JPH01241604A - Unmanned load working device - Google Patents

Abstract

PURPOSE:To decrease the equipment charge of guide facilities by executing the traveling of an unmanned load working device from guided traveling by a guide line, etc., to autonomous traveling when it is detected that the device goes into a prescribed block and executing the autonomous traveling of the device based on an approach tack which is prepared on data concerning a detected target position. CONSTITUTION:When a unmanned load working device 1, which travels with being guided by guide lines 12 and 12', etc., arrives at the prescribed block such as a load working block, etc., it is detected by a detecting means 7 that the device arrives at the prescribed block and the traveling is switched to the autonomous traveling. The target position such as a loading position, etc., is detected by a detecting means 8 and the approach track from a present position to the target position is prepared based on the data concerning the target position. Then, based on this approach tack, the device executes the autonomous traveling. Next, it is confirmed that the device arrives at the target position and after that, load working is automatically executed. When the work at the position is finished, the device goes back until a signal is detected from the guide line, etc. Then, the traveling is switched to the guided traveling again and the device moves to the next target position. Thus, investment in plant and equipment can be drastically reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application This invention relates to an unmanned cargo handling device such as an unmanned forklift that automatically travels within a warehouse or factory to load, unload, and transport cargo.

Conventional technology Conventional unmanned cargo handling devices such as unmanned forklifts that automatically load and unload cargo in warehouses and factories operate on a predetermined travel route while detecting signals from guide wires laid on the floor. When the vehicle automatically travels to a predetermined loading position, it stops traveling and automatically performs cargo handling work at that location, and when the cargo handling work is completed, it is guided along the travel route again to the next destination. In many cases, the vehicle is controlled to run automatically.

Problems to be Solved by the Invention However, in the case of the above-mentioned unmanned cargo handling device, since the traveling route is determined by guide lines, etc., the loading position etc. for carrying out cargo handling work must be set in advance based on the traveling route. In addition, if the cargo is placed in a position shifted from the set position, the position of the unmanned cargo handling device is also determined by guide lines, etc. It is not possible to correct the relative positional deviation between the unmanned cargo handling device and the cargo, making it impossible to perform cargo handling operations smoothly. Therefore, in the past, the current situation was to operate the system by modifying detailed operation programs through trial and error, which had the problem of poor workability. In addition, when unloading loaded cargo after stacking it in a predetermined location, if there is a shift in the position of the lower cargo, the unloading position cannot be corrected, resulting in the cargo collapsing and preventing normal cargo handling operations. The problem was that it couldn't be done.

On the other hand, Japanese Unexamined Patent Publication No. 54-114690 describes a control method for flexibly controlling the running of an unmanned carrier. This method includes two direction detection means that are placed at different positions in the work area of the unmanned carrier vehicle and detect the direction of the unmanned carrier vehicle from these positions, and an unmanned vehicle based on the detection of these two direction detection means. It is equipped with a position calculation circuit that calculates the position of the transport vehicle, and a calculation circuit that generates a running command for the unmanned transport vehicle based on the calculated position and the movement target position of the unmanned transport vehicle. Good luck! This is a method of controlling driving by feeding the inside of the vehicle.

Therefore, the position of the unmanned carrier can be detected by the two direction detection means, and the vehicle can be controlled to travel along a predetermined course such as the shortest route to a preset target position. The target position must be set in advance, and there is no function to deal with misalignment, so problems such as cargo collapse cannot be dealt with.

This invention was made against the above-mentioned technical background.
The object of the present invention is to provide an unmanned cargo handling device that can detect a target position within a predetermined area, autonomously travel to the target position, and perform cargo handling work.

Means for Solving the Problems As a means for solving the above-mentioned problems, the present invention provides an unmanned cargo handling system that has both a guided driving function using external signals and an autonomous driving function that does not rely on external guiding signals. The device includes a means for detecting entering a predetermined area such as a cargo handling work area, and a means for reversibly changing the driving method from guided driving using a guide line etc. to autonomous driving when detecting entering the predetermined area. means for detecting a target position such as a load position; means for generating an approach trajectory from the current position to the target position based on data about the detected target position; The vehicle is characterized by having a means for controlling the vehicle so that the vehicle travels autonomously.

