Planning technical processes with hierarchically structured tasks and parallelization options

Abstract

The invention describes a computer-implemented method for planning a technical process (P), wherein the technical process (P) has a hierarchically structured task (A), wherein the technical process (P) comprises a modification of a physical unit. The computer-implemented method has the following steps: - reading (S1) a dependency specification (AA) of the hierarchically structured task (A), - deriving (S2) parallelization options (PO) from the dependency specification (AA), and - generating (S3) a first action sequence (AS1) with incorporation of the parallelization options (PO), wherein the first action sequence (AS1) implements the hierarchically structured task (A). Furthermore, the invention describes a method for executing a technical process, a computer program product and a computer-readable medium.

Description

Technical Field

[0001] This invention relates to a computer-implemented method for planning a technical process, wherein the technical process has hierarchically structured tasks, and wherein the technical process (P) includes modifications to physical units. Furthermore, this invention describes a method for performing a technical process, a computer program product, and a computer-readable medium.

[0002] This invention can be used in particular for planning and executing technical processes with hierarchical structured tasks and parallelization options. Background Technology

[0003] The objective of this invention is to provide a solution for planning a technical process having hierarchical tasks with parallelization options.

[0004] In many fields, it is possible to solve tasks that are currently impossible to automate economically due to engineering costs, by leveraging autonomous systems. This means that flexible or dynamic fields require creating appropriate processing plans for constantly emerging new variations and then fully executing those plans. For frequently repeated fixed variations, especially in large-scale production, the cost of manual design may be worthwhile, but the required flexibility makes this strategy uneconomical.

[0005] The main focus is on the automation of tasks, which is primarily based on the parallel execution of processes.

[0006] Preferably, the following areas are mentioned for this purpose:

[0007] - Logistics and internal logistics: Multiple components can move goods within the factory or between production sites, where further production steps can be performed on those goods. Here, autonomy can be achieved through centralized or decentralized coordination of individual moving vehicles (AGVs). Furthermore, this can also be viewed as a sub-problem of coordinating the entire factory, thereby enabling autonomous control of the entire manufacturing process. Therefore, new production tasks can be easily input, the number of AGVs can be changed, or responses to constantly changing KPIs can be made.

[0008] - Road traffic: Multiple vehicles and some fixed infrastructure have a range of sensors, and some vehicles can drive autonomously or partially autonomously. The following question is particularly raised here: how to combine which sensors to generate sufficiently accurate data for motion planning or infrastructure control through sensor fusion.

[0009] -Personal Transportation: How should one select which combinations of public transportation components to dynamically accomplish the required transportation tasks? What is the optimal timetable? How to deal with congestion and delays, such as in rail transit?

[0010] - Process Industries: Multiple continuous production processes require coordinated control of pumps and machines used for mixing, heating, or cooling, as well as cleaning steps. Due to the complexity of the processes within each component, continuous adaptation of the processes can only be achieved after detailed simulations and testing. This raises the questions of which simulations and tests provide the necessary information and which controls minimize KPIs. Since both are time-consuming and costly, these steps must be coordinated and parallelized as much as possible.

[0011] -Perception: There are a large number of algorithms in image processing that are computationally intensive and only provide usable results in task-related links.

[0012] Therefore, all the advantages of modern computer architecture are utilized here to achieve as much parallelization as possible.

[0013] What all these application areas have in common is the need to coordinate numerous processes and control these processes in a strongly parallel manner. Therefore, autonomy in these areas is only possible if parallelism is considered during the creation and / or planning phase and utilized to the maximum extent during execution.

[0014] Therefore, one of the technical problems in the prior art is to automatically select elements suitable for solving a given problem in parallel from a large number of possible elements, to structure these elements in a network of multiple parallel sequences, and to adjust them to suit the task using appropriate parameters.

[0015] It is known that two approaches can be used to solve this special case. As mentioned above, the challenge lies particularly in the combination of problem size, the necessary parallelism in execution, and the flexibility in task formulation.

[0016] In the first approach, if only limited flexibility is required and the range of values ​​needed for these flexible components is fully known during the design phase, the process can be created manually in a classic manner. To do this, the process's dependencies on the flexible components are manually coded, and which processes run in parallel during execution are explicitly developed during the design process. This manual design process can, for example, include state machines, behavior trees, or HTN components.

