DE102017123877B4 - robot system - Google Patents

Abstract

Arrangement comprising a robot system for a forming machine and at least one mark (6) designed as an optical code, the robot system comprising:
a manipulator (2) suitable for manipulating a workpiece (3),
a musculoskeletal system (4) adapted to move the manipulator (2),
- A with the manipulator (2) coupled measuring device (5) - in particular a camera - which is adapted to detect a relative position between at least one arranged in the environment marker (6) and the manipulator (2), and
a calculation unit (7) connected to the measuring device, which is designed to calculate at least one correction value for a control and / or regulation of the musculoskeletal system (4) from the relative position,
wherein the at least one marker is suitable for detection by the measuring device (5),
characterized,
the at least one marker (6) is designed as a data matrix and / or QR code.

Description

The present invention relates to an assembly of a robot system for a forming machine and at least one designed as an optical code markers, wherein the robot system comprises a manipulator, suitable for manipulating a workpiece, and a musculoskeletal system, which is adapted to move the manipulator includes.

Manipulators are also referred to as "end-of-arm tools" (short EoAT, English: tool at the end of the arm) or as takeover heads and are used for receiving and storing of semi-finished products, workpieces and the like or in stations or machines. Manipulators may include holding devices for holding the workpieces and / or movable axes. In this case, stations are designated as areas in which semi-finished products can be provided or workpieces can be stored. Examples would be a sliding table with receiving positions for semi-finished products or a cooling table or conveyor belt for the storage of finished parts or workpieces.

As prior art applications and patents are known in which used by robot-mounted or externally mounted cameras to guide the robot.

• IBVS: Image Based Visual Servoing
• PBVS: Position Based Visual Servoing

There are in principle two different methods to control the difference between the recorded image and the robot position or rules. These methods are generally known by the term "visual servoing" and are subdivided into

• IBVS: Image Based Visual Servoing
• PBVS: Position Based Visual Servoing

These methods are used, for example, to follow parts or to pick up parts that have not been sorted ("handle in the box").

Especially in the application "handle in the box" are also 3D camera systems in use, which are joined together two images from different known positions to a three-dimensional information about the object in a computing unit.

Furthermore, methods are known to calibrate the robot to improve the absolute accuracy of a robot. These methods are either performed with external measuring systems such as laser interferometers or theodolites or are based on an iterative calculation of the JACOBIAN matrix by Newton's method or LQR method.

For example, an industrial robot system is equipped with a high-quality laser tracker system with special robot sequence programs and an interface to the laser tracker to determine the geometries of the manipulator used by the robot or the robot itself or correct by the inverse application.

In handling devices or linear robots is usually ignored by this calibration, since this would be particularly costly for the relatively large workspace and due to the design usually a backlash in the drive train is present, which complicates the use of these devices for accurate Cartesian processes.

These methods all have the disadvantages that on the one hand they are very complex to program and also very time-consuming to put into operation again for each application. A cost-efficient use in series machine construction is thus virtually eliminated.
pamphlet DE 102 49 786 A1 shows a method for referencing a robot to a workpiece or patient, with the following steps: First, at least one image of a workpiece (or patient) from at least two positions are recorded with at least one camera. This one camera is located on the robot, in particular on the robot arm, with which this also performs the later operations. As a workpiece while any objects and devices are referred to, which can be edited by a robot. For the medical technology field, but the application is by no means limited to this, the workpiece should be considered the patient or, more precisely, the surgical site. Consequently, at least two images are taken from different positions of this workpiece, these positions should be as far apart as possible, as this increases the accuracy of the process. An increase in the accuracy of the method can be further achieved by making more than two shots. According to the invention, it is further provided that a reference point on the workpiece is selected in at least one image. This reference point must be selected identically for the two images taken.

pamphlet US 2011/0 235 054 A1 shows the detection of articles by means of optically coded markers.

publications DE 10 2015 015 094 A1 . DE 60 2004 013 107 T2 . DE 103 45 743 A1 and US 2005/0 102 060 A1 show more robot or detection systems.