Operation By having the above configuration, when the unmanned cargo handling equipment traveling guided by the guide line etc. reaches a predetermined area such as a cargo handling work area, the detection means detects that it has reached the predetermined area. It can be detected and switched to autonomous driving. When switching to autonomous driving, the detection means detects the target position such as the cargo position, and an approach trajectory from the current position to the target position is generated based on data about the target position, and the target position is determined based on this approach trajectory. It runs autonomously until Once the vehicle has reached the target position and confirms that it has reached the target position, the cargo handling work is carried out on its own. When the work at that location is completed, the vehicle backs up until a signal from a guide line or the like is detected, switches to guided travel again, and moves to the next destination.

EXAMPLE Hereinafter, an example of the present invention will be explained based on FIGS. 1 to 15.

The unmanned forklift 1 unmannedly performs loading and unloading work such as loading and unloading cargo in warehouses and factories. In addition to being able to move up and down along the vertical axis, the tilting hydraulic cylinder 4 can change the inclination in the front and rear directions, and the left and right hydraulic cylinders 5 and 6 can move it in the horizontal direction. Further, a window 2a is provided at the tip of each of the forks 2, 2, and a television camera 7 disposed inside the tip, a distance sensor 8 for ultrasonic or optical distance measurement, etc. are connected to the window 2a. 2a toward the front, and a lifting height sensor 9 is provided at the top of the mast 3 to detect the height of the load placed on the floor and determine the lifting height of the fork 2. It is being

Although not shown, the unmanned forklift 1 can perform guided travel in which it travels on a predetermined travel route, autonomous travel in which it travels within a cargo handling work area, and cargo handling operation in which it loads and unloads cargo. It is equipped with a control device.

The control of guided travel by the control 11 device is as follows:
A guide Ij! laid on the floor surface on which the unmanned forklift 1 travels along a predetermined travel route. By detecting car signals etc. with a detection coil (not shown), the vehicle is guided and automatically travels.It also detects mark plates, etc. installed at predetermined positions on the travel route, and decelerates.
In addition to performing actions such as accelerating, stopping, or turning left or right,
The system detects when the vehicle enters a cargo handling area and switches its mode of travel from guided travel using guide lines to independent travel.

Furthermore, the control of autonomous travel by the control device 11 involves three-dimensionally capturing the position of the cargo within the cargo handling work area using a pair of television cameras 7 disposed within the tip of the fork 2.2. The distance sensor 8 measures the distance from the current position of the unmanned forklift 1 to the load, and an approach trajectory from the current position to the cargo handling position is generated from the obtained cargo position information, and based on the generated approach trajectory. The vehicle is controlled and runs autonomously.

Furthermore, the control of the cargo handling operation by the control device is performed based on the degree of directness of the unmanned forklift 1 with respect to the load obtained from the position information of the load three-dimensionally captured by the pair of television cameras 7, and the distance sensor 8. The approach locus is generated based on the measured distance to the load and the height information of the load obtained by the lifting height sensor 9, and the vehicle autonomously travels.

When the vehicle reaches the location of the cargo, it stops traveling and automatically performs cargo handling work at that location. When the cargo handling work is completed, it returns to the guide line level by traveling based on the map and traveling based on target object detection, and then resumes the traveling route. The vehicle is guided so that it automatically travels to the next destination.

In addition, the cargo handling operations carried out by this unmanned forklift 1 include unloading the cargo G brought in by the truck T and loading it onto the shelf S, and loading the cargo G on the shelf S onto the trolley (see Figure 3). , Direct stacking of loads separated by the stacking machine N/C onto a fixed position on the floor in a warehouse or factory (see Figure 4), or supply of empty pallets Pl to the processing machine H/C. and a pallet P containing processed products.
There are operations such as taking out and temporarily storing the parts 2 and loading them onto the truck T (see Fig. 5).

