JP2001072208A - In-warehouse automatic guided vehicle guidance device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic guided vehicle guidance device flexible for change of the layout of a warehouse with less stationary equipment, and high in self running accuracy. SOLUTION: This vehicle can independently run over a principal passage wide in width by comparing present position detected data from a present position detection device 3 with route data over digital map data, can take in a front image by means of a vehicle front camera acting as an image take-in device for a real environment recognition means 4 when the vehicle is to go into a passage between pallet shelves, to which accurate positional data is essential, so as to recognize the above passage by means of an image processing device, can correct its running direction to go into the passage between the pallet shelves when the running direction is found to have been out of the right direction, and can read in positional data mounted to the pallet shelf T when it goes through the pallet shelf T, so that the present positional data are thereby renewed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to guidance of an automatic guided vehicle in a warehouse, and more particularly to a guidance device having a small number of fixed facilities and having high flexibility for layout change.

[0002]

2. Description of the Related Art Recently, frozen food storage warehouses have become widespread due to an increase in the need for frozen storage of foods, and goods in and out of warehouses have become frequent due to the distribution system associated with changes in food culture. The transport vehicle greatly contributes to automation that eliminates a severe work such as a worker having to enter a warehouse and operate a transport vehicle, thereby improving work efficiency. A typical example of such an automatic guided vehicle is a storage device that stores digital map data such as a route in a warehouse, a yaw rate sensor that detects the position of the vehicle, and a position detection device that includes a travel distance sensor and the like. And the vehicle is controlled so that the current position data calculated from the position detecting device is adjusted to the digital map data in the storage device.

[0003]

However, in such a simple position detection method, an error occurs in the current position data due to an error in the yaw rate sensor or an error in the traveling distance data due to the slip of the wheels, so that the current position data deviates from the route. In addition, it is difficult to perform processing when there is an obstacle on the traveling route.

According to the present invention, an unmanned vehicle which reads a true position data by an image reading device, replaces the current position data, travels on an accurate route, finds an obstacle by performing image processing, and finds an obstacle, thereby facilitating a transfer operation. It is intended to provide a transport vehicle.

[0005]

According to the present invention, there is provided an unmanned vehicle guidance apparatus in a warehouse, comprising: a digital map storage means for digitizing and storing a map in the warehouse; and a route setting means for setting a traveling route of the vehicle on the map. A current position detecting means for detecting the position of the vehicle moved according to the set route, a real environment recognizing means for recognizing the real environment in the warehouse, and a real environment recognizing the position detected by the current position detecting means by the real environment recognizing means And a position matching control means for controlling to match the coordinates.

Further, the present position detecting means comprises a yaw rate sensor for detecting the turning angular velocity of the vehicle, an encoder for detecting the moving distance of the vehicle, and an arithmetic circuit for outputting the current position based on the integration of the outputs. Features.

The real environment recognizing means comprises a traveling path recognition visual sensor for recognizing a traveling path in a vehicle traveling direction and a shelf position recognizing visual sensor for detecting a predetermined shelf position in a direction perpendicular thereto. .

The real environment recognition means is characterized by comprising an omnidirectional visual sensor for simultaneously detecting a plurality of directions and recognizing a desired position.

[0009] The real environment recognition means is characterized by extracting and strengthening the contact portion of the row of shelves in the vehicle traveling direction, and enhancing the extraction for every pallet stored in the direction perpendicular to the portion.

[0010] The real environment recognition means is characterized in that an address designation display is provided on each pallet storage shelf appropriately arranged in the warehouse.

Further, a control for detecting a deviation between the current position outputted by the current position detecting means and the position recognized by the internal environment recognizing means and controlling the amount of the deviation to converge within a predetermined moving distance based on the actual environment coordinates. It is characterized by comprising means.