The object of the invention is to provide an arrangement of a robot system and a marker with respect to the prior art improved positioning - in particular with regard to the accuracy and / or the time to reach a certain position - provide.

This object is achieved by an arrangement having the features of claim 1.

• - ein mit dem Manipulator gekoppeltes Messgerät - insbesondere eine Kamera - welches dazu ausgebildet ist, eine Relativposition zwischen zumindest einem in der Umgebung angeordneten Marker und dem Manipulator zu erfassen, sowie
• - eine mit dem Messgerät verbundene Berechnungseinheit, welche dazu ausgebildet ist, aus der Relativposition zumindest einen Korrekturwert für eine Steuerung und/oder Regelung des Bewegungsapparats zu berechnen.

According to the invention, provision is made:

• a measuring device coupled to the manipulator, in particular a camera, which is designed to detect a relative position between at least one marker arranged in the surroundings and the manipulator, and
• a calculation unit connected to the measuring device, which is designed to calculate from the relative position at least one correction value for a control and / or regulation of the musculoskeletal system.

Protection is also desired for a molding machine having an arrangement according to the invention. Examples of molding machines are injection molding machines, transfer molding, pressing and the like.

• • Verbesserte Genauigkeit mit relativ einfachen, kosteneffizienten Mitteln
• • Absolut genaues Verfahren innerhalb eines bestimmtem Arbeitsbereichs
• • Einfache Programmierung von Robotern

Compared to the prior art, the following advantages result:

• • Improved accuracy with relatively simple, cost-effective means
• • Absolutely accurate procedure within a certain work area
• • Easy programming of robots

The measurement according to the invention of the relative position between the at least one marker and the manipulator takes place in at least one spatial direction but can also include all linear relative positions and one or more coordinates for establishing a relative orientation (eg angle). In other words, at least one relative coordinate is determined. Preferably, however, at least two or three relative coordinates (possibly together with relative orienting coordinates) are determined so that in the ideal case the relative position of an application in a plane is absolutely known.

For a general application (without limitation to one level) six relative coordinates (3 translations, 3 rotations) are necessary to determine the position and orientation of the marker. The relative position can be expressed, for example, as a deviation between a desired value and an actual value of an image position of the at least one marker or parts thereof.

As a musculoskeletal system, various types of robots can be used. Examples would be linear robots with two or three main axes arranged Cartesian and optional additional hand axes which are typically designed as rotary or pivot axes, Scara kinematics with two or three parallel axes of rotation and a linear axis, industrial robots with eg 6 axes, 7 axis robots or dual arm Robots with 15 or more axes.

Instead of optical centering with a camera, other non-contact systems are conceivable, which can interact as markers and measuring instruments. Examples would be electromagnetic systems based on transponders such as RFID chips. Magnetic cards are also possible. It may be advantageous if persistent digital identification features can be stored and preferably the non-contact passive readout of these features is possible. Alternatively, however, these identification features can also be stored in the control and / or regulating device of the musculoskeletal system or centrally on a server available for the robot control (as far as a connected cloud). This advantageously achieves a simpler or even central possibility for processing and changing the features without the read / write unit having to be in the vicinity of the transponder or RFID chip.

The coupling of the measuring device with the manipulator is preferably done by attaching the measuring device to the manipulator or in the area of the end-of-arm tooling near the manipulator such that movement of the manipulator (by the musculoskeletal system) also causes a corresponding movement of the measuring device , Of course, it can be provided that a relative positioning of the measuring device relative to the at least one marker can be changed or selected. This can be used to adapt the position of the measuring device with respect to the manipulator to expand the focusable area, but also, for example, to obtain depth information.

An important aspect is that, for the handling of parts always the relative position between the mounted on the musculoskeletal manipulator and the station is detected for the provision or recording of the parts to be handled. In contrast, when performing shaping processes usually an absolute position of the manipulator over a larger at least the process necessary area is necessary. But even in these applications, the invention can be used advantageously.

Further advantageous embodiments are defined in the dependent claims.

