Method and apparatus for split control of an end effector for performing a task by means of a control device

Abstract

Method for split control of an end effector (7) for performing a task using a control device (1), wherein the task is divided into at least two successive sub-steps (8, 9) and the end effector (7) is controlled by the control device (1) taking into account restrictions and user commands (4) specified for the respective sub-steps (8, 9), characterized in that the control device (1) performs the following when switching from the respective sub-step (1) to the respective next sub-step (9): - Check whether, during the transition from the current sub-step (8) to the next sub-step (9), the end effector (7) satisfies all the restrictions required for the transition to the next sub-step (9), and - Blocking the transition if not all restrictions required for the transition to the next sub-step are met, or - Move from the respective sub-step (8) to the respective next sub-step (9) when the required restrictions are met.

Description

[0001] The present invention relates to a method and a device for the split control of an end effector for performing a task by means of a control device according to the preambles of claims 1 and 11.

[0002] Shared control is a concept in robotics where the control of a system is divided between a user and an assistance system. For example, the user decides which task should be performed, while the user and the assistance system jointly control the system's state variables to successfully complete the task.

[0003] A typical example of such a system is a robotic arm with an end effector that enables people with motor impairments to perform tasks independently. The user specifies a general direction of movement for the end effector, and the assistance system supports the precise positioning and alignment of the end effector. In this way, the end effector can be aligned exactly for task execution.

[0004] Currently, however, shared control is mostly implemented in an open-loop control system. This poses a problem because the system cannot react to errors, inappropriate user input, or unexpected changes in the environment.

[0005] Furthermore, in a shared control system, a task is typically divided into a sequence of sub-steps that are executed sequentially. This means that in each sub-step, a support action assigned to that sub-step is performed, and only after the completion of this action can the next sub-step commence. In the case of the aforementioned robot arm, this means that in each sub-step, the end effector not only performs its assigned support action but must also be moved into the correct position and orientation, depending on the specific workspace conditions.

[0006] Currently, however, the movement speed of the end effector is often not taken into account in split control. This means that it cannot be guaranteed that the end effector is correctly positioned and aligned after completing an action assigned to a sub-step and transitioning to the next sub-step. This can lead to inconsistent support actions when the same task is executed repeatedly due to varying user input speeds, potentially resulting in reduced performance or even task failure.

[0007] The object of the invention is to enable a task to be performed as flawlessly as possible by a system with split control.

[0008] The problem is solved by the features of the independent claims. Advantageous embodiments are specified in the dependent claims. It should be noted that the features specified in the claims can be combined individually with each other and / or with the subject matter of the general description, thereby revealing further embodiments of the invention.

[0009] This involves a method for split-controlling an end effector to perform a task using a control device, wherein the task is divided into at least two successive sub-steps and the end effector is controlled by the control device taking into account restrictions and user commands specified for the respective sub-steps, and wherein the control device performs the following when switching from the respective sub-step to the respective next sub-step: - Check whether, during the transition from the current sub-step to the next sub-step, the end effector satisfies all the restrictions required for the transition to the next sub-step, and - Blocking the transition if not all restrictions required for the transition to the next sub-step are met, or - Switching from the respective sub-step to the respective next sub-step when the required restrictions are met.

[0010] The method according to the invention is particularly applicable in shared control, a control concept in which the control of a system - here an end effector - is divided between an automated control device and a human user.

[0011] In robotics, an end effector is the final link in a kinematic chain that interacts directly with the environment and performs tasks. A typical example of an end effector is the gripper of a robot arm, which can grasp and release objects, such as a drawer handle. A possible task would therefore be opening a drawer with the robot arm's gripper.

[0012] Unlike conventional control systems, such as manual or fully automatic control, shared control involves both the user and the control unit. This means the user can influence the end effector's control in real time, while the end effector autonomously performs certain tasks. The user can input commands before and during task execution via a user interface, such as a joystick. The control unit processes these commands by blending them with assistance guidelines and / or mapping them to the end effector's movements. The processed result is then used to control the end effector accordingly.