For example, the case of loading and unloading cargo at a predetermined position on the floor in a warehouse will be explained with reference to FIG.
On the floor surface, a guide line 12.12' is provided along the traveling route of an unmanned forklift (not shown) that transports the load 11.11' that has been separated into layers by the separation machine.
Point A on the driving route, 8. C..., A'
, B', C'...... each lead, Ii! 12. To the side of 12',
Places AI and A for placing the cargo 11.11'
2,...,A,n,31.

32, ...3n, C1, C2, ...
・Qn, ・・・・・・・・・Place A '1
．． A'2. -=-A'n, B'1. B'
2゜...8'n, C'1. C'2. ...
・・・C'n, ・・・・・・・・・ is each point A
, B, C, ... and A', B'.

It is set as a linear space extending from C'... in a direction perpendicular to the guide line 12, 12'.

Further, in the vicinity of points A and A' on the travel route where the guide lines 12 and 12' are laid, there are mark plates 13. 13' are provided respectively. For example, when transporting the load 11 to the storage location C1, the unmanned forklift loaded with the load 11 detects the mark plate 13 while being guided along the y, 4 line 12. A driving route is calculated from the coordinates of the position of the mark plate 13 on the route and each position data of point C and storage location C1, and the section to point C is first guided 1gJ12.12
' At the same time, the vehicle switches to autonomous driving at point C, and autonomously travels toward the storage location C1 by a distance corresponding to the distance from point C to the storage location C1. If there is no other load on C1, the height of the load G is measured by the lifting height sensor 9, and if another load G is already placed on the floor, the height of the load G is measured. Accordingly, the fork 2 is raised, stacked on top of the above-mentioned position G, and the load 11 is unloaded, then retreated towards the above-mentioned point C, and when the signal from the induction line 12 is detected, it is guided from autonomous running. After switching to line-guided travel, it moves to the position of the new load 11.11'.

In addition, the target storage location A1, 31 to be unloaded.

There are three main types of cargo handling work patterns when other cargo is already placed on the CI, Cn, etc., and each of these patterns will be explained with reference to Figures 7 to 12. .

The work pattern for cargo handling differs depending on the state in which the cargo G is placed, and the first pattern, as shown in FIG.
Center line L of the unmanned forklift 1 in the direction of travel
In this case, the unmanned forklift 1 can move straight ahead and perform cargo handling work. The second pattern, as shown in FIG. 8, is a case where the straight line L1 passing through the center of the position G is parallel to the vehicle center line L2 and deviates by the width W. If the loads 11 are stacked while ignoring the amount of shift of the load G to be stacked, there is a risk of the loads collapsing, so it is necessary to stack the loads 11 according to the shifted position of each load G. Load handling is performed by correcting the course of the unmanned forklift 1 so that the vehicle body center line L2 overlaps the straight line ILt passing through the center of the load G. Furthermore, the third pattern is as shown in Figure 9.
In the case where the straight line L1 passing through the center of the position G is deviated so as to intersect with the vehicle body center line L2 at an angle θ, it is necessary to prevent the cargo from collapsing as in the case of the second pattern, and in this case, Load handling is performed by correcting the course of the unmanned vehicle so that the center of the vehicle body 1112 overlaps the straight line L+ passing through the center of the load G and the - straight line.

Furthermore, the above-mentioned No. 1. Second. In the case of each third pattern, the lifting height sensor 9 provided on the mast 3 of the unmanned forklift 1 detects how many tiers the load G is stacked, and the fork 2 is adjusted according to the number of tiers of the load G. let it rise,
The load 11 supported on the fork 2.2 is correctly lowered onto the said position G and stacked.

Next, the first. Second. Control of cargo handling work in each of the third patterns will be explained.

