Method for controlling a kinematic system of an automatic machine for processing or packing articles and respective automatic machine

Abstract

A method for controlling a kinematic system (3), in particular a robot (15) or a manipulator (14), of an automatic machine (1) for processing or packing articles (2) and comprising the steps of: determining a mechanical configuration of the kinematic system (3); define, depending on the mechanical configuration of the kinematic system (3), a cloud (CP) of points (P) representing a set of possible positions in the joint space (JS) that the kinematic system (3) can reach during the performing of the motion profile (MP); wherein each of the points (P) corresponds to a precomputed possible position (q1, q2, q3) of the kinematic system (3) in the joint space (JS); wherein each of the points (P) corresponds to at least one precomputed position (f(P)) of the kinematic system (3) in the workspace (WS); and, in use, the sequential and cyclic steps of: detecting the current position (q1a, q2a, q3a) of the at least one electric actuator (6); define, depending on the current position (q1a, q2a, q3a) of the at least one electric actuator (6), a point (PA) corresponding to said current position by using said cloud (CP) of points as a reference; computing, as direct kinematics, a respective effective position (f(q1a, q2a, q3a)) of the kinematic system (3) in the workspace (WS) due to the corresponding point (PA) defined by using said cloud (CP) of points (P) as a reference; control the kinematic system (3), by way of the control system (9), to perform said motion profile (MP) as a function of said effective position (f(q1a, q2a, q3a)) of said calculated kinematic system (3) in the workspace (WS).

Description

[0001] “METHOD FOR CONTROLLING A KINEMATIC SYSTEM OF AN AUTOMATIC MACHINE FOR PROCESSING OR PACKING ARTICLES AND RESPECTIVE AUTOMATIC MACHINE”

[0002] Cross-Reference to Related Patent Applications

[0003] This patent application claims priority of the Italian patent application No. 102024000015472 filed on July 4, 2024, the content of which is incorporated by reference herein.

[0004] Technical Field

[0005] The present invention relates to a method for controlling a kinematic system, in particular a robot or manipulator, of an automatic machine for processing or packing articles and to a respective automatic machine.

[0006] The term “articles” refers to consumer articles or semi-finished products. In particular, the present invention finds advantageous application in the control of a kinematic system of an automatic machine for processing or packing articles of the tobacco industry (such as, for example, traditional or heat-not- burn cigarettes, filters, electronic cigarettes, cigars, snus). Alternatively, the articles could be articles of the food or pharmaceutical field or absorbent hygiene products.

[0007] The present invention finds application, for example, in a kinematic system of a machine for processing aerosol-generating material to form pieces of aerosol-generating material intended to be components of tobacco products.

[0008] Furthermore, the present invention finds advantageous but not limitative application in a kinematic system of an automatic machine for packing tobacco products, for example a packing machine that produces packets of cigarettes and in the method for controlling the same, by monitoring the production steps thereof, to which the following description will make explicit reference, without losing generality, in being applicable any type of automatic machine for processing or packing articles.

[0009] Prior Art

[0010] Automatic machines for producing or packing of (consumer / finished or semifinished) articles are known. Usually, said automatic machines comprise a plurality of movable operating members that act on articles (for example packets of cigarettes, foodstuffs, absorbent hygiene products., etc.) to modify the shape, structure, or position thereof. Generally, the movable operating members are mechanical parts of different shapes and sizes suitable for processing consumer articles.

[0011] Sometimes, said movable operating members are programmable trajectory kinematics used as construction elements of the so-called parallel robots, namely, mechanical systems with a plurality of arms controlled by a computer to support and move an end-effector.

[0012] In particular, the prior art comprises the so-called Delta robots, a known type of parallel robot formed by three arms, each of which is operated by a respective actuator mounted on a base.

[0013] In particular, the document WO2019166923 describes, among the possible embodiments, a Delta robot provided with non-spherical joints, namely, with non-incident rotation axes (for example universal joints).

[0014] As is known, the control of a kinematic system, for example of the aforementioned Delta robot, is based on a target position value in the workspace, which is given as input to a control system that uses an inverse kinematic transformation to obtain the desired value in the joint space, which is controlled by the respective actuators of the kinematics. In feedback, the current positions of the joints are detected (therefore the coordinates in the joint space) which, by means of the control system, are, by way of a direct kinematic transformation, converted again from the workspace to the joint space, in order to be compared with the next target position value, to calculate the tracking error and cyclically implement the respective control.