[0013] It is possible to divide the task to be performed into at least two consecutive sub-steps. For example, the task of opening a drawer can be subdivided into three sub-steps: moving the gripper to the position of the drawer handle with a predefined orientation, gripping the drawer handle, and opening the drawer. These sub-steps are executed sequentially by the end effector, meaning that the next sub-step may only begin after the previous sub-step has been completed. In particular, the user can input user commands into the control device at each sub-step to influence the end effector's control for executing that specific sub-step.

[0014] It can also be stipulated that certain restrictions must be met in each sub-step. These restrictions define specific parameter limits for the end effector, which it must not exceed during execution in order to perform its function correctly and safely. For example, the end effector may have to maintain a certain height in one sub-step, while a specific orientation is required in a subsequent sub-step. In this example, the specified height and orientation thus represent the respective restrictions for the corresponding sub-steps.

[0015] In particular, it can be provided that the end effector is controlled by the control unit, taking into account the restrictions and user commands specified for the respective sub-steps. The control unit can combine the user commands entered in real time with its own pre-programmed instructions for each sub-step, while adhering to the specified restrictions.

[0016] As mentioned earlier, shared control currently typically uses open-loop control, which lacks continuous feedback and correction. This can lead to the end effector being unable to respond to constraint violations due to unexpected changes in the work environment and / or inappropriate user input. Such a violation can, in turn, cause the task to fail.

[0017] To avoid this, it can be implemented that, during the transition from one sub-step to the next, a check is performed to ensure that the end effector meets all the restrictions required for the change. This check can be carried out by verifying the end effector's pose, i.e., its position and orientation. If the end effector has been moved into the correct position and orientation during the transition from the current sub-step, the required restrictions are considered fulfilled.

[0018] If it is determined that all required restrictions are met, the transition to the next sub-step is initiated. In this case, the execution of the current sub-step ends, it is placed in the deactivation state, and the execution of the next sub-step begins, which then enters the activation state.

[0019] If, however, the end effector is unable to meet all the necessary restrictions during the transition to the next substep, the transition is blocked. Blocking the transition means that the execution of the next substep cannot begin and it remains in a disabled state until the required restrictions in the current substep are met. However, blocking the transition does not mean that user command input is interrupted; on the contrary, the user can continue to input commands to the control device during this blocking phase.

[0020] Furthermore, it can be provided that the current sub-step remains in the activation state during the blocking phase, so that the end effector continues to be controlled until it fulfills all the restrictions required for the current sub-step and the transition to the next sub-step can take place.

[0021] The method according to the invention ensures compliance with the necessary restrictions in each step, thereby enabling the end effector to perform the task safely, precisely, and efficiently. Within the framework of shared control, the end effector is controlled by both the user and the control unit – even in an open control loop without continuous feedback or correction of the actual task execution. Even in the event of an error, the input of user commands does not need to be interrupted, which is particularly advantageous when the end effector is used as part of an assistive device for people with motor impairments, such as a robotic arm or a motorized wheelchair with an integrated robotic arm.

[0022] It is preferred that the end effector can still be controlled to a limited extent during the blocking phase, restricting only those user commands that would affect the transition from the current substep to the next. This has the advantage of avoiding a complete interruption of movement during the blocking phase by continuing to allow limited control of the end effector based on user input. This implementation variant ensures that the end effector can respond flexibly to user input at critical moments without losing efficiency due to waiting times. This ensures that the end effector is controlled only within the workspace of the current substep until the necessary restrictions are met.

[0023] It is preferred that during this testing phase the fulfillment of the restrictions is checked by verifying the pose of the end effector, whereby the required restrictions are deemed to be fulfilled if the end effector has been brought into the correct position and orientation during the transition from the current sub-step to the next sub-step.

[0024] It is preferred that the following be carried out during this examination phase: a) Capturing a target pose of the end effector provided by the control device based on user commands, b) Capturing intermediate poses of the end effector provided by the control device by tracking the target pose, with the end effector being guided along the motion trajectory formed by these intermediate poses, c) Comparing the target pose with the last intermediate pose provided, and d) Determine that the required restrictions are considered fulfilled if the difference between the target pose and the intermediate pose is less than a specified difference.

[0025] In robotics, pose refers to the position and orientation of an object—in this case, the end effector—in space. The pose of the end effector encompasses both its position (e.g., x, y, and z coordinates in three-dimensional space) and its orientation (e.g., rotations around the x, y, and z axes).

