates a route that avoids obstacles when there is an obstacle in the driving course while driving on a predetermined driving course. However, in the case of a driving course creation device that does not have the above-mentioned avoidance route creation function, the unmanned vehicle will stop until the obstacle is removed.

[Problem that the invention seeks to solve]

However, as mentioned above, conventional driving course creation devices either have pre-fixed driving course data built-in or store driving course data entered by the user at the time of startup, so the driving area before driving can be changed. The environment is not taken into consideration, and there may be obstacles on the driving course while driving. For this reason, conventional driving course creation devices need to stop once every predetermined distance to check whether there are any obstacles within the driving course, which increases driving time and reduces work efficiency. It had the following drawback.

SUMMARY OF THE INVENTION In view of the above-mentioned drawbacks of the conventional art, it is an object of the present invention to provide a highly practical driving course creation device that can create a driving course that allows the vehicle to travel to a target point without being obstructed by obstacles.

[Means for solving problems]

FIG. 1 is a functional block diagram of the present invention.

1, for example, recognition means for recognizing position information of targets and obstacles from image data obtained by imaging with a television camera, etc.;
Reference numeral 2 denotes a route creation means for creating a travel course in which the unmanned vehicle can travel from a starting point to a destination point without being obstructed by obstacles, based on the position information of the target and the position of obstacles added from the recognition means 1. be.

[For production]

The recognition means 1 recognizes targets and obstacles from image data obtained by imaging with a television camera or the like. Recognition of the target is
For example, this is performed by extracting image data of the specified color from the image data based on color information of a target object specified by operating an external switch. Further, the recognition of the obstacle is performed, for example, by extracting only one image data of the obstacle using a technique such as feature extraction based on the image data. After recognizing the targets and obstacles as described above,
Position information of targets and obstacles is obtained and added to the route creation means 2. The route creation means 2 is configured to move the unmanned vehicle from the starting point to the target point (a point where the target object can be grasped by a manipulator or the like) without being obstructed by obstacles, based on the position information of the target objects and obstacles. To create a driving course that can be driven without being hit. The above driving course is a driving course that is not obstructed by obstacles, and while the unmanned vehicle is driving to the target point, it is not necessary to stop once to recognize obstacles as in the past. , the travel time to the target point will be significantly shortened and the work efficiency of unmanned vehicles will be improved.

〔Example〕

Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 2 is a system configuration diagram of an embodiment of the present invention, and FIG. 3 is a system configuration diagram showing FIG. 2 in more detail.

The general controller 1-roller 10 controls the entire operation of the unmanned robot 7 according to the embodiment of the present invention and sets the running course, and as shown in FIG. 3(b), the processor 10-1
, serial input/output (SIO) 10-2 to 10-6, and parallel input/output (PIO) 10-7. Although not shown, the processor 10-1 internally includes a microprocessor unit MPU, a read-only memory (ROM) that stores programs for performing the above-mentioned control and other processing,
These MPUSROMs have random access memory (RAM) and the like to store data during various types of processing.

The RAMs are connected by pass lines. Also, the above-mentioned pass lines are 31010-2 to 10-6, P 101
0-7 are also connected in common. When the embodiment of the present invention starts operating, the overall controller 10 waits for control information such as commands to be added from an input device. PIOlo-7 of the general controller 10 is connected to various switches 11-1 to 11-11 of the switch unit 11, and manages the contact states, that is, on and off states of the switches 11-1 to 11-11. Switch unit 11 is Go
Switch 11-1, RESET switch 11-2, audio E/D switch 11-3, red Paul switch 11-4,
Blue pole switch 11-5, yellow ball switch 11-6
, green pole switch 11-7, far ball switch 11-
8, Near pole switch 10-9, Shape Eswitch 11-
10, consists of two shape switches 11-11, each switch 11-1 to 11-11 is a PIO of the central controller 10.
Connected to lo-7. For example, GO switch 11-1
11-1 is a switch for starting the operation of the unmanned robot of this embodiment, and by turning on this switch 11-1, operations such as recognition of a target object, setting of a running course, and handling, which will be described later, are started. switch 11-4
~11-7 is a switch for inputting the color designation of the target object (balls of each color in the embodiment of the present invention), and when the red ball switch 11-4 is turned on, the target object is a red ball. . The far ball switch 11-8 and the near pole switch 11-9 are used to determine whether, when the color of the target object is specified and there are multiple balls of that color, the farther ball of the same color should be the target object or the closer ball of the same color should be the target object. This is a switch that indicates whether the colored ball should be the target object. Further, the shape 1 switch 11-10 and the shape 2 switch 11-11 are switches for instructing, for example, the size and shape of the ball. This allows you to specify one of the balls even when there are many balls. Further RE
The SET switch 11-2 is a reset signal for this device, and by turning on this switch, if it is in operation, it stops the operation. Also, the audio E/D switch 11
-3 is a selection switch for enabling voice input, which will be described later. When the various switches 11-4 to 11-11 described above are turned on after the general controller 10 starts operating (the Go switch 11-1 is turned on), the general controller 10 instructs the image processing unit 12 to detect the target object. start the recognition operation.