[0015] The particular kinematic structure of the aforementioned Delta robot has led the Applicant to develop its direct kinematics, to carry out the aforementioned control of the kinematic system, which however consists of a system of equations with a degree greater than thirtieth. This means that, although it is possible to solve said systems, the same requires considerable computational resources, which limit the application to very slow dynamics (with a cycle time of a few seconds), certainly compromising accurate control with cycle times lower than a second, or even half a second.

[0016] There is, therefore, a need to obtain a control method for a kinematic system that facilitates the calculation of direct kinematics, allowing rapid and precise controls even in complex kinematic systems such as, but not limited to, the aforementioned Delta robot having non-spherical joints at both ends of its links.

[0017] Description of the Invention

[0018] The object of the present invention is to provide a method for controlling a kinematic system, in particular a robot or a manipulator, of an automatic machine for processing or packing articles and a respective automatic machine.

[0019] According to the present invention, a method for controlling a kinematic system, in particular a robot or a manipulator, of an automatic machine for processing or packing articles and a respective automatic machine is provided as claimed in the attached claims.

[0020] The claims describe preferred embodiments of the present invention and form an integral part of the present description.

[0021] Brief Description of the Drawings

[0022] The present invention will now be described with reference to the attached drawings, which illustrate some non-limiting embodiment thereof, wherein:

[0023] • Figure 1 schematically illustrates an automatic machine which is the object of the present invention and that performs a first part of the method according to an embodiment of the present invention;

[0024] • Figure 2 schematically illustrates a part of the automatic machine of Figure 1 that performs a second part of the method according to an embodiment of the present invention;

[0025] • Figure 3 schematically illustrates the transition from the joint space to the workspace according to a first embodiment of the method according to the present invention;

[0026] • Figure 4 schematically illustrates the transition from the joint space to the workspace according to a first embodiment of the method according to the present invention;

[0027] • Figures 5 illustrate a block diagram for the control of a kinematic system according to an embodiment of a control method according to the present invention.

[0028] Preferred Embodiments of the Invention

[0029] Figure 1 illustrates an automatic machine 1 for processing or packing articles 2, in particular articles of the tobacco industry. Obviously, the present invention can also be applied to different machines for the production or packing of (finished / consumer or semi-finished) articles.

[0030] The same reference numbers and letters in the figures identify the same elements or components having the same function.

[0031] In the context of the present description, the term “second” component does not imply the presence of a “first” component. These terms are in fact used as labels to improve clarity and are not to be considered in a limiting manner.

[0032] The elements and features illustrated in the various preferred embodiments, including the drawings, may be combined with one another without departing from the scope of protection of the present application as described in the following.

[0033] It should be noted that in the following of the present description, terms such as “upper”, “lower”, “front”, “back” and the like are used with reference to conditions of normal movement of the kinematic system 3 in the workspace (with Cartesian reference).

[0034] As illustrated in the non-limiting embodiment of Figure 1 , it is also possible to define:

[0035] - a longitudinal axis X;

[0036] - a transverse axis Y arranged, in use, horizontal and orthogonal to the axis X; and

[0037] - a vertical axis Z and arranged, in use, vertical and orthogonal to the axes X, Y.

[0038] The machine 1 comprises a kinematic system 3, which in turn comprises: at least one arm 4 by way of which to perform at least one processing on the articles 2 (for example a gripping and positioning operation from one conveyor 5 to another); at least one respective electric actuator 6, preferably comprising an electric motor 7, which is mechanically connected to the at least one arm 4 at a respective universal joint 8; and a control system 9, electrically connected to the electric actuator 6 and configured to control the at least one arm 4 to perform a motion profile MP in the workspace that allows said operation to be carried out.

[0039] The term “processing” means the mechanical transformation of the article 2 in shape, size, components, or position.

[0040] The automatic machine 1 comprises one or more end effector devices 10, such as grippers, drums, pushers, etc., which are arranged at a distal end 11 of the kinematic system 3 and carry out processing of articles, namely, production and / or packing operations, on the articles 2 (which in the non-limiting embodiment illustrated in Figure 1 are packets of cigarettes).

[0041] Advantageously but not limited to, and as illustrated in one of the examples of Figure 1 , the kinematic system 3 is a Delta robot, in particular of the type described in document WO2019166923.

[0042] In other non-limiting examples, the kinematic system is a manipulator robot, or any other type of kinematic system.

[0043] The machine 1 described so far, in particular the control system 9, is configured / programmed to perform the method described in the following.