A control and / or regulating unit connected to the calculation unit can be provided for controlling and / or regulating at least one kinematic parameter of the musculoskeletal system-in particular at least one position.

At least one position sensor may be provided which is adapted to detect a position of the musculoskeletal system (also referred to as a "robot pose"), and the control and / or regulating unit may be adapted to use the musculoskeletal system by means of the controlling and / or regulating at least one position sensor measured position of the musculoskeletal system.

The calculation unit may be connected to the at least one position sensor and configured to take into account the measured position of the musculoskeletal system in the calculation of the correction value.

The control and / or regulating unit can be designed to use the correction value for determining and / or correcting a desired value occurring during the control and / or regulation of the musculoskeletal system. This can be done in the context of a higher-level control and / or regulation and allows an increase in the precision of the control and / or regulation.

It can be provided that the musculoskeletal system is designed to position the measuring device within the scope of a search travel programmed by an operator and / or programmatically such that the at least one marker lies in a detection area of the measuring device. An automatic or semiautomatic reading of the at least one marker or setting up the robot system is thereby possible.

It can be provided that the measuring device is designed to detect encoded information provided by the at least one marker and pass it on to the calculation unit and / or the control and / or regulating unit. The at least one marker can be designed to provide encoded information for reading by the measuring device.

In a particularly preferred embodiment, a desired value or a correction in the control and / or regulating unit is determined with the coded information and used for the calculation of the setpoint value used for the musculoskeletal system. This can be done in the context of a higher-level control and / or regulation and allows an increase in the precision of the control and / or regulation.

Correspondingly coded information can also be a position offset between the marker and a point of use for the manipulator. However, correspondingly coded information can also be further a number and / or the position offset of further relevant markers, wherein a plurality of markers form a coordinate system of a station or a correction system for linearizing the robot kinematics.
However, corresponding coded information can also contain control or movement commands for further process steps.

It can be provided that the measuring device is designed to determine from the at least one marker in the calculation unit the relative position of the marker to the measuring device and / or store it in the control and / or regulating unit such that starting up by means of the coded information of the Marker's coded position is possible.

In a particularly preferred embodiment, a unique identification (Id, number, link) is stored as coded information in the marker and can be used by the control and / or regulating unit for looking up values stored there. Correspondingly coded information can also be an address (link) in a central computer system (server, cloud, etc.), which contains further information about the process to be used, the parts or the production.

The information stored for the unique identification of the marker in the control and / or regulating unit and / or in a central computer system can also during the teaching process (teaching) of the best manipulator target position for performing the process from the position determined by the position sensor relative position or from the position sensor determined relative position and from further information from the control and / or regulating unit are determined and stored.

It can be provided that the control and / or regulating device performs the storing of the relative position when learning a sequence of movements by an interaction of a user.

The additional information in the marker can be coded both optically and in another way (for example via magnetic markers or an RFID tag as identification via a radio-frequency signal).

It may be provided that information about relative positions for other stations a movement sequence when learning positions of the musculoskeletal system are stored together with the relative position of the measuring device. The coding of the information in the marker can be avoided thereby, whereby simpler markers can be used.

It may be provided that the measuring device is designed to detect a plurality of markers, wherein the calculation unit is designed to define an at least two-dimensional coordinate system by the positions of the markers. Due to the increased number of reference points and possibly the reduced distance between them, in such an embodiment, a work area with very high accuracy for the control and / or regulation of the manipulator can be created. With an additional interpolation between the now serving as a marker markers, a calibrated handling system can be achieved with approximately linear scale and an improvement of the absolute positioning accuracy according to the invention.

A calibration system may be provided which is designed to detect a relative position between the measuring device and the manipulator, and that the calculation unit and / or the control and / or regulating unit is designed to determine the relative position in the calculation of the correction value to take into account. In particular, relative positions between the measuring device and parts of the manipulator or the manipulator kinematics can also be detected.