The image processing unit 12 includes a camera 12-1, a camera controller 12-2, a color decoder 12-3, an image processing device 12-4, a camera tilter 12-5, and a camera tilter driver 12-6. The camera 12-1 is installed in the camera tilter 12-5, and the general controller 10 via the camera tilter driver 12-6 before starting the target object recognition operation so as to image the target object.
The camera tilter 12-5 is controlled so that the target object is within the imaging range of the camera.

On the other hand, the image data of the camera 12-1 is stored in the camera control.
- camera 12-2, for example camera 12-1
The image data obtained is converted into multivalued digital image data by the camera controller 12-2. The digital image data is then decoded by a color decoder 12-3 and applied to an image processing device 12-4. The decoded digital image data is added to the image processing device 12-4 as described above, but the decoded digital image data is stored in the internal image memory only when a target object recognition request is received from the general controller 10. Take in.

In the embodiment of the present invention, one of the four colors of balls is designated, so the image processing device 12-4 selects one color from among the image data of each color corresponding to the designated ball color. Select image data. In general, unmanned driving 1 unit
For example, in a factory, there are processes that generate various types of light within the range of movement J that ~ moves, and the level of image data differs depending on the reception of this light. In the embodiment of the present invention, although not shown in the drawings, the light illuminates the range of the 1R image of the camera 12-1, and in order to prevent malfunctions caused by this external light, the data captured by the camera 12-1 is Incorporate twice.

Then, when the levels of the image data of these two times match, it is assumed that the image data is normal and target object recognition processing is performed.

Furthermore, in the embodiment of the present invention, target object recognition processing is performed using color data. When recognition processing is performed using black and white data, images in the distance and nearby images have different brightness, so for example, it is possible to perform m i16 only in the vicinity of the camera, but in the present invention, using a specific color, Therefore, even if the levels of near and far image data obtained by the light are different, it is possible to clearly recognize which object is the target.

When the image processing device 12-4 recognizes the object, ii!
ii The image processing device 12-4 converts the image data and the position information of the target object into serial data and sends it to the general controller 1.
0 to processor 10-1 via SIOI O-4.

Since the position of the target was recognized by the above-mentioned operation,
Next, the general controller 10 creates a travel course to the target object from the image data obtained from the image processing unit 12.

Hereinafter, the operation of creating a travel course in this embodiment performed by the general controller 10 will be explained with reference to the flowchart shown in FIG.

First, image data and target object position information are input from the image processing device 12-4, and obstacle position information is determined from the image data (processing 5TI).

The process ST1 will be explained in detail below. Processor 10-
1 stores image data input from the image processing unit 12 via 5IO10-4 in a RAM (not shown).

Then, when all the image data is input, feature extraction processing is performed from the image data stored in the RAM, and all obstacles are extracted by comparison processing with the feature data of obstacles set in advance. The position information and shape data of are determined and stored in RAM.

Then, based on the position information of the target and the position information and shape data of each obstacle obtained as described above, the following processing is performed to create a driving course to the target.

In an embodiment of the invention, the outer shape of the body is rectangular. Further, as will be described later, the trackless traveling unit 13 moves the main body to a target position by running in a straight line and rotating. The central axis of rotation in this trackless traveling unit 13 is not the center of gravity of the rectangle, but is shifted in the longitudinal direction of travel. For this reason, in the embodiment of the present invention, in the process of creating a running course in the general controller 10 described above, the area from the start point to the target is defined as the required range on the left and right when traveling in a straight line, and the required range when rotating. We are looking for a course that can be divided into two areas.