[0026] According to a), the end effector's target pose is provided by the control device based on user commands. The user can use a user interface such as a joystick for this purpose. Joysticks with two degrees of freedom allow movement in two directions (e.g., forward / backward and left / right), while joysticks with three degrees of freedom additionally allow movement along a third axis (e.g., up / down). These movements are entered as user commands into the control device, which can then provide the target pose with six degrees of freedom—three for translation (movement along the x, y, and z axes) and three for rotation (rotation around the x, y, and z axes). These six degrees of freedom allow the end effector to assume any desired position and orientation in three-dimensional space.

[0027] According to b), the intermediate pose is provided by the control device based on the target pose. In particular, the intermediate pose can be generated by a dynamic system integrated into the control device, such as an interpolator, by this system following the target pose and generating intermediate poses in the process. These intermediate poses lie on the path to the target pose and allow the end effector to move incrementally toward the target pose. The end effector then follows the trajectory of motion formed by the intermediate poses instead of moving directly to the target pose. According to c), the target pose is compared with the intermediate pose. Here, the target pose is specifically compared with the most recently provided intermediate pose.

[0028] According to d), the required restrictions are considered fulfilled if the difference between the target pose and the intermediate pose is less than a predetermined difference. This difference indicates how far the intermediate pose is from the target pose, both in terms of position and orientation. Preferably, this difference is calculated in the special Euclidean group SE (3), taking into account both the linear distance (between the positions) and the angular distance (between the orientations). The required restrictions are considered fulfilled only if this difference is less than the predetermined difference.

[0029] It is preferred that during this blocking phase the end effector continues to be controlled in a restricted manner by guiding the end effector to a newly created target pose based on the user commands entered during the blocking phase, provided that these user commands would not affect the transition from the current sub-step to the next sub-step.

[0030] It is preferred that during this blocking phase the end effector continues to be controlled in a constrained manner by traversing the last provided intermediate pose along the motion trajectory in the work area of ​​the current sub-step and, starting from this traversed intermediate pose, providing further intermediate poses while tracking the target pose, so that the end effector is guided along the new motion trajectory formed by these new intermediate poses.

[0031] It is preferred that the control device includes a shared control template and that the task is modeled as a state machine in the shared control template. It is particularly preferred that the target pose is provided by the shared control template.

[0032] A Shared Control Template (FCT) is a framework that can model the sub-steps of a task using a Functional Structure Model (FSM). Within the FSM, input mapping (IM) and active constraints (ACs) can be defined for each sub-step. The input mapping maps the low-dimensional user command inputs (e.g., via a three-degree-of-freedom joystick) to the task-specific displacements of the high-dimensional target pose (e.g., with six degrees of freedom) of the end effector. The active constraints further limit the end effector's target pose to support successful task execution. Specifically, the Shared Control Template can provide the target pose as output, which depends on the user commands and takes into account the constraints in the various sub-steps of the task to be executed.

[0033] It is preferred that the control device includes an interpolator, and that the intermediate positions are provided by the interpolator.

[0034] An interpolator is a dynamic tool that uses interpolation algorithms to calculate and output intermediate poses, creating a smooth path between two poses for the end effector to follow. The target pose output by the shared control template can be input into the interpolator, allowing it to continuously follow the target pose and incrementally generate intermediate poses that define the path to the target pose.

[0035] It is preferred that the restrictions are specified in the form of fixed values, function values ​​and / or manifolds.

[0036] With fixed values, it can be stipulated that a degree of freedom (DoF) of the end effector is fixed to a specific value, such as a fixed height, so that the end effector can only move at that height. With function values, on the other hand, various functions such as inequalities, polynomials, dot products, or user-defined functions can be used. These functions use inputs such as distances or transformations to dynamically control the movement of the end effector. For example, a function can adjust the angle of the end effector based on its distance to a target point. Using manifolds, a specific range of motion in high-dimensional space can be defined, restricting the end effector to a flexible but bounded area.