[0017] In the second approach, a specialized planner can be used to solve very small tasks in a specific homogeneous domain (i.e., ideally purely symbolic tasks (also known as scheduling) or purely sequential tasks (also known as path planning)). The planner's results can incorporate parallelism (particularly in the form of Petri nets), which can then be utilized using corresponding execution components.

[0018] Therefore, the objective of this invention is to describe an improved solution for planning interdependent technical tasks and processes. Summary of the Invention

[0019] This invention derives from the features of the independent claims. Advantageous extensions and designs are the subject of the dependent claims. The design, applicability, and advantages of this invention are apparent from the following description and drawings.

[0020] This invention relates to a computer-implemented method for planning a technical process, wherein the technical process has hierarchically structured tasks, and wherein the technical process (P) includes modifications to physical units. The computer-implemented method comprises the following steps:

[0021] - Read the dependency description of the hierarchical structured task.

[0022] - Derive parallelization options from the dependency description, and

[0023] - Generate a first action sequence with the parallelization option included, wherein the first action sequence performs the hierarchical structured task.

[0024] Within the scope of this invention, hierarchical structured tasks are tasks structured at different hierarchical levels. Higher hierarchical levels may have tasks that are specified in more detail by further tasks at the hierarchical levels below them.

[0025] Specifically, each hierarchical structured task has a hierarchical description, wherein each hierarchical description defines a hierarchical level within the process, such that a first task at a first hierarchical level is further refined by a further task (or at least one second task) at a second hierarchical level, wherein the first task and the at least one second task / the further task are part of the hierarchical structured task. Specifically, the first hierarchical level may be determined by the first task "planning, execution, control, and response". At the second hierarchical level, the first task "control" at the second hierarchical level determines (at least one second task or) the further task "measuring actual values, recalling target values, comparing actual values ​​and target values, and calculating deviations".

[0026] In the context of this invention, modification of physical units includes modifications to physical units, particularly real-world objects, specifically alterations, adaptations, manufacturing processes, production steps, treatments, and / or modifications. This particularly includes milling of production parts, assembly of individual components, surface treatment of workpieces, shaping of workpieces, and / or other modifications to physical units performed within the scope of the manufacturing process.

[0027] In the context of this invention, the description of task dependencies is a description of the dependencies between the aforementioned hierarchical tasks. These dependencies may in particular be causal relationships and / or temporal relationships.

[0028] Causal dependency describes the causal dependence of a technical process, particularly the dependency arising from actions that can only be performed after previous (directly or indirectly preceding) actions are completed and their results exist, especially painting after molding. In particular, an action can also depend on a large number of preceding (parallel or consecutive) actions. Temporal information describes the temporal dependency of a technical process, particularly processing time, waiting time, delivery time, working time, and idle time.

[0029] In the context of this invention, the parallelization options for tasks are particularly for options that allow the execution of hierarchically structured tasks in parallel. Thus, options for parallelizing tasks at the same hierarchical level can be specifically provided.

[0030] The first sequence of actions within the scope of this invention is a sequence of actions that perform the hierarchical structured task. This sequence may contain continuous and / or parallel actions. The hierarchical task may define the cleaning of a machine, with associated actions being "sandblasting," "brushing," "washing," and "drying."

[0031] One aspect of the present invention is to plan a technical process with hierarchically structured tasks, and in doing so, to consider the parallelization options of the hierarchically structured tasks.

[0032] When reading the dependency description of the hierarchical structured task, the hierarchical planner creates a series of actions, also known as sub-plans, with progressively increasing levels of detail based on a hierarchical model of declarative and procedural knowledge about all relevant elements and processes in the technical process. For this purpose, appropriate action sequences for the technical process are transferred to specialized planning components.

[0033] Then, each step in the sub-plan obtained in this way is developed into a new hierarchical task at a more detailed level, until all processes contained in the plan are sufficiently detailed for the task.

[0034] Unlike sequential planning, hierarchical planners must provide sub-planners not only with relevant instances of the model, but also with temporal information about those instances. That is, when are specific instances created, or which components are occupied during which time periods? This is mandatory for considering the effects of parallelization during planning.

[0035] When deriving parallelization options from the dependency specification, the parallelization component obtains a plan that includes or implements parallelism from the hierarchical planner and implements parallel execution on the hardware. This corresponds to the plan launching multiple processes simultaneously.