First, it is determined whether an obstacle exists on the straight line from the starting point to the target object (processing 5T2). If there are no obstacles, it is possible to reach the target by just driving in a straight line, so a straight route to the target point is generated (processing 5T4), but if there are obstacles, they must be avoided.

In this case, a straight trajectory is determined on which the trackless traveling unit 13 can travel in contact with each obstacle. The trajectory at this time is the vIl'L path up to the point where it passes the obstacle until it can rotate, and this point is taken as the first candidate node. Then, the first candidate node that has the minimum sum of the distance from the starting point to the first candidate node and the straight-line distance assuming that there are no obstacles from the first candidate node to the target is selected as the first candidate node. The node (obstacle avoidance point) is determined (processing 5T3). Then, the trajectory is determined by repeating the operations ST2 to ST3 again using the determined first node as a starting point.

By repeating the processes ST2 and ST3, obstacle avoidance points are set one after another, and a straight path from the obstacle avoidance point to the target is found in process ST2, thereby creating a course (i.e., a trajectory) from the starting point to the target. ) is determined (processing 5T4).

The above-mentioned method is a course determining method that takes rotation and straight-line travel into consideration, but it is also possible to determine a course with the shortest travel distance or shortest travel time.

For example, all the courses that avoid obstacles are found (if there are multiple nodes, all the courses are found), and then the course with the minimum travel distance is determined from among the courses. If the vehicle takes a long time to rotate, etc., the sum of the time required for straight-line travel and the time required for rotation is determined, and the course that provides the minimum value is determined as the course with the shortest travel time.

After the general controller 10 determines the driving course as described above, the general controller 10 transfers the data of the determined driving course to the non-IFlti driving unit I3.
Send to. The trackless traveling unit 13 has a GLV: 27 to C'-ra 13-1, and the trackless traveling unit 13 operates under the control of this GLV controller 13-1.

The GLV controller 13-1 has the same configuration as the general controller 0, and includes a processor 13-3-5, serial input/output (SIO) 13-3-1, and parallel input/output (PI).
O) 13-3-2 to 13-3-4, which are similarly connected by a bus (not shown). The processor 13-3-5 has a microprocessor unit M inside.
P.U.

A read-only memory (ROM) that stores control programs for each device that makes up the trackless traveling unit 13, and a random access memory that stores data during various processes.
(RAM), etc., and these MPU, ROM, RAM
5S I Ol 3-3-1, PIO13-3-2~1
3-3-4 are directly connected by a pass line. When the above-mentioned general controller 10 joins the GLV controller 13-1, it is taken into the processor 13-3-5 via the 31013-3-1, and the running control corresponding to the running course data is executed by the processor 13-3-5. 5 does. First, control is performed to drive the trackless traveling unit 13 toward the first node in the necessary travel direction and distance.

The trackless traveling unit 13 outputs motor drive data to the D/A converter 13-2 via P 1013-3-2.

This drive data is a digital signal, which is converted into an analog signal by the D/A converter 13-2, and further applied to the drive controller 13-4 via the analog switch 13-3. Drive controller 13-4
is a circuit that drives the motor (L) 13-13 and the motor (R) 13-14. This motor 13-13.13-1
4 are mechanically connected to wheels (not shown), and the wheels are rotated by rotation of the motor. In the embodiment of the present invention, when the left and right wheels are rotated in the same direction, the trackless traveling unit 13 moves forward or backward, and when the left and right wheels are rotated in opposite directions, the trackless traveling unit 13 rotates. I do. Incidentally, the central axis of this rotation is the center of gravity position of the traveling unit determined at the time of creating the aforementioned traveling course.

The D/A converter 13-2 is provided with two channels so as to generate two analog signals for rotating the left and right wheels. In other words, D/A Combark 13-2
Control data to rotate the left and right wheels are independently added to the control data, and each outputs an independent analog signal.

