EP2851331B1 - Method for controlling a picking vehicle - Google Patents

Description

The invention relates to a method for controlling a picking truck. In particular, the invention relates to a method for controlling an industrial truck, especially in picking, wherein the truck has an optical sensor with a monitoring area, which is connected to a control device, and the control device by evaluating the data of the optical sensor, the position of a person within of the monitoring area is determined and tracked.

For picking, the compilation of goods deliveries in a warehouse, very often pedestrian pallet trucks are used as a form of material handling equipment. In this case, the pedestrian pallet truck is moved with a charge carrier on load forks, such as a pallet or a wire mesh basket, through the rack aisles of a warehouse in which the goods to be picked or picked are stored. Depending on the order, the order picker or the person removing the goods picks up the goods located on the shelf, places them on or in the load carrier of the drawbar lift truck and then has to drive to the next picking location within the goods warehouse or the rack aisle of the goods warehouse for the next goods. When the respective picking order has been completely processed, the order picker moves to a goods delivery with the pedestrian pallet truck. The routes from one location to the next are in the great majority short to very short straight distances within a rack aisle or along a shelf in the warehouse. In sum, however, due to the large number of required travel movements of the truck, which must be made by the order picker, a considerable amount of time, since the order picker has to go to a control handlebar of the pallet truck to operate the corresponding driving switch and the vehicle during the driving movement control, and then back to the position for picking up the goods.

There are therefore known methods and trucks for semi-automated order picking. This semi-automated picking takes place in the Principle similar, only the order picker, the truck must at least on the straight routes along a shelf no longer lead itself, but the truck moves automatically. The control of the truck while driving to the next receiving station can be done for example by a radio remote control, such as a wireless glove with a radio remote control function, or by controlling the truck via voice commands.

For example, a solution with a wireless glove is known, which attracts the picking person and through which the control of the truck is made possible. It can be integrated as a command button in the glove individual functions such as "forward", "back", etc., which can be pressed by the picking person or the order picker. After picking up a particular product at a pickup location, thus after picking, for example, the "before" command key on the glove can be pressed by the order picker and the truck continues to move in the aisle as long as the order picker presses the command key on the glove holds. Due to the integration of the radio command transmitter in the glove, it does not have to be worn as a separate object by the order picker and does not disturb the latter in its freedom of movement.

A disadvantage of this prior art, however, is that the use of such a glove by a plurality of persons can lead to acceptance problems. Also every time an operation with the other hand must be done. The wireless gloves have the further disadvantage that they wear out through daily handling and thus need to be replaced regularly. This causes additional, ongoing costs.

Another known method is the control by voice commands, in which, however, additional location information or information about the position is required to automatically lead the truck between the receiving locations of the goods of the picking process. In this case, the truck is automatically guided to the next location via a goods management systems and a voice input from the order picker, when the picking is completed at a recording location. Also here is the Goods removal manually.

A disadvantage of this prior art is that background noise can interfere. Also, the additional location information must be available. In general, partially automated systems based on commands via voice input have the disadvantage that they have to be laboriously installed. For example, installations in warehouses are required for navigation and communication, a goods management system must be connected and servers must be made available in order to achieve the functionality of such a system. However, this suffers from the flexibility of the logistics processes, as rapid changes are only possible to a limited extent. Also, the economic and time effort is often spared. In all the systems described above, there is also a fundamental need for the employees performing the order picking to be trained and instructed in the system.

From the EP 2 533 119 A1 a device for radio remote control of a truck with a wireless glove is known, which can be used for picking.

US 2002/190849 A1 discloses a method according to the preamble of claim 1.

The present invention has for its object to provide a method for controlling an industrial truck, in particular an industrial truck for picking, available, which avoids the aforementioned disadvantages and with which the truck in a simple manner with a person together when loading goods can be moved.

This object is achieved by a method having the features of independent claim 1 and by a truck with the features 17. Advantageously, the developments of the invention are specified in the subclaims.

The object is achieved in that in a method for controlling an industrial truck, in particular during order picking, wherein the truck has an optical sensor with a monitoring area which is connected to a control device, and the control device by evaluating the data of the optical sensor Position of a person within the surveillance area determines and tracks, within the surveillance area is defined at least one monitoring limit and is detected by the control device when the person's position has exceeded the monitoring limit, and the truck is driven by the control device via a traction drive of the truck until the position of the person is back on the original side of the monitoring limit ,