The unmanned forklift 1 controls guided travel based on signals from guides 1i112 and 12' laid along the travel route, and uses travel distance data based on a map and measurement wheels (not shown) in the warehouse. A television camera 7 and a distance sensor 8 provided within each tip of the pair of forks 2.2, and a lifting height sensor 9 provided at the top of the mast 3.
Based on various positional data obtained by the above methods, autonomous travel and cargo handling operations are controlled by generating approach trajectories to the target position.

11 and 12 show examples of cargo to be handled by the unmanned forklift 1. In the case of the box-shaped pallet 11 with supports at the four corners shown in FIG. 11, first, when loading, The box-shaped pallet 21 to be loaded is recognized. The box-shaped pallet 21 is recognized by two television cameras 7.2 installed within the tips of the forks 2.2 of the unmanned forklift 1, as shown in the block diagram of FIG.
7, the box-shaped pallet 21 is imaged, and the image processing unit 3m! Dressed W!
113.13 to extract the palette features,
From the obtained data, the three-dimensional position calculation unit 14 calculates the three-dimensional position of the box-shaped pallet 21 and outputs position information to the travel control unit 15 of the unmanned forklift 1. and,
Travel control during cargo handling work is performed based on the input traveling data, and as shown in the block diagram of FIG. An approach trajectory to the target position for unloading is generated in the approach trajectory generation unit 17 from the current position obtained from the map, the travel distance data by the measuring wheels, and the position information of the box-shaped pallet 21. The locus is given to the drive unit 18 to control autonomous travel and cargo handling operations.

That is, the centers of the fork insertion height dimension a and the fork insertion width dimension a are detected, and the vehicle body center line L2 of the unmanned forklift 1 is aligned with the center of the fork insertion width dimension a. Next, the unmanned forklift 1 is advanced, the forks 2.2 are inserted into the lower surface of the box-shaped pallet 21, and after tilting upward by the tilting hydraulic cylinder 4, the lifts 1 and 2 are raised. Lift up the box-shaped pallet 21. On the other hand, when unloading, the box pallet 2 is placed at the target position.
1 is not detected, calculate the travel route to the target location for unloading from the input map and travel distance data of the traveling wheels, perform autonomous travel or guided travel using guide lines, and mark The target position is detected by the plate and the box-shaped pallet 21 is properly unloaded. In addition, if another box-shaped pallet 21 has already been placed at the target position, first, the box-shaped pallet 21 placed is recognized in the same manner as during loading, and the box-shaped pallet 21 is detected by the lifting height sensor 9. The height H of the pallet 21 is detected, and the fork 2.2 is raised to a height corresponding to the number of stages of the box-shaped pallets 21 placed.

Next, the center of the insertion width dimension of the placed box-shaped pallet 21 is detected, and the two television cameras 7 provided at the tips of the forks 2゜2 are used to detect the center of the insertion width dimension of the placed box-shaped pallet 21. The positions of the upper ends e and f of the left and right pillars on the near side are detected, and the positions of the upper ends e and f are three-dimensionally processed as coordinates.

As a result of coordinate processing, if the straight line 1+ passing through the center of the placed box pallet 21 and the body center line L2 of the unmanned forklift 1 deviate from each other, that is, if
If the box-shaped pallet 21 supported by The lower ends of the left and right pillars on the front side of the pallet 21 supported on the fork 2.2 touch the upper ends e of the corresponding pillars on the left and right front sides of the pallet 21 that has already been placed, After moving the unmanned forklift 1 so that it is directly above the position f, the forks 2.2 are lowered and stacked on top, and after tilting the forks 2, 2 downward by the tilting hydraulic cylinder 4, the unmanned Forklift 1
to complete unloading.

FIG. 12 shows another type of pallet 31, in which the pallet position is recognized by the means shown in the block diagram of FIG. 13, as in the case of the box pallet 21, and The traveling of the pallet 31 is controlled by the means shown in the block diagram, and a cargo 31' is placed on the pallet 31. And this cargo 31
When loading the pallet 31 on which ' is placed, onto the unmanned forklift 1, first, the pallet 31 is recognized, the centers of the fork insertion height dimension a and the fork insertion width dimensions S and C are detected, and the unmanned forklift 1 Align the center line of the vehicle body with the center of the insertion width dimension. Next, move the unmanned forklift 1 forward and fork 2
．． 2 into the pallet 31, and tilt hydraulic cylinder 4.
After tilting upward by the fork 2.2, the pallet 31 is lifted by raising the fork 2.2.