[0012] Furthermore, image information detected by a visual sensor as real environment recognition means is transmitted to a monitor room provided at a location remote from the detection site, and an operator can remotely operate based on the image information. Features.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention compares a current position detection data from a current position detection device 3 with path data on a digital map data for a main road which has a wide path width and does not cause any problem even if a running error occurs. When the vehicle travels independently and enters a passage between pallet shelves T where accurate position data is required from a main passage, a front image is captured by a vehicle front camera 41 which is an image capturing device of the real environment recognizing means 4.
In step 3, the path between the pallet shelves T is recognized. If the traveling direction is deviated, the path is corrected and the path proceeds between the pallet shelves T. In the path between the pallet shelves T, the position data attached to the pallet shelves T is read and the current position data is read. To update. In addition, image data can be transmitted to the outside of the warehouse, and traveling can be switched from independent traveling to remote control by an operator's operation. By providing such a means, it is possible to travel to an accurate position.If there is an obstacle, for example, an abnormal signal is generated and the image data is sent to an operator outside the warehouse, and transport operation is performed by remote control. Can be continued.

[0014]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of an apparatus for guiding unmanned vehicles in a warehouse according to the present invention. FIG. 1 shows a part of a layout in a typical warehouse, and FIG. 2 shows a basic block diagram of a guidance device mounted on a vehicle.
T1 to T4 in FIG. 1 are pallet shelves arranged at predetermined positions in the warehouse.

In FIG. 1, the current position (work standby position) of a vehicle such as a forklift is assumed to be point A.
Point B indicates a point set in advance on the extension of the CD line designated as the passage of the forklift, which is a work vehicle, in the passage between the pallet shelves T1 and T2 and with a distance sufficient to correct the displacement, Point X is the destination position of the route, and points E, F, and G indicate corner positions (edges) of the pallet shelf T1. Similarly, points H, I, and J indicate corner positions (edges) of the pallet shelf T2.

In FIG. 2, reference numeral 1 denotes a route data to be traveled by a vehicle stored on a digital map in the warehouse as a work track path of the vehicle in a digital map storage means 2 in which a map in the warehouse is digitized and stored in advance. The route setting means to be set, the current position detecting means 3 for detecting the position of the moving vehicle, and the real environment recognizing means 4 for recognizing the state of the surroundings of the vehicle. And other environments as visual information.

Reference numeral 5 denotes a position matching control means for controlling the current position data of the current position detecting means 3 to be matched with the real environment coordinates recognized by the real environment recognition means 4, and 6 is set in the digital map storage means 2. Control means for comparing the route data obtained and the current position data detected by the current position detecting means 3 and controlling the vehicle so that the vehicle travels along the route.

The transmission means 7 includes a display operation panel 8 installed in a monitor room remote from the current position of the vehicle.
The visual data of the surrounding environment imaged by the real environment recognizing means 4 is transmitted, and an artificial operation signal of an operator operated by the environmental image displayed on the display operation panel 8 is received to manually operate the vehicle. The control means 6 can be remotely controlled to operate.

The current position detecting means 3 may employ a representative position detecting means such as a gyro, but a cheaper configuration for detecting the current position in a small area in the warehouse can be used. . That is, as shown in the block diagram of FIG. 3, the moving distance is calculated from the outputs of the yaw rate sensor 31 and the encoder 32 that detects the rotation of the wheels, which are attached to the center of the vehicle, and the position of the previous position. And an arithmetic circuit 33 for calculating the current position from

Further, since the real environment recognition means 4 visually captures the edge of the pallet shelf T in the warehouse and the traveling path, as shown in FIG. 4, it captures the environment in the traveling direction of the vehicle, and particularly recognizes the traveling path. Front camera 41 (traveling path recognition visual sensor), and a vehicle measurement camera 42 (shelf position recognition visual sensor) that captures an environment in a direction substantially orthogonal to the vehicle and recognizes the position of, for example, a pallet shelf T, and both cameras And an image processing circuit 43 for inputting image signals by 41 and 42 and processing it as position data. The image processing circuit 43 detects the floor of the warehouse and the edges of the pallet shelves T in the traveling direction of the vehicle. Between the center of the passage in the advancing direction is calculated and output, and the address display provided on the pallet shelf T is recognized to obtain the address data and to proceed. Outputs a coincidence signal when it is at the center of the image with respect to the direction.