The monitoring boundary provides a kind of virtual or invisible boundary established within the area of the environment monitored by the optical sensor, the surveillance area. Whenever a person exceeds this limit, the truck automatically tracks. This can be compared in the figurative sense as two magnets that are held together with the same polarity and thus the one is always pushed on too close approach of the other. In this case, the optical sensor can be any type of sensor from a laser scanner, which detects the environment alone in one plane, up to a camera. It can be a capture only one-dimensional act up to a three-dimensional detection of pixels, in which the distance to the optical sensor is detected. In the method, the data points corresponding to a person within the surveillance area are detected, and as these points move beyond the monitoring boundary, the truck is tracked by the control device, which may be part of a vehicle control, for example, or a separate computer, specifically for this purpose. In this case, the truck is moved and with this together the monitoring area including the monitoring limit until it is moved back over the position of the person and the person is back on the original side of the monitoring boundary. The inventive method allows a particularly advantageous semi-automated picking. An employee, the picker, has no time to track or move the truck. The truck, so to speak, automatically follows its running movements while, for example, it goes along a series of containers with the goods to be picked or a rack aisle and removes these goods. The employee can remove the goods completely freely and place them on, for example, a pallet or a container on the fork of the truck. Whenever all the goods have been picked in one area and the employee continues to move along the rack aisle, the truck follows him automatically. Advantageous for this is no fixed installation in For example, a warehouse needed and no additional device that would have to be carried by the employee. In particular, however, no elaborate training or briefing is required because the operation by the picking employee is ultimately only to pick the goods as usual and continue each. Industrial trucks with the method according to the invention can be used as a single or in a variety of vehicles in a warehouse without major preparations must be made.

Advantageously, the optical sensor can detect distances to objects in the surveillance area.

By detecting the distances to the objects in the surveillance area, tracing a person, and in particular distinguishing moving objects such as persons from fixed objects, is greatly simplified. Thus, not only the movement in one image plane but also in two dimensions including a movement out of this image plane can be judged. In this case, an assessment of the movement with respect to, for example, a continuous movement and the movement speed is possible, which, for example, must be kept within certain frame values in the case of a person walking on foot.

In a further embodiment of the method, the optical sensor can detect the monitoring area in three dimensions.

If the optical sensor detects not only distances in a plane but also with a certain height extent and thus the objects three-dimensionally, for example a time of flight camera compared to a simple laser scanner, the detection and the tracking or "tracking" of a person can continue be improved.

In a favorable embodiment, changes in the position of the person are tracked and the changes in the values are only judged as correct with continuous changes.

As a result, errors can be avoided. The fact that a person has to move continuously in the surrounding space makes sudden jumps in position, for example in the mathematical sense, inaccurate. It is conceivable in such a case for safety reasons to end the process or to stop the truck

In a further refinement of the method, the data of the position of the person are differentiated from data of further moving objects or persons in the case of an at least partial optical overlap by recognizing the image data with constant change as the position of the tracking person.

If a second person runs in the optical axis between the optical sensor and the tracked person and partially covers them, it can thus be recognized by the control device "which" person is the right one to carry on. The pixels of the second person have opposite the pixels of the first person an instant jump in the distance.

Advantageously, forecast values are determined for the future position of the person within which the position should be located.

From the course of the previous movement of a person or the pixels of the representation of this person as well as the values that are possible for the movement of a person in space, a prognosis value for temporally subsequent positions of the person to be tracked can be created. The person should therefore be in a certain position or within a certain range of positions after the previous movements up to the time of determining this forecast value for the future.

In a further development of the method, if the tracking person is optically covered by further objects or persons, the originally tracked person is detected by means of a search within the prognosis values.

For example, if a second person occurs in the optical axis in front of the tracked person, that it is completely covered at least for a short time, then a search using the forecast values makes it possible to track the person again capture. This applies at least, as long as the coverage does not last too long.

Advantageously, the position of a person to be tracked in a learning phase is detected by a movement of the person.

As a result, at the beginning of working with the truck in a simple manner, the person to be tracked can be identified by the control device. After the start, a short phase of the detection of a targeted movement of the serving or commissioning person, which can be triggered for example by a special circuit or command input in the truck.

The optical sensor can be a camera and the position of a person to be tracked can be detected by means of image processing methods.

There are many solutions available on the market for detecting people in the three-dimensional area that can be used here.

In a favorable embodiment of the method, the monitoring boundary on one side or on both sides of the industrial truck separates a front area of the monitoring area from a rear area with respect to a movement direction of the industrial truck.

During picking, an employee is usually taken from a shelf or containers in an area on the side of the truck and the goods are loaded onto the truck. Often, the truck is a pallet truck with a corresponding container or a pallet on the forks, which is moved forward with rearward oriented fork with a drawbar. Such a truck is then moved by the inventive method in an optimal way with the picking staff when the monitoring limit is positioned substantially transverse to the direction of travel in the front region of the vehicle.

Advantageously, the monitoring limit, starting from the truck perpendicular to the direction of movement of the truck to a maximum of 45 ° to Movement direction arranged.