On the other hand, at the time of unloading, if the pallet 31 is not detected at the target position, a traveling route to the target position for unloading is calculated from the input map and travel distance data of the traveling wheels, and autonomous driving is performed. Alternatively, the pallet 31 is unloaded appropriately by performing guided travel using a guide line and detecting the target position using a mark plate. Further, if another pallet 31 is already placed at the target position, the placed pallet 31 is recognized,
In addition to detecting the center of the insertion width dimension, the two television cameras 7 provided at the tips of the forks 2 and 2 detect the posture of the pallet 31 placed on the pallet 31. The respective positions of the left and right upper ends e and f are subjected to three-dimensional coordinate processing. As a result of coordinate processing, a straight line L+ passing through the center of the pallet 31 placed
If the body center line L2 of the unmanned forklift 1 is out of alignment with the vehicle body center line L2, that is, if the pallet 31 supported by the fork 2.2 is not facing directly, the straight line Ll and the vehicle body center line L2 will be aligned in a straight line. The approach trajectories of the unmanned forklift 1 are automatically generated so as to overlap, and the unmanned forklift 1 autonomously travels based on the obtained approach trajectories so that the lower left and right corners of the front side of the pallet 31 supported on the fork 2.2 are After moving the cargo 31' of the pallet 31 that has already been placed so that it is directly above the left and right upper ends e and f of the front side, the forks 2 and 2 are lowered and stacked on top, and the tilting hydraulic cylinder 4 After tilting the fork 2.2 downward, the unmanned forklift 1 is moved backward to complete unloading.

Furthermore, the pallet 21 loaded onto the unmanned forklift 1
．． The position of the fork 2.2 on the pallet 21.31 can be recognized by a sensor or the like, but by tilting the fork 2.2 inserted into the pallet 21.31 upward by a predetermined amount using the tilting hydraulic cylinder 4, the position of the fork 2.2 on the pallet 21. 2
1.31 by its own weight to a position where it contacts the mast 3, and the left and right hydraulic cylinders 5.6 are operated to move the forks 2. By restraining the pallet 21.31 in a fixed position on the fork 2.2, it is possible to carry out cargo handling operations without having to recognize the position of the pallet 21.31 on the fork 2.2.

Next, an example of the case where the unmanned forklift 1 performs cargo handling work on the cargo 11 will be explained based on the flowchart of the cargo handling work program shown in FIGS. 15(A) and 15(B).

When an unmanned forklift 1, which automatically travels according to a guidance signal from a guide line laid along a travel route, arrives at a cargo handling work area with cargo 11 loaded, it detects a mark plate and starts a cargo handling work program.

First, in step 1, the load G is detected in order to check whether the load G is already present at a target position determined in advance for unloading. Then, in step 2, the detection results in step 1 are checked.

In step 2, if the load G is not detected, the unmanned forklift 1 advances toward the target position and proceeds to step 26, detects a mark plate indicating the target position for unloading, and proceeds to step 27. Proceed and check the detection results. As a result of the detection, if the mark plate is not detected, the process proceeds to step 28, where the unmanned forklift 1 moves further toward the target position, returns to step 26, detects the mark plate, and detects the mark plate. Step 2 until
6. Repeat each process of 27.28. and,
If a mark plate is detected in step 27, the process proceeds to step 10. On the other hand, if the load G is detected at the target position in step 2, step 3
Proceed to.

In step 3, the number of stages of the load G is detected by the lifting height sensor 9 provided on the mast 3 of the unmanned forklift 1. Then, in step 4, it is checked whether or not the load G is in the 1st stage, and if it is not in the 1st stage, step 29
Proceed to and check whether the direction G is two stages. As a result, if it is determined that the cargo G is not stacked in one or two stages, the process returns to step 3 and the number of stages of the cargo G is detected again.