[0044] In fact, according to a further aspect of the present invention, a control method of a kinematic system 3 is provided, in particular of a robot or manipulator, for the automatic machine 1 for processing or packing articles 2.

[0045] The method provides, as illustrated in the non-limiting embodiment of Figure 1 , a step 12 of determining a mechanical configuration of the kinematic system 3. In particular, in this step the arms 4 that form the kinematic system 3 are defined and how they are rotary / l i nearly connected to one another by way of the joints 8. Preferably, the joints 8 are rotary joints, configured to rotate a respective arm 4 around a respective rotation axis relative to another arm or with respect to a frame.

[0046] In addition, the method comprises a step 13 of defining, depending on the mechanical configuration of the kinematic system 3, a cloud CP of points P (schematically illustrated in the central part of Figure 2) representing a set of possible positions in the joint space JS that the kinematic system 3 can reach during the performing of the motion profile MP. In particular, therefore, the cloud CP of points P is generated, for example by way of a simulation system PC, before the kinematic system 3 is operated, more specifically in the design step of the automatic machine 1 or during the first start of production of the automatic machine 1 .

[0047] In this text, the term "joint space" JS (or configuration space) means, as per convention, the space in which the vector q of the joint variables is defined (formed by the position of the individual joints 8, in particular, therefore, in a derivative manner, by the position of the electric motors 7). Its size is usually indicated with N.

[0048] Otherwise, the term “workspace” WS (or Cartesian space) means, as per convention, that space in which the vector (in this case three-dimensional) of the end effector device 10 is defined in the environment in which the same operates (having coordinates x, y, z relative to the axes X, Y, Z mentioned above).

[0049] Advantageously, each of the points P of the cloud CP corresponds to at least one precomputed possible position of the kinematic system 3 in the joint space JS. In other words, the points P of the precomputed cloud CP of points P, represent possible / probable positions, within a predefined tolerance range in the vicinity of the motion profile MP in which the end effector device 10 can be found during the performing of the processing operations on the articles 2.

[0050] In detail, each of the points P in the joint space JS also corresponds to a precomputed position f(P) (before the performing of the processing operations and therefore of the motion profile MP by the kinematic system 3) of the kinematic system 3 in the workspace WS. In other words, the cloud CP of points therefore allows the problem to be moved from the calculation of the direct kinematics of complex systems (if desired, also applicable to simple systems) to the determination of the point of the cloud closest to the current position of the kinematic system 3.

[0051] Advantageously, therefore, once steps 12 and 13 have been performed (one- off), the method comprises, in use, or during the normal operation of the automatic machine 1 (that is, when the same carries out the processing on the articles 2) the sequential and cyclic steps of: detecting the current position q1a, q2a, q3a of the at least one electric actuator 6; defining, depending on the current position q1 a, q2a, q3a of the at least one electric actuator 6, a point PA corresponding to said current position q1a, q2a, q3a by using said cloud CP of points as a reference;

[0052] - calculating as / instead of the direct kinematics, a respective effective position f(q1 a, q2a, q3a) of the kinematic system in the workspace WS depending on the corresponding point PA defined by using said cloud CP of points as a reference; controlling the kinematic system 3, by way of the control system 9, to perform said motion profile MP as a function of said calculated effective position f(q1 a, q2a, q3a) of said kinematic system 3 in the workspace WS.

[0053] In this way, it is possible to ignore the complexity of the kinematic structure of the kinematic system 3, and instead estimating where the end effector is located, namely, the end effector device 10, depending on the position of the at least one electric motor 7 and of the precomputed cloud CP of points.

[0054] Advantageously but not limited to, and as illustrated in the non-limiting attached embodiments, the kinematic system 3 comprises a manipulator 14 provided with a plurality of arms 4, in particular three arms 4, connected to one another by respective universal joints 8 in a serial manner (for example anthropomorphic) or in a parallel manner (for example a Delta robot). In particular, in these cases, each arm is moved by a respective electric actuator 6.

[0055] Advantageously but not limited to, the cloud CP of points extends over a number of dimensions equal to the number of electric actuators 6 provided for the movement of the kinematic system 3. For example, in the non-limiting embodiment of Figures 3 and 4, the cloud CP is three-dimensional as the kinematic system 3 comprises three electric motors 7.