In such a configuration, when the vehicle travels on the set route ABX, the vehicle first travels according to the route data of AB by comparison and adjustment with the data detected by the current position detecting means 3. Yaw rate sensor 3
The position error detected by the encoder 1 or the encoder 32 itself accumulates due to the traveling and deviates from the correct traveling route. This deviated point is represented by B1 and B2.

When the vehicle starts at point A, the current position data from current position detection means 3 and digital map storage means 2
The control means 6 adjusts the traveling route of the vehicle by comparison with the traveling route data set on the in-warehouse map data from the vehicle and travels toward the point B. When it is determined that the current position data from the current position detecting means 3 has reached the point B, the actual environment recognizing means 4 detects the actual surrounding environment and performs image processing to detect the direction of the path and approach the center line CD of the path. While heading toward X. At this time, the vehicle is in the range between B1 and B2, but does not proceed on an erroneous route because it is set in advance with a sufficient distance to correct the displacement.

At this time, the traveling direction proceeds in the direction of CD,
Since the current position data obtained by the yaw rate sensor 31 and the encoder 32 may include an error, the position matching control unit 5 outputs the address data and the coincidence signal output by the image processing circuit 43 of the real environment recognizing unit 4. Since the current position data obtained by the fetched current position detecting means 3 is updated, the current position data can be corrected to the true current position data. Based on this data, the traveling direction of the vehicle is controlled by the control means 6 so that the point X can be more accurately reached.

The address to be attached to the pallet shelf T may be displayed on the label in units of shelves such as T1 and T2, for example, or may be affixed, or may be more finely identified and displayed for each shelf step (full storage). If the actual environment recognition means 4 recognizes this and determines the work position, the work can be performed by accurately recognizing the shelf position for carrying in and out.

The edge of the pallet shelf T recognized by the front camera 41 of the vehicle may be recognized from the angular shape of the shelf itself. The shelf position can be recognized, and the current position and the traveling route can be easily determined. The image processing circuit 43 of the real environment recognizing means 4 extracts and strengthens the installation portion of the shelf row captured by the vehicle front camera 41 and the vehicle side camera 42, and furthermore, for each pallet storage of the pallet shelf T. The line is subjected to extraction enhancement processing so as to be easily processed as digital data for position determination.

The camera as the image capturing device here uses a plurality of cameras such as the vehicle front camera 41 and the vehicle side camera 42 as in the embodiment.
By using a so-called omnidirectional vision sensor that can detect multiple directions by combining a single CCD camera and a 360-degree reflection mirror, it can capture a wider range of real-world information and only the necessary data in it , The same real environment recognition can be performed.

The position matching control means 5 detects a deviation between the current position data detected by the current position detection means 3 and the position recognized by the real environment recognition means 4 and corrects the current position data. The deviation amount detected and compared is set within a predetermined moving distance (a permissible deviation range in which the traveling route does not largely deviate) based on the actual environment coordinates based on the warehouse digital map data stored in the digital map storage means 2. Is controlled so as to converge.

As described above, the actual environment recognizing means 4 recognizes the actual surrounding environment in which the vehicle travels, and controls the self-propelled vehicle while converging the detection error of the position data by the current position detecting means 3 to more accurate position data. In addition, it is possible to easily travel along the intended traveling route easily, and it is possible to easily cope with a change in the pallet rack layout in the warehouse by making a digital map using the warehouse layout design data.

Further, it is also possible to easily provide a recognition process for generating an abnormality alarm when an obstacle is recognized in the traveling direction by the real environment recognition means 4,
Such an alarm, the vehicle front camera 41 and the vehicle side camera 4
Sending the image of No. 2 to the display operation panel 8 installed in the monitor room outside the warehouse by the wireless device, and switching the position control by the position matching control means 5 or the direct vehicle control by the control means 6 to the remote control via the wireless device. Thus, the operator can confirm the situation and proceed with the operation.