[0036] After generating the first sequence of actions with the parallelization option included, the hierarchical planner (also known as the planning component) provides the first sequence of actions to the parallelization component, which is a set of partially ordered actions—ordering that initially indicates only causal dependencies and not explicitly temporal dependencies. Each pair of actions in this plan is a candidate for parallel execution, and the resource requirements between these two actions are described neither explicitly (i.e., in particular, a direct description of the order between actions) nor implicitly (i.e., in particular, a chain of other ordering relationships).

[0037] A lock is a time constraint between actions in a plan, corresponding to a causal sequence in the plan. This constraint requires the results of two distinct preceding actions as input variables; that is, there must be two results occurring at the same time, which allows the subsequent action to be initiated. The parallelization component recognizes such locks and uses them to determine a series of actions that can be executed in parallel.

[0038] Once all the input variables for a series are available, the series can be started independently of other running processes.

[0039] An extension of the invention includes a method for performing a technical process, wherein the technical process has been planned using the method according to the invention. The method for performing the technical process includes further method steps of performing a hierarchical structured task of the technical process by executing a first sequence of actions.

[0040] In order to perform the hierarchical structured task of the technical process by executing the first sequence of actions, it is equally applicable to perform it in the real world.

[0041] In another extension of the invention, the method has further method steps:

[0042] - Receive process data, wherein the process data includes the actual state of the technical process, and

[0043] - Generate a second action sequence based on the process data, wherein the second action sequence implements the hierarchical structured task. This includes performing the hierarchical structured task of the technical process by executing the second action sequence.

[0044] If there is a significant deviation from the result of the first action sequence, the parallelization component reports this back to the hierarchical planner, where a plan change, namely the second action sequence, can be created.

[0045] This parallel execution necessarily includes a monitoring component that checks whether all actions in a sequence of actions satisfy the conditions calculated during planning (also known as the target state).

[0046] In cases of significant deviation, particularly due to the result of the first action sequence falling outside the expected range, or in erroneous situations, particularly due to the first action sequence failing to terminate successfully, the monitoring component provides corresponding information to the planning component, enabling the creation of a new plan, or second action sequence in the terminology of this invention. Specifically, the planning component is informed which action, with which input and output values, caused the termination of execution. Based on the feedback, the planning component can decide whether to choose a new plan or continue planning. From there, the process restarts.

[0047] In one extension of the invention, the second action sequence replaces the first action sequence. The advantage of this is that it allows for a new planning of the technical process.

[0048] In one extension of the invention, the second action sequence follows the first action sequence in time. The advantage of this is that the technical process continues adaptively, thereby correcting deviations and errors.

[0049] In one extension of the invention, the second action sequence stops the technical process. The advantage of doing so is that no further action sequences are executed and the technical process is interrupted.

[0050] In one extension of the invention, the technical process is performed by at least one production unit, wherein the at least one production unit is configured as a production machine, production robot, and / or automated unit. The at least one production unit here performs the technical process in the real world.

[0051] In one extension of the present invention, the technical process is configured as follows:

[0052] - Manufacturing process, and / or

[0053] - Logistics process, and / or

[0054] -Traffic control process, and / or

[0055] -Transportation process.

[0056] In this context, it is considered particularly advantageous to construct the technical process as a manufacturing process.

[0057] In one extension of the invention, the dependency specification defines a hierarchical structured task with respect to the following aspects:

[0058] -Procedural order, and / or

[0059] - Causal dependence, and / or

[0060] - Interactions, and / or

[0061] -Time information, and / or

[0062] -Required resources, and / or

[0063] - Tasks that must be performed simultaneously at the same time, and / or

[0064] - Required tools.

[0065] The procedural sequence illustrates the process-dependent nature of the technical process.

[0066] Causal dependency describes the causal dependence of the technical process, particularly the dependency arising from actions that can only be performed after previous (directly or indirectly preceding) actions are completed and their results exist, especially painting after molding. In particular, an action can also depend on a large number of previous (parallel or consecutive) actions.

[0067] The interaction relationship illustrates the dependence of the technical process caused by interaction, especially when the execution components of the action must perform the action in combination with each other, particularly the gripping robot arm as the first component and the processing robot arm as the second component.

[0068] The time information illustrates the time dependence of the technical process, particularly processing time, waiting time, delivery time, working time, and idle time.

[0069] The required resources describe the resources needed for the described technical process, in particular materials, raw materials, goods, financial resources, human resources, space, and environmental conditions (especially temperature, pressure, and humidity).