A calibration system can be used for two tasks. First of all, by imaging and evaluating a test pattern, the distortion of the optical system can be determined and subsequently corrected (this is also referred to as the intrinsic parameters). Furthermore, one can record the same test pattern from several robot positions (axis positions) and from this determine the exact mounting position and / or orientation of the camera on the manipulator (one speaks of the extrinsic parameters). A very simple test pattern would be given in the form of a regular grid with known dimensions (cf. Chessboard with black and white fields).

It can be provided that the measuring device is designed to detect a distance and / or a relative orientation between the at least one marker on the one hand and the measuring device and / or the manipulator on the other hand within the scope of the detection of the relative position. If, for example, a camera is used in conjunction with an optical marker, the size of the marker, as detected by the camera, can easily be used to obtain distance information. Also in embodiments with, for example, a magnetic marker, the distance can be measured directly or indirectly by the magnetic field strength.

By determining the distance between the at least one marker and the manipulator or measuring device, a relative position in all three spatial dimensions-and therefore also a correction value for all three spatial dimensions-can be obtained in a simple manner. By using two-dimensional geometries as markers-preferably standardized data matrix or QR codes-the position of the marker relative to the orientation of the manipulator or measuring device can also be determined.

If, in addition, the size of the marker selected for the particular application is known or stored as coded information in the marker, the distance determination or the robustness can be increased. The size of the marker could also vary due to the space available and then for each relevant position, the relevant area or relevant process in the robot program or the control and / or regulating unit during teaching (teaching) are deposited.

It can be provided that the manipulator is designed to receive and deposit a workpiece. In this case we speak of a handling robot (as opposed to an industrial robot).

The at least one marker can be designed to define at least two coordinate directions.

Preferably, the at least one marker can be embodied as an optical code in interaction with a camera as a measuring device. However, it is also conceivable to form the at least one marker as a predefined geometric object (rectangle, triangle, arrangement of circles, registration marks, etc.).

According to the invention, the at least one marker is designed as a data matrix or QR code.

As a result, a particularly simple two-dimensional design is possible.

It can be provided to use different types of markers for different tasks (filing, recording, removal, insertion) of the manipulator.

These types of markers can be standardized for several robot systems and, depending on the stored information (positions, sequences, Periphery / tool types, types of manipulators), be retrievable from a database.

Manipulators are also referred to as "end-of-arm tools" or in the special case of a manipulator for receiving and storing of semi-finished products, workpieces and the like as transfer heads. Manipulators may include holding devices for holding the workpieces and / or movable axes for the linear and rotational movement of the workpieces.

In a preferred embodiment, the at least one marker can be formed and / or positioned by elevations and / or depressions in a surface carrying the at least one marker. This can allow a particularly precise measurement of the relative position, in particular if the elevations and / or depressions are produced by means of CNC milling in a workbench or the like.

The determination according to the invention of the relative position between the at least one marker and the manipulator can also be used if the at least one marker is arranged on a movable component. The robot system can then be used to move the manipulator - especially synchronized - to move the component. A further intended use would be if at least one marker is arranged on a machine component which still oscillates after a rapid movement and the robot system can already synchronize the manipulator during the decay process to carry out the next process step. Thereby, the throughput of the production of the robot system can be increased.

The movable component may be formed by a part of the arrangement or a precursor of a part to be produced. It is also possible to arrange the marker on a handset to be moved by an operator, whereby the handset becomes a kind of remote control for the robot system. It can be used to move the robot system or "teach" movements, i. E. the executed movement or target position is stored in order to be carried out repeatedly, for example, in a cyclically running process. The moving component can be very simple and can be reduced to a printed marker (eg QR code printed on paper).

The robot system according to the invention may have a positional accuracy with deviations of less than 1 mm, preferably less than 5 tenths of a millimeter and particularly preferably less than 5 hundredths of a millimeter. The measuring device can be designed to detect the relative position with deviations of less than 1 mm, preferably less than 5 tenths of a millimeter and particularly preferably less than 5 hundredths of a millimeter.