[0037] The invention further comprises a robot system for the split control of an end effector of a robot for performing a task, wherein the task is divided into at least two successive sub-steps, comprising at least one control device, wherein the control device is configured such that the end effector is controlled by the control device taking into account restrictions and user commands specified for the respective sub-steps, and wherein the control device is configured to perform the following when switching from the respective sub-step to the respective next sub-step: - Check whether, during the transition from the current sub-step to the next sub-step, the end effector fulfills all spatial restrictions required for the transition to the next sub-step, - Blocking the transition if not all spatial restrictions required for the transition to the next sub-step are met, and - Transition from the respective sub-step to the respective next sub-step when the required spatial restrictions are met.

[0038] It is preferred that the control device be designed in such a way that during this blocking phase the end effector can still be controlled to a limited extent by the control device, restricting only those user commands that would affect the transition from the current sub-step to the next sub-step.

[0039] It is preferred that the control device includes a shared control template trained to model the task in the form of a state machine and to provide a target pose based on the user commands.

[0040] It is preferred that the control device includes an interpolator designed to provide intermediate poses by tracking the target pose.

[0041] It is preferred that the control device is designed such that the end effector is controlled taking into account restrictions in the form of fixed values, function values ​​and / or manifolds.

[0042] The solution presented here and its technical context are explained in more detail below with reference to the figures. It should be noted that the invention is not intended to be limited by the illustrated embodiments. In particular, unless explicitly stated otherwise, it is also possible to extract partial aspects of the situations explained in the figures and combine them with other components and / or findings from other figures and / or the present description. The figures show schematically and by way of example: Fig. 1 a method according to the invention for the split control of an end effector, Fig. 2 a known method for split control of an end effector, Fig. 3 a problem scenario in the procedure according to Fig. 2, Fig. 4 Comparison of problem scenarios in the procedure according to Fig. 2 at different user command input speeds, Fig. 5 a variant for limited control during the blocking phase and Fig. 6 another implementation variant for limited control during the blocking phase.

[0043] Fig. Figure 1 shows a method according to the invention for the split control of an end effector, while Fig. 2 represents a known method for the split control of an end effector.

[0044] In Fig. 1 and Fig. 2. It is evident that the method according to the invention, in comparison to the known method, additionally comprises process steps a), b), c), and d). In order to illustrate the differences between the method according to the invention and the known method, as well as the resulting positive technical effect, more clearly, Fig. 1 and Fig. 2 are compared here. First, based on... Fig. 2. The known procedure is described in order to illustrate the problem that arises. Subsequently, based on... Fig. 1 The method according to the invention is explained in order to illustrate how the process steps a), b), c) and d) can contribute to solving the problem.

[0045] Fig. Figure 2 shows an end effector 7 that is shared-controlled by a user (not shown) and a control device 1. The control device 1 comprises a shared control template 2 and an interpolator 3. The shared control template 2 provides a target pose 5 of the end effector 7 based on user commands 4, while the interpolator 3 generates intermediate poses 6 of the end effector 7 by following the target pose 5 provided by the shared control template 2, so that the end effector 7 is controlled based on the intermediate poses 6.

[0046] Fig. Figure 2 also shows that a task to be performed with the end effector 7 is divided into sub-step _A 8 and sub-step _B 9, which are to be executed sequentially. In both sub-step _A 8 and sub-step _B 9, the end effector 7 is to be moved, firstly in compliance with the restrictions assigned to the respective sub-step 8, 9 (e.g., maintaining a predefined orientation and / or height of the end effector), and secondly, controlled to execute an action assigned to the respective sub-step 8, 9. The task with sub-steps _A 8 and _B 9 is modeled by a state machine 11 in the shared control template 2.

[0047] Fig. Figure 2 further shows that the target pose 5 provided by the Shared Control Template 2 is fed back into the template (see dashed line), so that target pose 5 is provided cumulatively based on the last target pose 51 and the user commands 4.

[0048] As mentioned previously, the intermediate poses are provided based on the target pose, and the end effector is controlled based on these intermediate poses. Therefore, during step change 10, the last provided intermediate pose 6 should be as close as possible to the target pose 5.

[0049] However, with the known procedure according Fig. The following two problems can occur: The user typically enters the user commands 4 into the Shared Control Template 2 via a user interface, such as a joystick (not shown). If the user enters the commands 4 too quickly, the target pose 5 is also deployed correspondingly quickly. This can lead to a situation where, during a partial step transition 10, the difference between the last deployed intermediate pose 6 and the target pose 5 is too large. It can happen that the intermediate pose cannot follow the target pose, which, especially during partial step transitions, results in the required restrictions not yet being met.