In the embodiment of the present invention, the information on the joystick 13-5 is transferred to the arithmetic circuit 1 so that the main unit can also be moved by operating the joystick 13-5.
In addition to 3-6, an analog signal for controlling the motor is input to the drive controller 13-4 based on the information. D/
A calculation circuit 13 selects the output of the A converter 13-2.
The signal that controls whether to select the output of -6 is PIO13-
It is output from 3-2, and analog switch 13-3
When a positive control signal is applied to , the output of the D/A converter is selected and output to the drive controller 13-4,
When the negative output is inverted via the inverter 13-7 and applied as a positive signal to the analog switch 13-8, the analog control signal output from the arithmetic circuit 13-6 is passed through the analog switch 13-8 to the drive controller 13-8. Join 4. Note that a selection signal for automatic movement or movement by joystick operation is output from the processor PIO 13-3-2, and this is an automatic/manual changeover switch 13- of the trackless traveling unit 13.
It can be switched in the on/off state of 9. The on/off signal of automatic/manual changeover switch 13-9 is PIO13-3-4
each time it is added to processor 13-3-5 via
Analog switch 13-3.13 whose purpose is this control signal
Turn -8 on and off. Furthermore, in the embodiment of the present invention, starting, stopping, and even braking can be performed manually, and this is done by using the start switch 13.
-10, the stop switch 13-11, and the brake switch 13-12 are turned on and off by being applied to the processor 713-3-5 via the PIO 13-3-4. The drive controller 13-4 also has a control terminal for controlling the start of driving, and outputs the output from the GLV controller 13-1 (via PIO 13-3-2).
, an arithmetic circuit 13-6, and a signal obtained by inverting the drive signal of the brake 13-20 via an inverter 13-21 are applied to the control terminal via an OR gate 13-22, so that when the brake is applied, the drive signal is inverted. The drive is stopped when the target position is reached.

The drive controller 13-14 includes a tachogenerator (L) 13-15 which is mechanically connected to the motor 13-13.13-14 in the same way as the wheels. Tacogen (R) 1
3-16 output is added. This tachogen 13-15.
The output of 13-16 is an analog signal proportional to rotation, and the drive controller 13-4 rotates the motor 13 so that the wheels rotate at a constant speed. -13.13-14
control. Through this control, the wheels are rotated at a target rotational speed, and as a result, the traveling unit 13 is moved at a constant speed.

Generally, the wheels may slip, for example with respect to the floor, and this slip increases displacement errors. In order to reduce this movement error, in the embodiment of the present invention, two encoders (measurement M (L
) Encoder 13-17, measuring wheel (R) encoder 13
-18. This encoder 13-17.13
-18 is mechanically connected to a measuring wheel provided near the left and right wheels, for example, and determines the rotation of the measuring wheel.

The encoders 13-17, 13-18 output one pulse at a specific angle of rotation and also output a signal representing the direction of rotation. These pulses and rotational direction signals are applied to counters 13-19, which count the number of pulses output from each encoder. Note that the counters 13-19 are up/down counters, and the rotation direction signal is applied to the up/down control terminal of this up/down counter.For example, it counts down when rotating in reverse and counts up when rotating forward. The distance traveled by each wheel rotation can be determined with high accuracy using the count value. The output of this counter 13-19 is applied to the processor 13-3-5 via the PIO 13-3-3.
calculates the distance traveled by the main body and the current direction of rotation from this value.

The trackless traveling unit 13, which moves according to the traveling course added from the general controller 10 by the above-mentioned operation, has a measuring wheel to measure the amount of travel, etc., and obtains highly accurate movement data with this measuring wheel. However, the error becomes large when moving over a long distance. For this reason, after performing a specific movement, for example, when the target position is reached, the target object from the image processing unit 12 is recognized again, and the traveling error is corrected.

In the above-mentioned driving, the position of the center of gravity of the main unit (
However, when lifting a target object with the manipulator 14-2, the calculation is performed using the position of the manipulator 14-2 as the central axis. That is, since the range in which the manipulator 14-2 can move is limited, the main body is controlled to move the central axis of the manipulator 14-2. With this control, even if a running error occurs, the manipulator 14-2
The target can be lifted by moving.

The manipulator unit 14 is connected to the general controller 10 via 31010-2.