Since the monitoring boundary, viewed from the industrial truck, is oriented laterally obliquely forward with respect to a direction of movement, the optical sensor can advantageously be arranged and yet the monitoring limit be located at the front of the industrial truck.

The truck can be moved with the steering set in the straight-ahead position.

In many warehouses with industrial hall floors such track control is often sufficient for rack aisles and no controlled control is required.

It can be carried out using the data of the optical sensor and / or the data of a personal protection device, a lane detection.

This allows targeted control of the steering of the truck. In this case, both the data of the optical sensor can be used, and also or alternatively data of a personal protection device, as far as such is provided and installed in the truck. Such personal protection devices are present in autonomous vehicles and monitor a nearby area for collisions with obstacles or people. These are usually based on monitoring at a low altitude above the ground, for example by laser scanners, which scan surfaces in accordance with standards at a height of approx. 150 mm above the ground.

In an advantageous development of the edge and the orientation of one or both sides of a rack aisle are detected and at a distance from the edge or both edges a target track is set.

For example, the shelf columns are detected as components of the shelves of the rack aisle. From the positions of the components of the shelf can be determined at a distance from a target track on which the truck should move. This setpoint track can also be placed in the middle of a rack aisle in double-sided shelves or, depending on the wishes of the serving person On one side of the rack aisle, so that the most ergonomic work possible. It is also possible that such a specification of a target track, an alignment of the truck takes place and it is not necessary at the beginning of order picking in a rack aisle, that the truck is aligned exactly by hand. Rather, the truck must only be driven into the rack aisle and aligns with the first movement movements to this rack aisle.

In an advantageous embodiment of the method, the end of a rack aisle is detected and terminated on reaching the end of the rack aisle the process.

The truck can then stop automatically or follow the succession to the person.

It can be detected from the data obstacles and the vehicle control can avoid this during the driving movement.

The data obtained by the optical sensor, in particular when it comes to three-dimensional data, in addition to the tracking of the person and the determination of a desired track also be used in general for the vehicle guidance, for example in the sense of an autonomously driving vehicle. Thus, with appropriate design of the control device, the truck can be designed so that after completion of a picking on request or by entering a command the truck automatically moves to a goods delivery and drops the container or pallet. This leads to a further time saving, since the commissioning person does not have to control the truck there. This is optimally supplemented if it is provided that the truck also drives autonomously on request to a first working position of a picking person.

Furthermore, it is also conceivable to use the data for a command control of the truck by gestures or movements of the tracked person. As a result, for example, further commands that do not affect the travel movement, such as a lifting movement, etc., are controlled. Finally, these data can also be used to detect and monitor the height of a load on a pallet on the forklift.

The object is also achieved by an industrial truck with an optical sensor having a monitoring area, which is connected to a control device, which performs a method described above.

The truck has the advantages already described.

Fig. 1

schematisch in Aufsicht ein Flurförderzeug, bei dem das erfindungsgemäße Verfahren zur Anwendung kommt, mit einer kommissionierenden Person,

Fig. 2

schematisch in Aufsicht das Flurförderzeug und die Person der Fig. 1 bei einer Fahrbewegung des Flurförderzeugs und

Fig. 3

das Flurförderzeug der Figur. 1 in einer seitlichen Ansicht.

Further advantages and details of the invention will be explained in more detail with reference to the embodiments illustrated in the schematic figures. This shows

Fig. 1

schematically in plan view of an industrial truck, in which the inventive method is used, with a commissioning person,

Fig. 2

schematic in plan the truck and the person of Fig. 1 at a driving movement of the truck and

Fig. 3

the truck of the Figure. 1 in a side view.

The Fig. 1 1 shows schematically in plan view an industrial truck 1, in which the method according to the invention is used, with a picking person 2. The industrial truck 1, here in the embodiment as a pallet truck with forwards directed forks in relation to a direction of movement, is located in a rack aisle 3 with a lateral load shelf 4, wherein a load shelf located on the other side is not shown. An observation area 7 is detected within the rack aisle 3 by an optical sensor 5, which in the present example is designed as a camera 6 with an all-round pick-up area and the ability to detect clearances. In a control, also not shown, the data of the optical sensor 5 are processed and the position of the person to be tracked 2 detected. In this case, a monitoring limit 8 is stored in the control device, which includes a front area 9 of the monitoring area 7 from a rear area 10 of the monitoring area 7 separates. While the truck is 1, the picking person 2 removes goods from the load rack 4 and places them on a transport pallet 11.