• 1 ein Ausführungsbeispiel einer erfindungsgemäßen Anordnung,
• 2 das in 1 dargestellte, erfindungsgemäße Robotersystem aus einer anderen Perspektive,
• 3a ein Beispiel für die Ausbildung des zumindest einen Markers,
• 3b ein weiteres Ausführungsbeispiel mit zwei Markern,
• 3c ein weiteres Ausführungsbeispiel mit einer Vielzahl von Markern,
• 4a und 4b Ausführungen mit Markern, die durch Erhöhungen und/oder Vertiefungen gebildet sind,
• 4c bis 4e weitere Ausführungsbeispiele von Markern sowie
• 5 und 6 zwei weitere Ausführungsbeispiele für Anordnungen mehrerer Marker.

Further details and advantages of the invention will become apparent from the figures and the associated description of the figures. Showing:

• 1 an embodiment of an inventive arrangement,
• 2 this in 1 illustrated, inventive robot system from a different perspective,
• 3a an example of the training of the at least one marker,
• 3b another embodiment with two markers,
• 3c another embodiment with a plurality of markers,
• 4a and 4b Embodiments with markers formed by elevations and / or depressions,
• 4c to 4e further embodiments of markers as well
• 5 and 6 two further embodiments of arrangements of multiple markers.

In 1 is an arrangement according to the invention 10 from a robot system according to the invention 1 and markers 6 shown.

The markers 6 are in this case on a component 12 arranged. The component 12 is stored on a conveyor belt and can be moved thereby. The component has several places of use 11 on which workpieces 3 can be used.

The task of the robot system 1 is to recognize which job site 11 the component facing up and which workpiece 3 which form in the upward-facing deployment site 11 is to use. In 1 is the job site 11 a circular opening and the corresponding workpiece 3 is a cylinder with the base of a circular disk.

The robot system has it 1 over the meter 5 - in this case, a camera. But instead of complex image recognition but markers 6 used which at the component 6 are arranged. These can be detected by the camera and provide information about the offset (or a unique identifier on which of the Control system when teaching deposited offset can be retrieved), which between the respective marker 6 and the job site 11 is present, and about which workpiece 3 must be used.

The robot system 1 in 1 is arranged hanging. In 2 is the robot system 1 enlarged and shown from a perspective from below. Part of the musculoskeletal system 4 , the manipulator 2 and the meter 5 in the form of a camera are easy to recognize.

The calculation unit is also schematic 7 and the control unit 8th located. The calculation unit 7 is with the meter 5 connected and calculates the at least one correction value from the measured values. In addition, the calculation unit 7 with the control unit 8th connected.

In the embodiment shown here is the regulation of the musculoskeletal system 4 according to the position of the manipulator, for which (in drives of the musculoskeletal system 4 integrated and therefore not recognizable) position sensors on the musculoskeletal system 4 available.

The connection between these position sensors and the drive controls for the musculoskeletal system 4 on the one hand and the control unit 8th on the other hand is also shown schematically.

In the illustrated embodiment, measurements of the position sensors are used in the calculation of the correction value, for which they are provided by the control unit 8th to the calculation unit 7 be transmitted.

It should be noted that the calculation unit 7 and the control unit 8th do not need to represent separate physical objects. In many cases, both the calculation unit 7 and the control unit 8th be executed as program modules of a central system control. Of course it is also possible to use the calculation unit 7 and the control unit 8th to be carried out separately. For example, the control of the musculoskeletal system 4 in the drive modules of the musculoskeletal system 4 be integrated. In other embodiments, the entire calculation unit or subtasks of the calculation unit may be integrated in the camera system.

The manipulator 2 has at least one rotary axis in the present case, around a corresponding workpiece 3 to place accordingly. The linear axes present in this embodiment are not shown for the sake of clarity.

The measuring device 5 is mounted in front of the axis of rotation of the manipulator 2, so that an image plane with the plane of the manipulator 2 best possible covers.

In the embodiment of the 1 and 2 can be the insertion of the workpiece 3 into the job site 11 both at a standstill component 12 (ie at dormant conveyor belt) as well as moving component 12 (with movement of the conveyor belt) happen. The movable conveyor belt is part of the station, consisting of several conveyor belts.