[0050] In the inventive method according to Fig. 1. The aforementioned problems can be avoided.

[0051] Fig. 1 shows that in addition to the steps of the known procedure according to Fig. 2. The process steps a), b), c), and d) are executed during substep change 10 from substep _A 8 to substep _B 9. The illustrated steps a), b), c), and d) with blocks 110, 120, 130, and 140 serve as examples. In block 110, the target pose 5 of the end effector 7 is detected at this point in time. In block 120, an intermediate pose 6 of the end effector 7 is also detected at this point in time. In block 130, the detected target pose 5 is compared with the detected intermediate pose 6. In block 140, substep change 10 only occurs if the detected intermediate pose reaches the detected target pose.

[0052] The process steps a) to d) of the method according to the invention ensure that the end effector 7 is moved into the correct position and orientation at the time of the sub-step change 10, in accordance with the spatial restrictions of the respective sub-steps 8 and 9, without blocking user commands. Furthermore, it is ensured that the entire task can be accomplished without any loss of performance, even if the control loop is an open-loop system.

[0053] Fig. Figure 3 shows a typical problem scenario that occurs in Fig. The two procedures described above can occur. Fig. The crosses 12, 13, 16, and 17 represent the positions of the end effector, and the arrows 14, 15, 18, and 19 represent the orientations of the end effector. The end effector should be guided according to the initial movement direction 22 entered by the user.

[0054] In Fig. 3. It can be seen that a translation shift takes place between the target position _t 12 of the end effector in sub-step _A 8 and the target position _t+1 13 of the end effector in sub-step _B 9. According to the procedure according to Fig. 2. The target position _t 12 is provided based on the user commands from, for example, the Shared Control Template at time t, while the target position_t+1 13 is provided based on the user commands from, for example, the Shared Control Template at time t+1.

[0055] In Fig. It can also be seen that both the target orientation _t 14 in sub-step _A 8 and the target orientation _t+1 15 in sub-step _B 9 remain horizontal. This could mean that the restrictions for sub-step _A 8 are such that the orientation of the end effector must be set horizontally before sub-step _B 9 is executed.

[0056] In Fig. 3 In particular, it can be seen that between the intermediate position _t 16 of the end effector in sub-step _A 8 and the intermediate position_t+1 17 of the end effector in sub-step _B 9 a translation shift also takes place, while between the intermediate orientation _t 18 of the end effector in sub-step _A 8 and the intermediate orientation_t+1 19 of the end effector in sub-step _B 9 only a slight rotation takes place.

[0057] According to the procedure Fig. 2. The intermediate position _t 16 and the intermediate orientation _t 18 (i.e., the intermediate pose at time t) are provided by the interpolator at time t based on the target position _t 12 and the target orientation _t 14 (i.e., the target pose at time t), and the end effector is thus controlled based on the intermediate position _t 16 and the intermediate orientation _t 18 at time t.

[0058] Accordingly, the intermediate position _t+1 17 and the intermediate orientation _t+1 19 based on the target position _t+1 13 and the target orientation _t+1 15 are also provided by the interpolator at time t+1, so that the end effector is controlled based on the intermediate position _t+1 17 and the intermediate orientation _t+1 19 at time t+1.

[0059] As mentioned above, the target position _t 12 and the target position _t+1 13 are provided based on the user commands at time t and time t+1, respectively. This can lead to the problem scenario that in sub-step _B 9, the intermediate orientation _t+1 19 is not yet set horizontally like the target orientation _t+1 15 if the user enters the user commands too quickly. Consequently, the restrictions set for sub-step _A 8 are violated. Such restriction violations can lead to a degradation in performance or the failure of the entire task.

[0060] Fig. Figure 4 schematically and exemplarily shows a comparison of problem scenarios in the procedure according to Fig. 2 at different user command input speeds. This shows Fig. 4 links the case of a high user command input speed 20, while Fig. Figure 4 on the right shows the case of a relatively low user command input speed 21.