[0056] According to some preferred embodiments, such as the one illustrated in Figure 4, the points P of said cloud CP of points P are equidistant from one another in the joint space JS forming a lattice (preferably three-dimensional or even with a greater number of dimensions) that can be subdivided into multidimensional figures PDF (such as the cubes illustrated in Figure 4, one of which is highlighted in grey) delimited by a number n (eight in Figure 4, since it is a cube) of vertices V corresponding to the points P of the cloud CP of points. In particular, during the step of defining the corresponding point PA in the joint space JS, the method provides for detecting the position q1 a, q2a, q3a of each of the electric actuators 6 and, due to said positions q1 a, q2a, q3a, the multidimensional figure PDF is identified, in particular the cube, wherein the point PA corresponding to said current position q1a, q2a, q3a is located.

[0057] In other words, the cloud CP of points is defined by a plurality of points P (each having coordinates q1 , q2, q3), which define areas, namely, the figures PDF, which can be identified because, for each area, the value of each position q1a, q2a, q3a falls within a respective interval. Therefore, taking the example of Figure 4 as a reference, finding that q1 a is comprised between qT and q1 ” (precalculated as they belong to the cloud CP of points) it is possible to understand in which layer of figures PDF the corresponding point PA is located. Similarly, finding that q2a is comprised between q2’ and q2” (precalculated as they belong to the cloud CP of points) it is possible to understand in which row of figures PDF the corresponding point PA is located. Finally, finding that q3a is comprised between q3’ and q3” (precalculated as they belong to the cloud CP of points) it is possible to understand in which figure PDF the corresponding point PA is located.

[0058] Preferably but not limited to, and as illustrated in the non-limiting case of Figure 4, the point PA corresponding to said current position q1 a, q2a, q3a, is identified / calculated as a function of its respective distance dqkfrom the vertices V, P of the multidimensional figure PDF wherein the same is located. In particular, therefore, the control system 9 is configured, among other things, to calculate the respective distance dqkfrom the eight vertices V, P of the cube containing the point PA corresponding to the positions of the motors 7 in the joint space. In this way, it is possible to uniquely define the position in the joint space JS of the point PA with respect to the predefined cloud CP of points. Therefore, the position of the corresponding point PA is obtained with respect to a precalculated dataset available at the time of performing of the motion profile MP.

[0059] Advantageously but not limited to, due to the corresponding point PA, a precompiled look-up table LUT is consulted, correlating positions P in the joint space JS to respective positions f(P) in the workspace WS.

[0060] Preferably, furthermore, the look-up table LUT is a structure that, as a function, contains both the position f(qk) and the velocity, namely, the Jacobian J(qk), where qk=(q1 k, q2k, q3k).

[0061] Therefore, the method preferably involves consulting the look-up table LUT to determine the area of the workspace WS corresponding to the multidimensional figure PDF, or the cube in the case of a Delta robot with three actuators. In this way, therefore, it will be necessary to define, within an area as small as the resolution of the lattice formed by the figures PDF, where the point f(q1 a, q2a, q3a) is located. In particular, consistently, the smaller the volume of the figure PDF, or of the cube, the smaller the area in which the end effector device 10 can be located in the workspace WS.

[0062] In any case, advantageously but not limited to, the method involves transforming the vectors dqkfrom the joint space JS to the workspace WS. Said transformation can be obtained by using the formula f qk~) +Jk* dqk), where the vector qk is transformed from the joint space JS to the workspace WS and due to the Jacobian contained in the look-up table LUT.

[0063] In particular, once the distance dqkhas been calculated for each vertex V of the figure PDF, the control system 9 is configured to calculate, in light of what is extracted from the LUT, for each of the vertices V, P of the figure PDF, for example of the identified cube, the following function

[0064] In this way, by averaging / interpolating

[0065] (where n is the number of vertices of the figure PDF; in the case of Figure 4, n=8) and in a few steps the function that allows to solve, without having to resort to complex calculation systems, the direct kinematics of the kinematic system 3, regardless of its complexity is obtained. Furthermore, due to the interpolation between the calculated transforms, the position error is further reduced, thus allowing the achievement of values lower than a tenth of a millimetre. According to other non-limiting cases, as illustrated schematically in the embodiment of Figure 3, the points P of said cloud CP of points are differently spaced from one another. In this case, it is more complex to use a specific lookup table LUT. On the contrary, the definition of the point PA corresponding to said current position requires the use of space partitioning algorithms, in particular of the k-d tree type (known in itself, especially from the field of vision systems and therefore not detailed in the following). In this case, therefore, an iterative algorithm that allows to converge towards the desired solution could be preferable. In other words, these embodiments involve detecting the positions q1 , q2, q3 of the electric motors 7 and establishing, based on the closest points of the precomputed cloud CP of points, where the end effector device 10 is presumably located in the workspace WS. Once said point has been identified, the reverse step is performed, calculating (by way of inverse kinematics) in which position the electric motors 7 should be in order to determine that position in the workspace WS. Having compared the deviation with the detected positions q1 , q2, q3, we proceed in an operational manner until reaching a value below a predefined threshold, for example with an error of less than half a millimetre.