To 3a By using at least one two-dimensional code as at least one marker which is attached to the workbench (storage / receiving station) or the forming machine (insertion station) and detected by an optical camera mounted on the manipulator, the deviations or vibrations existing on the manipulator become in comparison with the mere control and / or regulation of the musculoskeletal system improves without disadvantages as they occur in the prior art to accept.

Since the workbench of the robot system is typically constructed from CNC-machined parts, in this step, a positioning aid in the form of at least one marker according to the invention for the image codes to be applied in a further step can preferably be provided in a recess or a border.

An optical camera for identifying at least one two-dimensional or multi-dimensional marker in the form of an image code (also called "data matrix code") is carried along on the manipulator.

The image code can also be combined with a radio- or magnet-based code (RFID).

The data matrix codes are, for example, attached to system components (for example: peripherals, conveyor belt, tools, etc.). A preferred possibility is to produce different data matrix codes in a shaping machine (injection molding machine, marking laser, 3D printer, etc.) from two components with different properties (for example, colors). This can be a series process for a manageable number of different codes or produced in an additive manufacturing process (lot size 1: 3D printing, etc.). In this case, materials with particularly suitable surfaces (no reflection) with sufficient robustness against scratches, contamination, humidity or the like can be used.

If, for example, the at least one marker is attached in the vicinity of a part to be deposited or picked up, the robot can store the absolute position offset once. In this embodiment, the position offset consists of a two-dimensional distance and an angle relative to the data matrix code. If, for some reason, the pick-up / drop surface changes relative to the robot coordinate system, re-teach-in does not need to take place since the robot only deposits or picks up the corresponding part based on the Datamatrix code. The prerequisite for this, of course, is that the distance between the storage point or pick-up point and the data matrix code no longer changes, but this is easy thanks to the CNC machining of the station and the permanent attachment of the marker (gluing, insertion, press-fitting, screwing, etc.) can be achieved.

The examination of the data matrix code can take place acyclically preferably after changes to the configuration data (parts data) or after filling and positioning of the provision system (eg, sliding table) or cyclically preferably at relatively long intervals or particularly preferably after temperature changes. For reasons of optimization, a check of the data matrix code is dispensed with in the remaining cycles and it is also possible to record the stored position offset to the destination to be approached without having to approach intermediate positions. In a particularly secure variant, the measurement of the relative position and / or the determination of the at least one correction value is carried out in each cycle of a cyclically running process.

The initial finding of the data matrix code can be solved either manually or automatically by means of a search.

Finding the data matrix code with additional RFID code can be simplified with the aid of an RFID receiver, and preferably from a greater distance.

The position offset can either be stored in the robot program, optically integrated into the code or stored in a combined RFID tag or on a dedicated server. The retrieval is thus either directly by means of optical or electromagnetic measuring device or indirectly with the example, optically determined identifier of the at least one marker as a key for a table (database).

To 3b If several, preferably at least three, data matrix codes are arranged in one receiving area or storage area, then an improved positioning can be achieved over the entire area bounded by the data matrix codes (within certain limits, of course, also outside thereof). By starting and calibrating the data matrix codes, the robot can automatically correct the absolute errors of the manipulator in the area of the data matrix codes. From the measurement data, a detailed two- or multi-dimensional coordinate system can be created for the work surface identified by the codes. If the work surface is moved to a different position, preferably after a configuration change, no re-measurement must take place. All you have to do is find the appropriate marker (s).

The third dimension of the correction grid is either from the marker size imaged in the camera (smaller marker is farther away, larger marker is closer) via movement of the camera, the depth information of the camera (image size of the code), and / or additional in the third level additional markers are determined.
The data matrix codes or the RFID-extended data matrix codes may contain additional information (eg: program parameters or even program sequences) which enable the robot system to place or remove parts in any machine part without them being explicitly specified in the section Robot controller are programmed (intelligent marker). As a result, a storage of parts data in the markers in the individual stations can be achieved, whereby after a conversion, the usually necessary change to the parameters can be omitted.

Additional information may include images, positions, programming commands, temperature curves, position offset, etc.