[0061] In Fig. 4 are the restrictions for substep _A 8 as in Fig. 3 is specified. This means that the intermediate orientation (dashed arrows) must be set horizontally before activating sub-step _B 9.

[0062] Compared with Fig. 3 shows Fig. 4 more clearly shows how the speed of user command input affects compliance with these restrictions.

[0063] In the case of high user command input speed 20, the user enters the user commands into the shared control template faster than in the case of low user command input speed 21. Therefore, in case 20, the end effector has less time to adjust its horizontal orientation than in case 21 before substep _B 9 is activated. That is, the faster the user commands are entered, the greater the deviation from compliance with the spatial restrictions. Thus, in Fig. 4 also to recognize that in sub-step _B 9 the deviation between the horizontally set target orientation (solid arrows) and the intermediate orientation not yet set horizontally (dashed arrows) is greater in case 21 than in case 21.

[0064] Fig. Figure 4 thus illustrates the cause of the non-compliance with the restrictions, namely the excessive speed of user command input by the user.

[0065] Fig. Figure 5 showed a variant design with which the in Fig. 3 or Fig. The non-compliance with restrictions shown in section 4 can be avoided.

[0066] Fig. Figure 5 initially shows the same problem scenario as in Fig. 3 or Fig. 4, i.e., at the time of the sub-step change 10 from sub-step _A 8 to sub-step _B 9, the intermediate orientation (dashed arrows) has not yet reached the target orientation (solid arrows). This means that the required restrictions at sub-step change 10 are not yet fulfilled.

[0067] To ensure that the necessary restrictions are met during substep change 10, it is checked whether the end effector meets all the restrictions required for the transition to the next substep, i.e., in the case of Fig. 5. Whether the intermediate orientation (dashed arrows) has reached the target orientation (solid arrows). If this is not the case, the partial step change 10 is blocked.

[0068] Fig. Figure 5 further shows a preferred embodiment in which, during the blocking phase, the end effector is controlled in a restricted manner until the required restrictions are met.

[0069] In Fig. Figure 5 shows that during the blocking phase, the currently captured intermediate pose in sub-step _B 9 (dashed arrows with corresponding cross in sub-step _B) is returned along the end effector's movement trajectory within the workspace of sub-step _A 8 (double dashed arrows with corresponding cross in sub-step _A). This allows the interpolator to generate further intermediate poses within the workspace of sub-step _A 8, starting from this returned intermediate pose and following the user-initiated initial movement direction 22, until the last newly generated intermediate pose reaches the initial target pose.

[0070] Fig. Figure 6 shows another preferred execution variant in which the end effector does not follow a traversed intermediate pose during the blocking phase, but is controlled based on a newly generated target pose candidate until the required restrictions are met. This so-called target pose candidate is generated based on the user commands entered during the blocking phase, e.g., by the shared control template. If additional user commands are entered during the blocking phase, they are not blocked. Instead, based on these commands, at least one further target pose is provided as a target pose candidate, also by the shared control template. Subsequently, it is checked whether this target pose candidate immediately triggers the substep change 10.

[0071] In Fig.Figure 6 shows that, in addition to the initial movement direction 22 (solid arrows), a total of seven further movement directions are entered by the user in the form of user commands during the blocking phase: five dashed arrows and two double-dashed arrows. The movement directions represented by the five dashed arrows could directly lead to sub-step change 10 and therefore must not be used. In contrast, the two double-dashed arrows indicate movement directions that allow a return to the workspace of sub-step A. As a result, sub-step change 10 does not occur, so the effector can be controlled along these two movement directions until the necessary restrictions are met. Reference symbol list 1 Control unit 2 Shared Control Template 3 Interpolator 4 User commands 5 Target pose _t 51 Target pose _t-1 6 Intermediate pose 7 End effector 8 Sub-step _ A 9 Sub-step _ B 10 partial step changes 11 State machine 12 Target position _t 13 Target position _t+1 14 Goal orientation _t 15 Goal orientation _t+1 16 Intermediate position _t 17 Intermediate position _t+1 18 Intermediate orientation _t 19 Intermediate orientation _t+1 20 high user command input speed 21 Low user command input speed 22 initial direction of movement 110 Procedure step a) 120 Procedure step b) 130 Procedure step c) 140 Procedure step d)