In addition, if the load G is determined to be in the 1st stage in step 4, or if it is determined to be in the 2nd stage in step 29, the process proceeds to the next step, and in step 5, the fork 2°2 is moved in accordance with the number of stages in the direction G. The robot will rise to the desired height and proceed to step 6.

In step 6, the 2
Three-dimensional measurement of the position in the direction G is performed by the television camera 7 on the stand, and the process proceeds to step 7, where the load G is measured based on the measurement result.
Detects positional deviation.

If it is determined in step 8 that there is a positional shift of the load G, the process proceeds to step 30, and the posture of the unmanned forklift 1 is determined based on the position information of the load G obtained by the three-dimensional measurement in step 6. After the control is performed, the process returns to step 6. If it is determined in step 8 that there is no displacement of the load G, the process proceeds to step 9.

After completing step 9, the unmanned forklift 1 will start running.
The driving based on the map is switched to the driving based on target object detection, and the unmanned forklift 1 moves forward based on the approach trajectory generated from the position information of the load G obtained by the three-dimensional measurement performed in step 6, and the process proceeds to step 10.

In step 10, a check is made to see if the unmanned forklift 1 has reached the target position, that is, the unloading position, and if it is determined that the unmanned forklift 1 has not reached the target position, the process returns to step 7 and again. Detection of positional deviation is performed. As a result of the check in step 10, if it is determined that the target position has been reached, the process proceeds to step 11.

After proceeding to step 11, the fork 2.2 is lowered and the load 11 supported by the fork 2.2 is stacked on top of the placed load G, proceeding to step 12, the unmanned platform η- In order to prevent load shift at 30 o'clock after lift 1, the load on lift 2.2 is detected, and in step 13, if no load is detected, that is, load is detected, lift 2.2 is detected. Determining that the drop of ゜2 is insufficient, step 1 is carried out.
1, and the fly 4-2.2 further descends. and,
If it is determined in step 13 that there is no load, the process proceeds to step 14.

Proceeding to step 14, the unmanned forklift 1 starts moving backward, and in step 15, a guidance signal is detected. If it is determined in step 16 that the trigger signal is not detected, the process proceeds to step 31, where the attitude control of the unmanned forklift 1 has been performed, and the process returns to step 14, where the unmanned forklift 1 further moves backward.

If the guidance signal is detected in step 16, the process proceeds to step 17.

When proceeding to step 17, the unmanned forklift 1 returns from autonomous driving to automatic driving based on the guidance signal, and step 18
The process proceeds to step 19, moves along the travel route toward the new loading position of the load 11, and proceeds to step 19.

In step 19, the fork insertion position of the new load 11 is measured, and in step 20, the fork insertion position of the new load 11 is measured.
Detection of positional deviation is performed.

If the positional deviation of the footrests 2, 2 is detected in step 21, the process proceeds to step 32, and after the posture control of the unmanned forklift 1 is performed, the process returns to step 19, and the fork insertion position is measured. will be performed again.

If there is no displacement of the forks 2, 2 in step 21, the process proceeds to step 22, where the unmanned forklift 1 moves forward and the forks 2, 2 are inserted, and then in step 23, the forks 2.2 are raised. After that, step 24
Proceed to.

In step 24, the fork 2.2, which is in a raised state supporting the load 11, is tilted upward by the tilting hydraulic cylinder 4. As a result, the load 11 held by the fork 2.2 moves under its own weight to a position where it comes into contact with the mast 3. Furthermore, in step 25, fork 2.
2 is driven by the left and right hydraulic cylinders 5 and 6 to move in the left and right directions, respectively, so that the direction 11 is moved to a fixed position, that is, a straight line passing through the center of the load 11 supported on the forks 2 and 2.
After correcting the position so that 1Lt is in the same vertical plane as line L2, which is the center of the vehicle body of unmanned lift 4-lift 1, the process returns to the start state of this flowchart and steps 1 are repeated.
Perform each process from

Although this embodiment has been described in the case of an unmanned forklift, it can be similarly implemented in the case of unmanned cargo handling devices of other cargo handling methods such as a grasping type and a hanging type.