However, further possibilities for intelligent markers could also be that the size or a derived size information of the at least one marker (for example 10 × 10 mm or 20 × 20 mm) is supplemented in the optical code of the marker. This would be expedient if, for example, for reasons of space technology, different marker sizes are used in order to determine the distance of the manipulator as accurately as possible even with different marker sizes.

A further embodiment of an intelligent marker would be in particular, when using multiple markers eg. For the described more accurate orientation of the coordinate system in spatially extended storage or recording areas, information about the other markers or the number of other markers and their approximate position with respect to to supplement the present marker. As a result, the first detected marker would a priori have the information as to how many additional markers are present, as a result of which search journeys can be saved by finding the other markers without losing time.

Another possibility for an intelligent marker would be that production information (time stamp, ambient conditions, parts counter or batch number, current production specification values, actual production values, etc.) is stored in the at least one marker. This can take place on the manipulator, in the production cell, preferably on the shaping machine (in particular injection molding machine) or on a centralized system (Manufacturer Execution System,...). The next processing station can thus read out, reuse and supplement the externally stored information from the key stored in the at least one marker.

Another possibility of an intelligent marker would be an active intelligent marker consisting of a display for displaying the 2-dimensional marker, preferably an e-ink display (electronic paper). This would allow the marker to adjust its (additional) information, such as the number of parts removed or added by the manipulator In this way, information relating to a unit of delivery stocked with different parts numbers or a semifinished product tray could be taken directly from the at least one marker This information could also be stored on the central system to the code of the at least one marker.

Another possibility of an intelligent marker would be an active intelligent marker equipped with a communication option with the manipulator and other processing stations, preferably via a communication standard, for example Near Field Communication (NFC) and a printed code or optionally with a display. The robot system could, in particular via the manipulator, pass on further information in a similar manner as previously described to the at least one marker. The at least one marker could then store this information and, for example, make another processing station available in a linked production. In this case, the at least one marker can provide this changeable information optionally via a display or via the communication option.

To 3c : By using a larger number of Datamatrix codes, the nonlinearities in the position scale of the manipulator can be compensated:

The marker can also be mounted on a moving machine or tool part and thus a synchronization of the robot to this movement can be achieved.

To 4a and 4b : By projecting spines as a form of elevation ( 4a , Spines circular) or surface depressions ( 4a and 4b) markers, for example in the form of QR codes, can be positioned or even formed with high precision.

To 4c . 4d . 4e : In principle, there are also markers 6 without the possibility to code additional information optically possible. One possibility is to store the coding in an RfID chip or other medium. The image processing algorithms are particularly simple if only the 2D or 3D position of the marker 6 must be determined.

To 5 : Several (as markers 6 used) registration marks which are evaluated exclusively according to the position (and not the orientation) to a coordinate system and also for scale correction of the robot system 1 be used.

To 6 : By an additional optically evaluable scale in the form of a ruler 14 can the position of the robot system or the position of the musculoskeletal system 4 be corrected continuously. It is sufficient if the correction is acyclic after changes to the robot system 1 or the station or the robot system 1 and the station surrounding conditions (temperature, etc.) is executed.

For the sake of clarity, not all the elements are always provided with reference symbols in the figures if an element in the corresponding figure has already been provided with the corresponding reference symbol. This concerns in particular the markers 6 , the workpieces 3 and the job sites 11 ,

Claims (22)