[0070] The description of the technical process refers to tasks that are required to be performed simultaneously in time, particularly tasks with complementary functions, such as cleaning and water supply.

[0071] The required tools description specifies the tools needed for the described technical process, in particular the tools required for simultaneous / parallel operation, expected tool wear, tool space requirements, tool availability duration, tool application conditions, and / or application resources.

[0072] In one extension of the invention, the dependency description indicates dependencies that arise when performing a hierarchical structured task. Specifically, in this case, reference is made to the dependencies mentioned above.

[0073] In one extension of the invention, the hierarchical structured tasks each have a hierarchical description, wherein each hierarchical description defines a hierarchical level within the process, such that a first task at a first hierarchical level is further refined by a further task at a second hierarchical level (or, in the wording of the invention, by at least one second task at the second hierarchical level), wherein the first task and the at least one second task / the further task are part of the hierarchical structured task. Specifically, the first hierarchical level may be determined by the first task "planning, execution, control, and response". At the second hierarchical level, the first task "control" at the second hierarchical level determines at least one second task or the further task "measuring actual values, calling target values, comparing actual values ​​and target values, and calculating deviations".

[0074] In one extension of the invention, the method has further method steps:

[0075] -Receive user input, and

[0076] - Generate a third action sequence based on the user input.

[0077] Optionally, human interaction, particularly user input, and adaptation to action sequences are possible. This is especially meaningful for debugging and partial automation. Here, the hierarchical model enables the graphical representation of data and processes corresponding to specific user roles.

[0078] This partially automated operation allows access to the non-modeled experience of personnel familiar with the process. The hierarchical structure of tasks allows for focus on each relevant aspect in this case. Thus, interactive components could be selecting a favorite plan from a set of comparable plans, specifying additional time locks or combining series, or providing feedback to supplement attributes or indicate a level of further planning.

[0079] In one extension of the invention, the third action sequence replaces the first action sequence and / or the second action sequence, or the third action sequence follows the first action sequence and / or the second action sequence in time, or the third action sequence stops the technical process. The advantage of this is that the technical process can be re-planned, deviations and errors can be corrected, and no action sequence is executed, thus interrupting the technical process.

[0080] In one extension of the invention, the parallelization option provides a description of the time parallelization of the hierarchical structured task.

[0081] The present invention also includes a computer program product comprising a computer program, wherein the computer program can be loaded into a storage device of a computing unit, wherein when the computer program is executed on the computing unit, the steps of the method according to the present invention are performed by the computer program.

[0082] The present invention also includes a computer-readable medium having a computer program stored thereon, wherein the computer program is loadable into a storage device of a computing unit, and wherein when the computer program is executed on the computing unit, the steps of the method according to the present invention are performed by the computer program.

[0083] Manually creating action sequences is time-consuming, rigid, and error-prone. Modifications during flexible combinations or parallel processes further exacerbate this problem. This invention significantly reduces time consumption because modeling only needs to be done once. The planning component then creates flexibility and ensures that a large number of errors can be eliminated in association with the monitoring unit. The core of this approach is the use of a hierarchical structure in modeling to obtain a hierarchical structure that manages the planning task.

[0084] Alternative solutions are generally not applicable because either the problem is too complex or it cannot be solved without parallelization.

[0085] Parallelization is considered in the planning stage, allowing for highly detailed strategies tailored to existing hardware. This can include, for example, the number of CPU cores or the number of AGVs available. Furthermore, detailed dependencies can be considered, particularly which robots share the workspace; hierarchical planning allows for the layered specification of parallelization-related aspects: from discrete or symbolic to continuous or sub-symbolic. Maintaining the complexity of multiple such dependencies is a major advantage of hierarchical schemes.

[0086] This invention provides the further advantage of allowing multiple components to interact to ensure flexible, error-free, and parallel execution in large, heterogeneous domains. In this case, models, sub-plans, and, if necessary, error states are exchanged.

[0087] The fully automated calculation and parallel execution of the plan lead to real-world hardware tuning. Through the interaction of execution with hierarchical planning aided by biases and errors, an adaptive approach to responding to external influences is generated. Consequently, sensor measurements used for execution alter the behavior of the autonomous system.

[0088] Hierarchical planning of parallel processes is relevant in a wide range of industrial applications. It can be addressed through standardized modeling and planning in all areas where it holds potential in autonomous systems and various aspects of Industry 4.0. Examples mentioned using existing technologies provide an impression of the application areas. These examples range from advanced MES to simulation tools such as process simulation and co-simulation, and even traffic control systems.