The Fig. 2 schematically shows in plan the truck 1 with the transport pallet 11 and the person 2 of Fig. 1 in a driving movement of the truck 1 in the direction of movement shown by the arrow. The truck 1 in the aisle 3 with the lateral load shelf 4 detected by the optical sensor 5 in the form of the camera 6, the monitoring area 7, which is separated by the monitoring boundary 8 in a front portion 9 of the monitoring area 7 and a rear portion 10 of the monitoring area 7 , The position of the person 2 is now in the front area 9 of the monitoring area 7. Therefore, the control device controls a traction drive of the truck 1 so long and moves the truck 1 in the direction of movement until the person 2 again behind the monitoring limit 8 in the Rear area 10 of the monitoring area 7 is located. In this case, the edge 12 of the load rack 4 is determined from the data of the optical sensor 5 and set at a distance from this edge 12 a target track on which the truck 1 is moved by the control device.

Advantageously, no briefing of the picking person 2 is required and follows the truck 1 of the person 2 without time-requiring operator actions of the person 2 would be required.

The Fig. 3 shows the truck 1 of Figure. 1 in a side view with the load shelf 4 of the rack aisle 3 and additional goods containers 13 from which goods can also be removed during order picking. The truck 1 with the optical sensor 5, which is designed as a camera 6, has the monitoring area 7. In this embodiment, there is a three-dimensional monitoring, since the camera 6 can record, for example as a time of flight camera (ToF) via runtime measurements, the distance of the pixels to the camera 6. In an alternative embodiment, simultaneously in the Fig. 3 is drawn, the optical sensor 5 can also be designed as a laser scanner 14, which scans in a plane 15 and thereby detects the distances to target objects. This corresponds to a two-dimensional detection.

Claims (17)

1. Method for controlling an industrial truck (1), in particular during picking operations, wherein the industrial truck (1) has an optical sensor (5) which has a monitoring range (7) and is connected to a control device, and the control device determines and tracks the position of a person (2) within the monitoring range (7) by evaluating the data of the optical sensor (5),
   characterized
   in that within the monitoring range (7) at least one monitoring limit (8) is defined and detected via the control device if the position of the person (2) has exceeded the monitoring limit (8), and the industrial truck (1) is moved by the control device by means of a traction drive of the industrial truck (1) until the position of the person (2) is again on the original side of the monitoring limit (8).
2. Method according to Claim 1,
   characterized
   in that the optical sensor (5) can detect distances from objects in the monitoring range (7).
3. Method according to Claim 2,
   characterized
   in that the optical sensor (5) can detect the monitoring range (7) in a three-dimensional fashion.
4. Method according to one of Claims 2 to 3,
   characterized
   in that changes in the position of the person (2) can be tracked, and the changes of the values can be evaluated as correct only in the case of continuous changes.
5. Method according to Claim 4,
   characterized
   in that the data of the position of the person (2) can be differentiated from data of further moving objects or persons in the case of at least partial visual overlap in that the image data with continuous change are detected as the position of the tracked person (2).
6. Method according to Claim 4 or 5,
   characterized
   in that prediction values within which the position should be located are determined for the future position of the person (2).
7. Method according to Claim 6,
   characterized
   in that in the case of visual overlap of the tracked person (2) by other objects or persons, the originally tracked person (2) is determined on the basis of a search within the prediction values.
8. The method according to one of Claims 1 to 7,
   characterized
   in that the position of a person (2) who is to be tracked is detected in a learning phase by a movement of the person.
9. Method according to one of Claims 1 to 8,
   characterized
   in that the optical sensor (5) is a camera (6), and the position of a person (2) who is to be tracked is detected by means of image processing methods.
10. Method according to one of Claims 1 to 9,
    characterized
    in that the monitoring limit (8) divides, on one or both sides of the industrial truck, a region (9) of the monitoring region (7), which is the front one in the direction of movement of the industrial truck, from a rear region (10).
11. Method according to Claim 10,
    characterized
    in that the monitoring limit (8) is arranged starting from the industrial truck (1) perpendicularly with respect to the direction of movement of the industrial truck (1) as far as a maximum of less than 45° with respect to the direction of movement.
12. Method according to one of Claims 1 to 11,
    characterized
    in that the industrial truck (1) is moved with the steering fixed in the straight-ahead position.
13. Method according to one of Claims 1 to 11,
    characterized
    in that lane detection is carried out for a setpoint lane of the industrial truck (1) to be followed, using the data of the optical sensor and/or the data of a person protection device.
14. Method according to Claim 13,
    characterized
    in that the edge (12) and the orientation of one side or of both sides of a rack aisle (3) are sensed, and a setpoint lane is defined at a distance from the edge (12) or both edges.
15. Method according to Claim 14,
    characterized
    in that the end of a rack aisle (3) is detected, and when the end of the rack aisle (3) is reached the method is ended.
16. Method according to one of Claims 13 to 15,
    characterized
    in that obstacles are detected on the basis of the data, and the vehicle controller avoids said obstacles during the driving movement.
17. Industrial truck having an optical sensor (5) with the monitoring range (7), wherein the optical sensor (5) is connected to a control device which carries out a method according to one of Claims 1 to 16.

Images