Arrangement of a robot system for a forming machine and at least one designed as an optical code marker (6), the robot system comprising: - a manipulator (2), suitable for manipulating a workpiece (3), - a musculoskeletal system (4), which is formed, the manipulator (2) to move, - with the manipulator (2) coupled measuring device (5) - in particular a camera - which is adapted to a relative position between at least one arranged in the environment marker (6) and the manipulator (2) to detect, and - a calculation unit (7) connected to the measuring device, which is designed to calculate at least one correction value for a control and / or regulation of the musculoskeletal system (4) from the relative position, wherein the at least one marker is suitable for detection by the measuring device (5), characterized in that the at least one marker (6) is designed as a data matrix and / or QR code. Arrangement according to Claim 1 , characterized in that a control and / or regulating unit (8) connected to the calculation unit (7) is provided for controlling and / or regulating at least one kinematic parameter of the musculoskeletal system (4) - in particular at least one position. Arrangement according to Claim 2 , characterized in that at least one position sensor is provided which is adapted to detect a position of the musculoskeletal system (4), and that the control and / or regulating unit (8) is adapted to the musculoskeletal system (4) using the be controlled by the at least one position sensor measured position of the musculoskeletal system (4) and / or to regulate. Arrangement according to Claim 3 , characterized in that the calculation unit (7) is connected to the at least one position sensor and is adapted to take into account the measured position of the musculoskeletal system (4) in the calculation of the correction value. Arrangement according to one of Claims 2 to 4 , characterized in that the control and / or regulating unit (8) is adapted to use the correction value for determining and / or correcting a desired value occurring in the control and / or regulation of the musculoskeletal system (4). Arrangement according to one of the preceding claims, characterized in that the musculoskeletal system (4) is designed to position the measuring device (5) in the context of a search travel programmed by an operator and / or programmatically such that the at least one marker (6) in a detection range of the measuring device (5). Arrangement according to one of the preceding claims, characterized in that the measuring device (5) is adapted to detect the at least one marker (6) provided coded information and to the calculation unit (7) and / or the control and / or regulating unit ( 8) pass on. Arrangement according to Claim 7 , characterized in that the measuring device (5) is designed to determine from the at least one marker (6) in the calculation unit (7) the relative position of the marker (6) to the measuring device (5) and / or in the control and / or or control unit (8) store so that a start of coded by the coded information of the marker (6) position is possible. Arrangement after the Claim 8 , characterized in that the control and / or regulating device performs the storing of the relative position when learning a sequence of movements by an interaction of a user. Arrangement according to one of the preceding claims, characterized in that the measuring device (5) is adapted to detect a plurality of markers (6), wherein the calculation unit (7) is adapted to, by the positions of the marker (6) an at least two-dimensional coordinate system set. Arrangement according to one of the preceding claims, characterized in that a calibration system is provided, which is designed to detect a relative position between the measuring device (5) and the manipulator (2), and that the calculation unit (7) and / or the Control and / or regulating unit (8) is adapted to take into account the relative position in the calculation of the correction value. Arrangement according to one of the preceding claims, characterized in that the measuring device (5) is designed, within the scope of the detection of the relative position, a distance and / or relative orientation between the at least one marker (6) on the one hand and the measuring device (5) and On the other hand, to detect the manipulator (2). Arrangement according to one of the preceding claims, characterized in that the manipulator (2) for receiving and depositing a workpiece (3) is formed. Arrangement according to one of the preceding claims, characterized in that the at least one marker (6) is designed to define at least two coordinate directions. Arrangement according to one of the preceding claims, characterized in that the at least one marker (6) is designed with integrated intelligence. Arrangement according to one of the preceding claims, characterized in that the at least one marker (6) is formed and / or positioned by elevations and / or depressions in a surface bearing the at least one marker (6). Arrangement according to one of the preceding claims, characterized in that the at least one marker (6) is designed to provide encoded information for reading by the measuring device (5). Arrangement according to Claim 17 , characterized in that the at least one marker (6) is adapted to provide a position offset between the marker (6) and a point of use (11) for the manipulator (2) as information for the measuring device (5). Arrangement according to one of the preceding claims, characterized in that information about relative positions for further stations of a movement sequence when learning positions of the musculoskeletal system (4) are stored together with the relative position of the measuring device (5). Arrangement according to one of the preceding claims, characterized in that the at least one marker (6) on a movable component (12) is arranged and that the robot system (1) is adapted to the manipulator (2) tuned - in particular synchronized - with the Move component (12). Arrangement according to one of the preceding claims, characterized in that the at least one marker (6) is a 2-component injection-molded part. Shaping machine with an arrangement according to one of the preceding claims.

Images