[0089] This invention includes the following key algorithms:

[0090] - Hierarchical planning components for parallelism: To date, hierarchical planning components have not collected time information during plan creation and have created sequential plans. Extending the hierarchical scheme to scheduling is a core aspect of this invention, where the time availability of instances is essential, thus prioritizing plans with a high proportion of parallelization.

[0091] - Parallelization Unit: Analysis planning and identification sequences are essential for parallel control. This component does not exist in sequential hierarchical planning. Similar task creation has been considered in other domains, particularly Petri nets.

[0092] - Execution and Feedback: Parallel execution is based on the coordinated initiation and subsequent synchronization of multiple processes. Such algorithms are known in numerous applications. A specific aspect of this invention is the adaptation of control to deviations and errors through feedback. For this purpose, the feedback must be detailed enough to accurately reconstruct which process, with which input data, caused the feedback. Furthermore, it may be necessary to terminate other parallel processes based on this detailed information, or at least not to initiate new processes on the parallel series.

[0093] - Adaptation after feedback: Execution information must be incorporated into the adaptation process to prevent recurrence of problems. In other words, simply restarting the planning process does not solve the problem. Backtracking in hierarchical planning allows for appropriate responses to this information. That is, it involves continuously searching for suitable levels of abstraction where alternative plans exist that can achieve the goal without a failed process. Then, the process proceeds based on these alternative plans, just as in nominal hierarchical planning for parallel processes. Attached Figure Description

[0094] The particularity and advantages of the present invention will become apparent from the following explanation of several embodiments based on schematic diagrams.

[0095] Figure 1 A flowchart of the method according to the present invention is shown.

[0096] Figure 2 A flowchart for generating action sequences is shown, and

[0097] Figure 3 The layering is illustrated. Detailed Implementation

[0098] Figure 1A flowchart of a method for planning a technical process according to the present invention is shown, wherein the technical process has hierarchical structured tasks, and wherein the technical process (P) includes modification of physical units.

[0099] It includes the following steps:

[0100] - Step S1: Read the dependency description of the hierarchical structured task.

[0101] - Step S2: Derive parallelization options from the dependency description.

[0102] Step S3: Generate a first action sequence with the parallelization option included, wherein the first action sequence performs the hierarchical structured task.

[0103] Step S4: Receive process data, wherein the process data includes the actual state of the technical process.

[0104] Step S5: Generate a second action sequence based on the process data, wherein the second action sequence performs the hierarchical structured task.

[0105] Step S6: Receive user input, and

[0106] - Step S7: Generate a third action sequence based on the user input.

[0107] Then, in particular, the hierarchical structured task A of the S3b (not shown) technical process P can be performed by executing the first action sequence AS 1.

[0108] Furthermore, the hierarchical structured task A of the S5b (not shown) technical process P can then be executed by executing the second action sequence AS2.

[0109] This flowchart should not be viewed as strictly sequential. Steps S1 through S6 can also be performed in other orders. In particular, if input is received from a sensor, especially process data (step S4) or from a user (step S6), the already performed steps can be repeated. Input from a sensor, especially process data (step S4) or from a user (step S6) can, in particular, lead to changes in the planning or parallelization of the task.

[0110] Figure 2 A flowchart illustrating the action sequence of the generation technique P is shown, wherein the technique has a hierarchical structured task A. The hierarchical structured task A is read, and a first action sequence AS1 is generated through steps S1, S2, and S3 (see...). Figure 1 Then, the second action sequence AS2 is generated through steps S4 and S5 (see...). Figure 1The third action sequence AS3 can be generated at any time through steps S6 and S7. The second action sequence AS2 can replace, supplement, and / or provide an alternative to the first action sequence AS1. This also applies to the third action sequence AS3 relative to the second action sequence AS2.

[0111] Figure 3 The hierarchical description HA of hierarchically structured tasks A1 and A2 is shown, where each hierarchical description HA determines the hierarchical level within the process, such that the first task A1 of the first hierarchical level HL1 is composed of multiple second tasks A2a, ..., A2c on the second hierarchical level HL2. Figure 3 The process is further refined into three second tasks, where the first task A1 and the second task A2 are part of a hierarchical structured task A. The planning then determines which tasks should be combined in the second hierarchical level to achieve the goal of the more abstract task A1.