[0030]

According to the present invention, for example, the trunk path travels independently by comparing the current position detection data from the current position detector 3 with the path data on the digital map data, and requires accurate position data from the trunk path. When entering the passage between the pallet shelves T, the real environment recognizing means 4 recognizes the passage between the pallet shelves T, corrects the displacement of the current position, and proceeds to the passage between the pallet shelves T. In the aisle, the position data attached to the pallet shelf T is read and the current position data is updated so that the vehicle can travel independently along an accurate traveling route, and there are few fixed facilities such as traveling tracks and the layout of the warehouse It is possible to provide a guidance device having high flexibility with respect to a change. If there is an obstacle, for example, an abnormal signal is generated and the image data is sent to an operator outside the warehouse, and the operator can perform remote control to continue the transport operation.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of an in-warehouse map showing a part of an in-warehouse layout to which the in-warehouse unmanned vehicle guidance device of the present invention is applied.

FIG. 2 is a basic block diagram illustrating a configuration of a warehouse unmanned vehicle guidance device according to the present invention.

FIG. 3 is a block diagram showing one embodiment of a current position detecting means in the unmanned vehicle guidance device in a warehouse of the present invention.

FIG. 4 is a block diagram showing one embodiment of a real environment recognition means in the unmanned vehicle guidance device in a warehouse of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Route setting means 2 Digital map storage means 3 Current position detection means 4 Real environment recognition means 5 Position matching control means 6 Control means 7 Transmission means 8 Display operation panel

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 3F022 FF02 LL06 MM08 MM70 NN01 NN13 NN31 NN57 PP06 QQ03 QQ17 5H301 AA02 AA10 BB05 CC03 CC06 DD01 DD06 FF11 FF27 GG03 GG09 GG12 GG16

Claims (8)

[Claims] 1. A digital map storing means for digitizing and storing a map in a warehouse, a route setting means for setting a traveling route of a vehicle on the map, and a current position detecting means for detecting a position of a vehicle moved according to the set route. And real environment recognition means for recognizing the real environment in the warehouse, and position matching control means for controlling the position detected by the current position detection means to match the real environment coordinates recognized by the real environment recognition means. An unmanned vehicle guidance system in a warehouse. 2. The current position detecting means includes a yaw rate sensor for detecting a turning angular velocity of the vehicle, an encoder for detecting a moving distance of the vehicle, and an arithmetic circuit for outputting a current position based on these outputs. The unmanned vehicle guidance device in a warehouse according to claim 1, wherein: 3. The actual environment recognizing means is substantially similar to the traveling path recognizing visual sensor for recognizing the traveling path in the vehicle traveling direction and the traveling path recognizing visual sensor for detecting a predetermined shelf position arranged in a warehouse. The unmanned vehicle guidance device in a warehouse according to claim 1, comprising a shelf position recognition visual sensor for recognizing a direction orthogonal to the rack. 4. The unmanned vehicle guidance system in a warehouse according to claim 1, wherein the real environment recognition unit is configured by an omnidirectional vision sensor that detects a plurality of directions simultaneously and recognizes a desired position. 5. The real environment recognizing means extracts and strengthens a contact portion of a row of shelves in a vehicle traveling direction arranged in a warehouse,
2. The unmanned vehicle guidance system in a warehouse according to claim 1, wherein extraction enhancement is performed for each pallet stored at right angles thereto. 6. The unmanned vehicle guidance system in a warehouse according to claim 4, wherein the real environment recognition means performs an address designation display on each pallet storage shelf arranged in the warehouse. 7. The position matching control means detects a deviation between a current position output by the current position detection means and a position recognized by the real environment recognition means, and determines the amount of the deviation based on real environment coordinates. 2. The unmanned vehicle guidance system in a warehouse according to claim 1, wherein control is performed to converge within a moving distance. 8. An image information detected by the real environment recognizing means is transmitted to a monitor room provided at a location distant from the detection position, and an operator can remotely control the vehicle based on the image information. Claim 1 characterized by the following:
An unmanned vehicle guidance device in a warehouse according to the item.