[0066] Preferably but not limited to, the kinematic system 3 is a Delta robot 15 comprising:

[0067] - three arms 4;

[0068] - three electric actuators 6, connected to respective first ends 16 of the arms 4 and configured to move the three arms 4;

[0069] - an end effector device 10 connected to respective second ends 17, opposite to the first ends 16 of the three arms 4 and moveable by the electric actuators 6 by way of the arms 4; wherein the three arms 4 are connected to the end effector device 10 and to the respective electric actuator 6 by way of respective universal joints 8.

[0070] In the non-limiting embodiment of Figure 5, a flow chart of the control of a kinematic system 3 with three different actuators, such as the previously described robot 15, is schematically illustrated. In particular, a set of three target coordinates x*, y* z* is ordered for the end effector of the kinematic system 3. Subsequently, a trajectory generator 18 performs the inverse kinematics and therefore generates respective position setpoints q1*, q2*, q3* in the joint space JS for each of the three electric motors 7 to be controlled. Said values are fed to respective position controllers 19 (defining the so-called position loop). The output of said controllers 19 is fed to respective speed controllers 20 configured to determine a drive current for each of the motors 7. The current position q1a, q2a, q3a of each of the electric motors 7 is read by a respective encoder 21 and sent as feedback to the position controllers 19 and speed controllers 20 (deriving the same with respect to time). Furthermore, the current position q1 a, q2a, q3a is sent to a block 22 (which represents, for example, part of the industrial PC or PLC of machine 1 ) which is responsible for computing the direct kinematics as previously described, in light of the precomputed cloud CP of points and the pre-loaded look-up table LUT (for example when the automatic machine 1 is switched on).

[0071] Although the invention described above refers precisely to a specific example of embodiment, it is not to be considered limited to this example of embodiment, since its scope comprises all those alternatives, modifications or simplifications that would be clear to the expert technician in the field, such as for example: the addition of further actuators, another type of automatic machine, a different shape of the motion profiles, a different kinematic system, etc.

[0072] The present invention has multiple advantages.

[0073] First of all, it allows to improve the efficiency of the control of electric motors, especially in the case of complex kinematics, whose resolution of the calculations, especially of direct kinematics, can also involve several seconds and, therefore, are impractical for real-time controls.

[0074] Furthermore, the machine and the method described above make it possible to reduce the computational load of the main control unit, thus improving the performance thereof.

[0075] In fact, further advantages linked to the method according to the present invention concern the fact that a direct kinematics based on the search for a position, rather than on the resolution of systems of equations of a higher degree, is delegated to the control system 9.

[0076] LIST OF REFERENCE NUMBERS OF THE FIGURES

[0077] 1 machine

[0078] 2 articles 3 kinematic system

[0079] 4 arm

[0080] 5 conveyor

[0081] 6 electric actuator system

[0082] 7 electric motor

[0083] 8 joint

[0084] 9 control system

[0085] 10 end effector

[0086] 11 distal end

[0087] 12 step

[0088] 13 step

[0089] 14 manipulator

[0090] 15 Delta robot

[0091] 16 first ends

[0092] 17 second ends

[0093] CP cloud of points

[0094] JS joint space

[0095] LUT look-up table

[0096] MP motion profile

[0097] N dimension of joint space

[0098] PA corresponding point JS

[0099] PC simulation system

[0100] PDF multidimensional figure q1 position q1' position q1" position q1a current position q2 position q2' position q2" position q2a current position q3 position q3' position q3" position q3a current position