Claims

Method for the split control of an end effector (7) for performing a task using a control device (1), wherein the task is divided into at least two successive sub-steps (8, 9) and the end effector (7) is controlled by the control device (1) taking into account restrictions and user commands (4) specified for the respective sub-steps (8, 9), characterized in that the control device (1) performs the following when switching from the respective sub-step (1) to the respective next sub-step (9): - Checking whether, during the transition from the current sub-step (8) to the next sub-step (9), the end effector (7) fulfills all restrictions required for the transition to the next sub-step (9), and - Blocking the transition if not all restrictions required for the transition to the next sub-step are fulfilled, or - Switching from the respective sub-step (8) to the respective next sub-step (9).if the necessary restrictions are met. Method according to claim 1, wherein during this blocking phase the end effector (7) can still be controlled to a limited extent, whereby only those user commands (4) are restricted which would affect the transition from the current sub-step (8) to the next sub-step (9). Method according to claim 1 or 2, wherein during this testing phase the fulfillment of the restrictions is checked by testing the pose (5, 6) of the end effector (7), wherein the required restrictions are deemed to be fulfilled if the end effector has been brought into the correct position and orientation during the transition from the current sub-step (8) to the next sub-step (9). The method of claim 3, wherein during this testing phase the following is performed: - detecting a target pose (5) of the end effector provided by the control device (1) based on user commands (4), - detecting intermediate poses (6) of the end effector provided by the control device (1) by following the target pose (5), the end effector (7) being guided along the motion trajectory formed by these intermediate poses, - comparing the target pose with the last intermediate pose provided, and - determining that the required restrictions are considered to be satisfied if the difference between the target pose and the intermediate pose is less than a predetermined difference. Method according to claim 4, wherein during this blocking phase the end effector (7) continues to be controlled in a limited manner by guiding the end effector (7) to a newly generated target pose based on the user commands (4) entered during the blocking phase, provided that these user commands would not affect the transition from the current sub-step to the next sub-step. Method according to claim 4 or 5, wherein during this blocking phase the end effector (7) continues to be controlled in a restricted manner by traversing the last provided intermediate pose along the motion trajectory in the working area of ​​the current sub-step and, starting from this traversed intermediate pose, providing further intermediate poses while following the target pose, so that the end effector (7) is guided along the new motion trajectory formed by these new intermediate poses. Method according to one of the preceding claims, wherein the control device (1) comprises a shared control template (2) and the task is modeled in the form of a state machine (11) in the shared control template. Method according to any one of claims 3 to 7, wherein the target pose (5) was provided by the shared control template (2). Method according to claim 8, wherein the control device (1) comprises an interpolator (3), and the intermediate poses (6) were provided by the interpolator. Method according to one of the preceding claims, wherein the restrictions are specified in the form of fixed values, function values ​​and / or manifolds. A robot system for the split control of an end effector (7) of a robot for performing a task, wherein the task is divided into at least two successive sub-steps (8, 9), comprising: at least one control device (1), wherein the control device is configured such that the end effector is controlled by the control device taking into account restrictions and user commands specified for the respective sub-steps, characterized in that the control device (1) is configured to perform the following when transitioning from the respective sub-step to the respective next sub-step: - Checking whether, during the transition from the current sub-step to the next sub-step, the end effector fulfills all spatial restrictions required for the transition to the next sub-step, - Blocking the transition if not all spatial restrictions required for the transition to the next sub-step are fulfilled.and- transition from the respective sub-step to the respective next sub-step, when the required spatial restrictions are met. Robot system according to claim 11, wherein the control device (1) is designed such that during this blocking phase the end effector (7) can still be controlled to a limited extent by the control device, whereby only such user commands (4) are restricted which would affect the transition from the current sub-step to the next sub-step. Robot system according to claim 11 or 12, wherein the control device (1) comprises a shared control template (2) configured to model the task in the form of a state machine and to provide a target pose (5) based on the user commands. Robot system according to claim 13, wherein the control device (1) comprises an interpolator (3) configured to provide intermediate poses (6) by tracking the target pose. Robot system according to one of the preceding claims 11 to 14, wherein the control device (1) is configured such that the end effector (7) is controlled taking into account restrictions in the form of fixed values, function values ​​and / or manifolds.

Images