In the above, when the main body moves to the target,
In reality, the vehicle will stop just before a target to avoid hitting it, just like in the case of an obstacle. When this stops, a lifting control signal is applied from the general controller 10 via the 31010-2, and a signal for controlling the movement of the manipulator 14-2 is also applied to drive the manipulator 14-2. The manipulator unit 14 includes a manipulator controller 14-1, a manipulator 14-2, and a grip sensor 14-3. Manipu controller 14
-1 is added with a signal output from the processor 10-1 via the aforementioned SIOI O-2, and this control signal drives the manipulator 14-2. This control causes the manipulator 14-2 to operate and lift the target object. The grip sensor 14-3 of the manipulator unit 14 is a sensor that detects whether or not the target object has been lifted, and when the target object has been lifted, it is output to the processor 10-1 via the PIOIO-7 as detecting the target object. When the target object cannot be lifted due to a positional error, no detection signal is applied from the gripping sensor 14-3 even when a control signal for lifting the manipulator 14-2 is applied, so the processor 10-1 determines the position of the target object again. In other words, the position is corrected and the lifting is controlled again.

With the above operations, the main body that has traveled from the starting position can reach the target object and lift the target object. After this, for example, the position to which the target object should be moved is determined, and the target object can be moved to the desired position by creating the travel course and controlling the movement described above.

In the embodiment of the present invention, in addition to the above-mentioned apparatus, as shown in FIG. 3(b), a voice-input crab knit 15 is provided. R5-232C receiver 15-1 of this unit 15
are provided on the main body, including a microphone 15-2 and a voice recognition device 15.
-3, R3-232C transmitter 15-6 is provided, for example, near an operator or the like. Then, the operator's voice is recorded, and data such as the F indication of each color ball, the front and back of the target, the operation instruction and line input of the switch unit 11 mentioned above, and the desired position to move the target, etc. can be added. Note that the data from the audio input crab unit 15 becomes valid when the audio E/D switch 11-3 is turned on.

Through the above-described operations, in the embodiment of the present invention, the unmanned robot can travel to a target object or a desired position.

Next, the operation of the present invention will be explained in more detail.

FIG. 4 shows a general controller 10 in an embodiment of the present invention.
and is an operation flowchart of the GLV controller 13-1. For example, when the power is turned on, the main controller 1
0 and the GLV controller 13-1 start processing operations.

First, each controller 10.13-1 performs initial setting processing S.
Perform CI and SNI. Then, when a start signal SC2 is input (indicating the color of the target object, turning on the GO switch, etc.), target position recognition processing SC3 is performed. The target position recognition process SC3 is performed by the image processing unit 12 under the control of the general controller 10. When the target position recognition by the image processing unit 12 is completed, the position information of the target object is added to the general controller 10, so next the traveling course creation process S
Perform C4 and create a driving course. Then, a process SC5 is performed to send the result to the GLV controller 13-1, and reception of end data is detected in a travel end reception process SC6.

OL, V': J7 to C)-La13-1 is the initial setting SNI
Thereafter, a driving course reception process SN2 is performed, and when the driving course is transmitted by the above-mentioned process SC5, this data is received and a trackless driving process SN3 is performed. No IPI
The road traveling unit 13 uses this information to perform the aforementioned traveling operation to the target position. When the running operation is completed, the running end signal sending process SN4 is performed, and the overall controller l
Sends a termination signal to O. At this time, the general controller 10 is performing a travel end reception process SC6, and when the GLV controller 13-1 sends an end signal, it can receive the end signal, and performs the next ball (target object) position re-recognition process SC7. This re-recognition process SC7 is performed by the image unit 12 under the control of the general controller 10. Then, from the result obtained by the re-recognition process SC7, a process S for determining whether the current position is within the movable range of the manipulator 14-2
Do step 8.

In this determination, if it is not within the movable range of the manipulator 14-2, a fine movement travel control process SC9 is performed. This control signal is applied to the GLV controller 13-1, and the GLV controller 13-1 performs fine movement processing SN5 to cause the main unit to move slightly. When the general controller 10 finishes the fine movement process SC9, it repeats the process again from the ball position re-recognition process SC7. For example, when the manipulator 14-2 enters the movable range due to the above-mentioned repetition (determination SC8 is Y), the general controller 10 performs handling control processing 5CIO on the manipulator unit 14, and the manipulator 14-2 Perform an action to grasp a target object, such as a ball. Then, the general controller 10 performs a determination 5C1l as to whether or not the manipulator 14-2 has grasped the ball. This determination is made by the grip sensor 14-
This is done by the overall controller 10 determining whether or not the detection signal from 3 is added. In this determination, if the ball is not gripped (N), it is assumed that the ball position is not accurate, and the ball position recognition process S is performed again.
Repeat from C7. When this is repeated, fine movement control processing SC
When step 9 is executed, the GLV controller 13-1 performs a slight movement process SN5 corresponding to that process.