[0112] Although the invention has been described and illustrated in more detail by way of examples, the invention is not limited to the disclosed examples, and those skilled in the art can derive other variations therefrom without departing from the scope of protection of the invention.

[0113] List of reference numerals

[0114] A hierarchical structured task

[0115] AS 1 First Action Sequence

[0116] AS2 Second Action Sequence

[0117] AS3 Third Action Sequence

[0118] A1 First Task

[0119] A2a, b, c At least one second task

[0120] HA Layering Explanation

[0121] HL1 First tier level

[0122] HL2 Second Tier

[0123] Si step i,i=[1;7]

[0124] P technology process

Claims

1. A computer-implemented method for performing a technical process (P), wherein the technical process (P) has hierarchically structured tasks (A), and the technical process (P) is performed by at least one production unit. The method comprises the following steps: - Read the dependency description of the hierarchical structured task (A) described in (S1), - Derive the (S2) parallelization options from the aforementioned dependency specification. - Generate (S3) a first action sequence (AS1) with the parallelization option included, wherein the first action sequence (AS1) performs the hierarchical structured task (A). - The hierarchical structured task (A) is performed (S3b) in the technical process (P) by executing the first action sequence (AS1). - Receive (S4) process data, wherein the process data includes the actual state of the technical process (P), - Generate a second action sequence (AS2) based on the process data (S5), and - A hierarchical structured task (A) of the technical process (P) to be executed (S5b) based on the second action sequence (AS2). in: - The second action sequence (AS2) replaces the first action sequence (AS1), or - The second action sequence (AS2) follows the first action sequence (AS1) in time, or - The second action sequence (AS2) stops the technical process (P). During parallel execution, a check is performed to determine whether all actions in the first action sequence (AS1) satisfy the conditions calculated during planning. If the result of the first action sequence (AS1) is outside the expected range, or if the first action sequence (AS1) cannot be successfully terminated, then the second action sequence (AS2) is created.

2. The method according to claim 1, wherein the at least one production unit is configured as a production machine.

3. The method of claim 1, wherein the at least one production unit is configured as a production robot.

4. The method of claim 1, wherein the at least one production unit is configured as an automated unit.

5. The method of claim 1, wherein the technical process (P) is configured as a manufacturing process.

6. The method of claim 1, wherein the technical process (P) is configured as a logistics process.

7. The method of claim 1, wherein the technical process (P) is configured as a traffic control process.

8. The method of claim 1, wherein the technical process (P) is configured as a transportation process.

9. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) with respect to procedural order.

10. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) regarding causal dependencies.

11. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) regarding interaction relationships.

12. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) regarding time information.

13. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) for the required resources.

14. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) for tasks that are concurrently performed in a mandatory time.

15. The method of claim 1, wherein the dependency specification defines a hierarchical structured task (A) for the required tool.

16. The method according to any one of claims 1-15, wherein the dependency description indicates dependencies that occur when performing the hierarchical structuring task (A).

17. The method according to any one of claims 1-15, wherein the hierarchical structured task (A) has a hierarchical description (HA). Each hierarchical description (HA) defines a hierarchical level within the process, such that a first task (A1) of a first hierarchical level (HL1) is further refined by at least one second task (A2a, A2b, A2c) of a second hierarchical level (HL2), wherein the first task (A1) and the second task (A2) are part of the hierarchical structured task (A).

18. The method according to any one of claims 1-15, Further methodological steps are provided: - Receive (S6) user input, and - Generate a third action sequence (AS3) based on the user input (S7).

19. The method of claim 18, wherein: - The third action sequence (AS3) replaces the first action sequence (AS1) and / or the second action sequence (AS2), or - The third action sequence (AS3) follows the first action sequence (AS1) and / or the second action sequence (AS2) in time, or - The third action sequence (AS3) stops the technical process (P).

20. The method according to any one of claims 1-15, wherein the parallelization option provides a description of the time parallelization of the hierarchical structured task (A).

21. A computer program product comprising a computer program, wherein the computer program is loadable into a storage device of a computing unit, wherein when the computer program is executed on the computing unit, the steps of the method according to any one of claims 1 to 20 are performed by the computer program.

22. A computer-readable medium having a computer program stored thereon, wherein the computer program is loadable into a storage device of a computing unit, wherein when the computer program is executed on the computing unit, the steps of the method according to any one of claims 1 to 20 are performed by the computer program.

Images