[0101] V vertices

[0102] WS workspace X axis

[0103] Y axis

[0104] Z axis

Claims

1) A method for controlling a kinematic system (3), in particular a robot (15) or manipulator (14), of an automatic machine (1 ) for processing or packing articles (2); the kinematic system (3) comprising:- at least one arm (4) in order to perform at least one process upon the articles (2);- at least one respective electric actuator (6), preferably comprising an electric motor (7), mechanically connected to the at least one arm (4) at a respective joint (8);- a control system (9), electrically connected to the electric actuator (6) and configured to control the at least one arm (4) to perform a motion profile (MP) in the workspace (WS) that allows said processing to be carried out; the method comprising the steps of:- determining a mechanical configuration of the kinematic system (3);- defining, depending on the mechanical configuration of the kinematic system (3), a cloud (CP) of points (P) representing a set of possible positions in the joint space (JS) that the kinematic system (3) can reach during the performing of the motion profile (MP); wherein each of the points (P) corresponds to a precomputed possible position (q1 , q2, q3) of the kinematic system (3) in the joint space (JS); wherein each of the points (P) corresponds to at least one precomputed position (f(P)) of the kinematic system (3) in the workspace (WS); wherein the method comprises, in use, the sequential and cyclic steps of:- detecting the current position (q1 a, q2a, q3a) of at least one electric actuator (6);- defining, depending on the current position (q1 a, q2a, q3a) of the at least one electric actuator (6), a point (PA) corresponding to said current position (q1 a, q2a, q3a) by using the said cloud (CP) of points (P) as a reference;- computing, as direct kinematics, a respective effective position (f(q 1 a, q2a, q3a)) of the kinematic system (3) in the workspace (WS) as a function of the corresponding point (PA) defined by using the said cloud (CP) of points (P) as a reference;- controlling the kinematic system (3), by way of the control system (9), to perform said motion profile (MP) as a function of the said calculated effective position (f(q1 a, q2a, q3a)) of said kinematic system (3) in the workspace (WS).2) The method according to claim 1 , wherein the kinematic system (3) comprises a manipulator (14) provided with a plurality of arms (4), specifically three arms (4), connected to one another by respective joints (8) in a serial or parallel manner; wherein each arm (4) is moved by a respective electric actuator (6); wherein the cloud (CP) of points (P) has a number of dimensions equal to the number of electric actuators (6) provided for moving the kinematic system (3).3) The method according to claim 2, wherein the points (P) of said cloud (CP) of points (P) are equidistant from one another in the joint space (JS) forming a lattice that can be subdivided into multidimensional figures (PDF) bounded by a number n of vertices (V, P) corresponding to the points (P) of the cloud (CP) of points (P).4) The method according to claim 3, wherein the position of each of the electrical actuators (6) is detected and, due to said positions, the multidimensional figure (PD) is identified, in particular a cube, wherein the point (PA) corresponding to said current position (q1 a, q2a, q3a) is located.5) The method according to claim 4, wherein the point (PA) corresponding to said current position (q1 a, q2a, q3a) is identified as a function of its respective distance from the vertices (V, P) of the multidimensional figure (PDF) wherein the same is located.6) The method according to claim 4 or 5, wherein, due to the corresponding point (PA), a precompiled look-up table (LUT) is consulted, correlating positions in the joint space (JS) to respective positions in the workspace (WS).7) The method according to claim 2, wherein the points (P) of said cloud (CP) of points (P) are differently spaced from one another; wherein the definition of the point (PA) corresponding to said current position (q1 a, q2a, q3a) involves the use of space partitioning algorithms, in particular of the k-d tree type.8) The method according to any one of the preceding claims, wherein the kinematic system (3) is a Delta robot (15) comprising: three arms (4);three electric actuators (6), connected to respective first ends (16) of the arms (4) and configured to move the three arms (4);- an end effector device (10) connected to respective second ends (17), opposite to the first ends (16), of the three arms (4) and movable by the electric actuators (6) by way of the arms (4); wherein the three arms (4) are connected to the end effector device (10) and the electric actuator (6) by way of respective universal joints (8).9) An automatic machine (1 ) for processing or packing articles (2); the automatic machine (1 ) comprising a kinematic system (3) comprising in turn:- at least one arm (4) by way of which performs at least one processing upon the articles (2);- at least one respective electric actuator (6), preferably comprising an electric motor (7), mechanically connected to the at least one arm (4) at a respective universal joint (8);- a control system (9), electrically connected to the electric actuator (6) and configured to control the at least one arm (4) to perform a motion profile (MP) in the workspace that allows said processing to be carried out; the control system (9) being configured to carry out the method according to any one of the preceding claims.10) The machine (1 ) according to claim 9, wherein kinematic system (3) is a Delta robot (15) comprising:- three arms (4);- three electric actuators (6), connected to respective first ends (16) of the arms (4) and configured to move the three arms (4);- an end effector device (10) connected to respective second ends (17), opposite to the first ends (16), of the three arms (4) and movable by the electric actuators (6) by way of the arms (4); wherein the three arms (4) are connected to the end effector device (10) and the respective electric actuator (6) by way of universal joints (8).

Images