As described in detail, the unmanned cargo handling device of the present invention has both a guided traveling function using external signals and the ability to autonomously travel without relying on external guidance signals, and is capable of operating in cargo handling areas, etc. means for detecting entering a predetermined area; means for reversibly switching the driving method from 4-wire driving to autonomous driving; and means for detecting target positions such as cargo positions; Since it has a means for generating an approach trajectory from the current position to the target position based on data about the target position, and a means for controlling the vehicle to travel according to the generated approach trajectory, If the position of the cargo deviates from its normal position, the deviated position is detected and the cargo is stacked in accordance with the deviated position, thereby preventing cargo handling failure that may occur if the cargo is handled without adjusting to the deviated position. Cargo handling work can be carried out smoothly by preventing the occurrence of problems such as damage and collapse of cargo.Also, since the cargo is loaded and unloaded by running under its own weight, it is easy to change the layout of the stockyard for cargo such as inside the warehouse. At the same time, it is possible to significantly reduce the amount of equipment required to change the layout.Also, the degree of freedom in modifying the control program of unmanned cargo handling equipment is increased by changing the layout.Furthermore, the installation distance of guidance facilities such as guide lines can be increased. In addition to reducing the number of mark plates for confirming cargo handling positions, etc., this method has the effect of significantly reducing equipment costs.

[Brief explanation of the drawing]

1 to 15 show embodiments of the present invention,
Fig. 1 is a perspective view of an unmanned forklift, Fig. 2 is a perspective view of the main parts showing the inside of the fork tip with a cutaway, and Figs. Figure 6 is an explanatory diagram showing the layout inside the warehouse when an autonomous driving area is provided, and Figures 7 to 9 show the cargo handling pattern of the relative positional relationship between the cargo that has already been placed and the cargo that will be unloaded. , Figure 7 shows the case where there is no positional deviation, and Figure 8 shows the case where there is positional deviation in the parallel direction.
Fig. 9 is an explanatory diagram showing the case where there is a positional deviation in the rotational direction, Fig. 10 is an explanatory diagram showing the method of detecting the height of the load,
Fig. 11 is a perspective view of a box-type pallet, Fig. 12 is a perspective view of a flat pallet with a load placed on it, Fig. 13 is a block diagram of pallet position recognition, Fig. 14 is a block diagram of travel control, Figures 15 (^) and (B) are flowcharts of the cargo handling work program. 1...Unmanned forklift, 2...Fork,
2a...Window, 3...Mast, 4...Hydraulic cylinder for tilting, 5, 6...Left and right hydraulic cylinders, 7
...TV camera, 8...Distance sensor, 9...
- Lifting height sensor, 11... Load, 12... Guide line, 13... Image processing section, 14... Three-dimensional position calculation section, 15... Travel control section, 16... Current position Coordinate measurement unit, 17... approach trajectory generation unit, 1
8... Drive unit, 21... Box-shaped pallet, 31...
・Pallet, Ll... Straight line passing through the center of the load, L2
...The body center line of an unmanned forklift. Applicant Toyota Motor Corporation Agent Patent Attorney Rei 1) Oxygen deficiency (1 other person) Figure 6 Lightning 1 X Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 13

Claims (1)

[Claims] In an unmanned cargo handling device having both a guided traveling function based on external signals and an autonomous traveling function not based on external guidance signals, means for detecting entering a predetermined area such as a cargo handling work area; , a means for reversibly switching the driving method from guided driving using a guide line etc. to autonomous driving when entering the predetermined area is detected, means for detecting a target position such as a cargo position, and a detected target. An unmanned cargo handling device characterized by having a means for generating an approach trajectory from a current position to a target position based on data regarding the position, and a means for controlling autonomous travel based on the generated approach trajectory. .