When the ball is grasped, that is, in discrimination 5CII,
When it is determined that the object has been grasped (Y), processing for returning to the starting point is performed.

First, the general controller 10 inputs image data (images in the return direction) from the image processing unit 12, and performs return course creation processing 5C12. In addition, on the way back, it is also possible to determine the return course by finding the route taken to reach the target, that is, the reverse of the going trajectory. Then, transfer the obtained return course data to C, LV Sicon Roller 13.
-1, and the GLV controller 13-1 receives the return course SN6. After this reception is completed, the CLV controller 13-1 performs trackless running processing SN7,
For example, when you go to the target, you return to the starting point on the opposite trajectory. When the vehicle returns, it sends out a travel end signal SN8.

The general controller 10 receives this travel end signal 5N1.
In step 3, in order to end the processing for the target object instruction at the first start signal input SN2, an operation end signal is sent to the GLV controller 13-1 in step 5N14, and the processing is repeated again in step S.
Execute from C2. Note that the process SC2 is an input process,
The system waits until an operator or the like makes an input from the switch unit 11, voice input unit 15, or the like.

On the other hand, when the GLV controller 13-1 also receives the operation end signal from the general controller 10 (SN9), it completes all processing, performs the driving course receiving process SN2 again, and waits to receive the driving course from the central controller 10.

Then, when it joins again, the above-described operation is repeated.

In this embodiment, the positional information of targets and obstacles is recognized by analyzing image data captured by a television camera. It may also be possible to recognize positional information of objects or obstacles. .

Further, the present invention can be incorporated into a tracked unmanned vehicle having a plurality of travel courses from a starting point to a destination point.

A specific example is shown in FIG. In the same figure, the trajectory a~ is
The i&michi for guidance and unmanned vehicles are equipped with electromagnetic pickup, pick-up,
The line sensor detects the guidance signal output from the trajectory a-k and travels to the target point E. At this time, the trajectory f
, t, and g, the present invention can recognize the obstacle, so the trajectory f+ 'r g
It is possible to select a driving course from the starting point S to the destination point E that avoids the above, for example, the driving course a-1-b-1-C-+k or the driving course 6-j-+1-+k. It becomes possible. Therefore, it is possible to prevent a situation where the vehicle stops while traveling.

[Effects of the Invention] As described above in detail, according to the present invention, after confirming the positions of targets and obstacles within the travel area before traveling, the vehicle can travel from a starting point to a destination point while avoiding obstacles. By creating a course, you can get the following effects.

a. It is possible to travel to the target point without being obstructed by obstacles.

b. As a result of the above a, the travel time to the target point is shortened.

C3 No need to input complicated driving course data.

[Brief explanation of drawings]

Fig. 1 is a functional block diagram of the present invention, Fig. 2 is a system configuration diagram of a trackless unmanned vehicle to which the present invention is applied, and Figs. 3 (a) and (b) are detailed system configuration diagrams of the trackless unmanned vehicle. FIG. 4 is an operational flowchart for creating a running course according to the present invention, FIG. 5 is an operational flowchart for an embodiment of the present invention, FIG. 6 is a diagram for explaining another embodiment of the present invention, and FIG. 7 is a diagram for explaining a conventional driving course. Flowchart 1 showing the operation of the course creation device. FIG. 8 is a diagram specifically explaining the operation of the conventional travel course creation device. 1... Recognition means, 2... Route creation means. Patent Applicant: Toyoda Automatic Loom Works Co., Ltd. Machine R block diagram of each machine Fig. 1 Fig. 2 44υ Monthly running speed diagram Fig. 4 4! Fig. 5 Flowchart of the implementation of the present invention while removing the reciprocal sequence of the present invention Flowchart shown in Fig. 7. Conventional running speed. Figure 8.

Claims (1)

[Claims] A recognition means for obtaining positional information of a target and an obstacle, and a route creation means for creating a running course to the target that is not obstructed by obstacles based on the positional information of the target obtained by the recognition means. A driving course creation device characterized by: