Program creation device and computer-readable storage medium

Abstract

A program creation device according to the present disclosure: receives processing content in a multimodal format; analyzes the processing content and a processing condition according to the format of the processing content; creates a prompt including the processing content, the processing condition, and a processing program creation request; transmits the prompt to a generative AI; receives the processing program created by the generative AI; and outputs the processing program created by the generative AI.

Description

Program creation device and computer-readable storage medium

[0008]

[0001] This disclosure relates to a program creation device and a computer-readable storage medium.

[0002] [Correction based on Rule 91, 28.04.2026] Conventionally, there is a technology for generating programming code using artificial intelligence. (For example, Patent Document 1.)

[0003] Japanese Patent No. 7382095

[0004] [Correction based on Rule 91, 28.04.2026] The machining program of a machine tool is created in CAD (Computer Aided Design) to create a 3D model of the workpiece. When the model design is completed, in CAM (Computer Aided Modeling), a tool path is generated and the tool path is converted (post-processed) into a machining program.

[0005] When changing the machining content (machining shape), it is necessary to go through the processes of CAD and CAM, and there is a transfer of work between the numerical control device and the work on the PC (Personal Computer), which is cumbersome.

[0006] It is desired to create a machining program easily.

[0007] A program creation device according to one aspect of the present disclosure includes a reception unit that receives machining content in a multimodal format, a condition analysis unit that analyzes the machining content and machining conditions according to the format of the machining content, a prompt creation unit that creates a prompt including the machining content, the machining conditions, and a request for creating a machining program, a transmission unit that transmits the prompt to a generation AI, a reception unit that receives the machining program created by the generation AI, and an output unit that outputs the machining program created by the generation AI.

[0008] This is a block diagram of the program creation device. This is an example of image data showing the machining process. This is an example of modeling data showing the machining process. This is a diagram showing an example of a prompt when the user's instructions are in natural language only. This is a diagram showing an example of a prompt when the user's instructions are in natural language and image data. This is a flowchart explaining the operation of the program creation device. This is an example of image data showing the marking process. This is an example of modeling data showing the marking process. This is a diagram showing an example of a prompt when the user's instructions are in natural language only. This is a diagram showing an example of a prompt when the user's instructions are in natural language and image data. This is an example of a template that accepts changes to the machining program and marking process. This is a diagram showing an example of a change in the machining program. This is a diagram showing an example of a prompt. This is a schematic diagram showing changes in the machining program and marking process. This is a hardware configuration diagram of the program creation device.

[0009] Embodiments of this disclosure will be described below with reference to the drawings. In the following description, components having the same or similar functions will be denoted by the same reference numerals. Duplication of these components may be omitted.

[0010] In this application, "based on XX" means "based on at least XX," and includes cases where it is based on another element in addition to XX. Furthermore, "based on XX" is not limited to cases where XX is used directly, but also includes cases where it is based on something that has been calculated or processed. "XX" is any element (for example, any information).

[0011] [Correction based on Rule 91, April 28, 2026] The program creation device 100 according to this embodiment is implemented, for example, in a numerical control device. The program creation device 100 can also be implemented on a personal computer attached to the numerical control device, or on a personal computer, cell computer, fog computer, cloud server, or other computer connected to the numerical control device via a wired / wireless network.

[0012] The program creation device 100 is connected to the AI ​​server via a network. The AI ​​server may be distributed. The AI ​​server includes a generating AI. When predetermined data is input, the generating AI learns the probability of the predetermined data being the response. By learning from a large number of samples, the generating AI learns to output text with a high probability as a response to predetermined data when it is input. The generating AI can be used for purposes such as dialogue, question and answer, text summarization, text editing, text translation, text transformation, text modification, text optimization, text interpretation / detection, recognition, prediction, judgment, image generation, and overall judgment.

[0013] The generating AI is assumed to have learned enough data related to processing programs to generate them. This learning data includes processing programs, manuals, etc.

[0014] [First Embodiment] Figure 1 is a block diagram of the program creation device 100 according to the first embodiment. The program creation device 100 comprises a reception unit 1, a condition analysis unit 2, a prompt creation unit 3, a transmission unit 4, a receiving unit 5, an output unit 6, a program update unit 7, and a storage unit 8.

[0015] The reception unit 1 receives requests for the creation of a machining program and the machining details. The machining details refer to information representing the machining shape. The machining details can be specified in a multimodal manner. That is, the reception unit 1 accepts data in various formats, such as natural language (text), image data, and modeling data. Image data is a bit image, where an image is represented by pixels. Modeling data is vector data that represents straight lines, curves, etc., by combining points, lines, polygons, etc. Modeling data is created using CAD.

[0016] The condition analysis unit 2 analyzes the current machining conditions and the machining content specified by the user. Machining conditions are necessary for creating a machining program, but they are conditions that cannot be expressed through natural language dialogue. Typically, in dialogue, only the parts to be changed are indicated, assuming that preconditions are shared. When creating a machining program from information obtained through dialogue, there is a possibility that necessary information will be missing. Therefore, the condition analysis unit 2 holds the machining conditions that are expected to be necessary for creating the machining program at this stage, and adds the machining content specified by the user via prompts.

[0017] The machining conditions are assumed to be already set in the numerical control device or machining program. If they are not set, the condition analysis unit 2 may prompt the user to set the machining conditions. The machining conditions are not particularly limited. In this embodiment, the machining conditions relate particularly to the machining shape of the workpiece. For example, in the case of drilling, these may include the center coordinates and radius of the hole. The machining conditions may also include information about the workpiece. This information may include the shape of the workpiece, the machining surface, and the machining start position.

[0018] The processing content is multimodal. Reception unit 1 accepts processing content in various formats, including natural language, image data, and modeling data.

[0019] The method of conditional analysis differs depending on the data format of the processing content. If the processing content is "natural language only," the processing content is text. The reception unit may accept natural language in the form of structured text entered into a template (described later), or it may accept text in a conversational style (voice spoken by the user) as natural language. In the conversational style, a dialogue screen or similar is displayed to receive instructions from the user, such as "Add a circular hole with radius XX at position XX" or "Change the corner radius to XX."

[0020] When user instructions are in the form of "natural language and image data," the processing details are represented by an image. Figure 2 shows an example of image data indicating processing details. In this example, a circular hole is specified as the processing detail. The condition analysis unit 2 may pass the image data directly to the prompt creation unit 3, or it may convert the information obtained from the image data into text and pass that information to the prompt creation unit 3. By analyzing the image data using technologies such as shape recognition, processing details such as coordinates and processing shape can be converted into text. Furthermore, if the image data is a drawing, processing details such as dimensions and processing method are expressed in text, so the text in the drawing may be recognized and converted into text according to existing drafting rules.

[0021] When user instructions are in the form of "natural language and modeling data," the processing content is represented by combinations of points, lines, polygons, etc. Figure 3 shows an example of modeling data illustrating the processing content. The processing content of modeling data can also be converted into text using character matching or character recognition, similar to image data.

[0022] The prompt generation unit 3 creates prompts based on the analysis results of the condition analysis unit 2, the processing details entered by the user, and the user's instructions. The content of the prompts varies depending on the type of data.

[0023] [Correction based on Rule 91 28.04.2026] When user instructions are in "natural language only", a prompt is created that includes the current machining conditions and the machining content. Figure 4 is an example of a prompt. The prompt in Figure 4 includes the current machining conditions and the machining content instructed by the user. The current machining conditions are text such as "Create a round hole", "The depth is XX", "The machining center of the round hole is XX, XX, XX", "The radius is XX", "The tool is XX". The machining content instructed by the user is text such as "Add a round hole", "The machining center is XX", "The radius is XX".

[0024] If the user's instructions consist of "natural language + image data," the prompt generation unit adds the image data to the prompt created for natural language only, creating a prompt such as "Create a machining program to process the workpiece so that it has the same shape as the image data," as shown in Figure 5. Alternatively, machining details determined using image recognition or character recognition may be added to the prompt. For example, "Add a round hole," or "The machining center is XX, and the radius is XX."

[0025] [Correction based on Rule 91 28.04.2026] When the user's instruction is "natural language + modeling data", the prompt creation unit 3 adds the modeling data to the prompt created when only natural language is used, and creates a prompt such as "Please create a machining program that processes the workpiece so that it has the same shape as the modeling data". In addition, machining details determined using image recognition or character recognition may be added to the prompt. For example, "Add a round hole", "The machining center is ○○, and the radius is ○○", etc.

[0026] The prompt generation unit 3 may add a processing program as an example to the prompt. Examples include the processing program being modified this time, or processing programs created in the past. Past processing programs are, for example, processing programs with similar processing content. The language model performs in-context learning by reading example sentences, which may improve the accuracy of program creation.

[0027] The transmission unit 4 sends the generated prompt to the AI ​​server. The language model creates a processing program based on the prompt. However, since the language model is based on statistics, the same processing program will not be created every time. Therefore, it may be possible to create multiple processing programs and select the one that works as expected.

[0028] The receiving unit 5 receives the machining program created by the language model. The program update unit 7 stores the machining program received by the receiving unit 5 in the storage unit 8. The program update unit 7 determines whether the machining program operates as expected. It also selects the better of the machining programs that operate as expected. One example of a selection method is simulation. Based on the simulation results, it is determined whether the machining content is as expected. Alternatively, simulation may be used to select a machining program that speeds up the cycle time.

[0029] The operation of the program creation device 100 will be explained according to the flowchart in Figure 6. The program creation device 100 receives the machining content and a request for machining program creation (step S1). The machining content is multimodal. The format of the machining content can be natural language, image data, modeling data, etc.

[0030] The condition analysis unit 2 analyzes the machining process. In analyzing the machining process, first, the current machining conditions are read out (step S2). In this embodiment, the machining conditions relate to the machining shape of the workpiece.

[0031] If the processing instructions are in natural language only (Step S3: Natural Language), the condition analysis unit 2 passes the received natural language and the current processing conditions to the prompt creation unit 3. The prompt creation unit 3 creates a prompt that includes the user's instructions and the current processing conditions (Step S4).

[0032] [Correction based on Rule 91 28.04.2026] If the processing content is natural language and image data (Step S3: Image data), the condition analysis unit 2 may determine the processing content using image recognition, character recognition, etc. The condition analysis unit 2 passes the received natural language and image data to the prompt creation unit 3. The prompt creation unit 3 creates a prompt that includes the user's instructions, the current processing conditions, the image data, and the processing content determined from the image data (Step S5).

[0033] If the processing content consists of natural language and modeling data (Step S3: Modeling data), the condition analysis unit 2 may determine the processing content using character matching, character recognition, etc. The condition analysis unit 2 passes the received natural language and modeling data to the prompt creation unit. The prompt creation unit 3 creates a prompt that includes the user's instructions, the current processing conditions, the modeling data, and the processing content determined from the modeling data (Step S6).

[0034] The transmitting unit 4 sends the created prompt to the AI ​​server (step S7). The language model creates a processing program based on the prompt. The receiving unit 5 receives the processing program created by the language model (step S8).

[0035] The program update unit 7 determines whether the created machining program operates as expected (step S9), and stores the created machining program in the storage unit 8 (step S10).

[0036] As explained above, the program creation device 100 uses generation AI to create machining programs for machine tools. Machining details can be expressed using natural language, image data, modeling data, etc. The program creation device creates prompts based on the current machining conditions and the natural language, image data, modeling data, etc. that instruct the machining details, and causes the language model to create the machining program.

[0037] Users can change the processing content simply by entering (or speaking) text, or by specifying image data or modeling data. Because it uses generative AI, it offers greater processing flexibility than creating processing patterns from predetermined characters and action patterns. Creating diverse processing programs may also improve cycle time.

[0038] The generating AI may create processing programs from modeling data rather than from text. Modeling data, which represents the shape of strings using mathematical formulas, may be advantageous for conversion into processing programs.

[0039] The program creation device 100 includes a storage unit 8 that stores machining programs created in the past. The program update unit 7 can compare previously created machining programs with newly created machining programs and select a better machining program.

[0040] Furthermore, it is possible to store and use previously created processing programs for learning. By creating prompts that include the processing programs created for processing and having the generating AI read them, it is also possible to perform in-context learning on the language model.

[0041] [Second Embodiment] The program creation device 100 of the second embodiment creates a marking processing program. Marking is a processing method in which a pattern is drawn on the surface of a workpiece using a drill or laser. The reception unit 1 receives a request to create a processing program that performs marking, and the marking content. The marking content refers to information that represents the pattern to be marked. The marking content can be specified in a multimodal manner. That is, the reception unit 1 accepts data in various formats such as natural language (text), image data, and modeling data. Image data is a bit image in which an image is represented by pixels. Modeling data is vector data that represents straight lines, curves, etc., by combining points, lines, polygons, etc. Modeling data is created using CAD.

[0042] The condition analysis unit 2 analyzes the current marking conditions and the marking content specified by the user. The marking conditions are necessary for creating the machining program but cannot be expressed through natural language dialogue. The condition analysis unit 2 holds the marking conditions that are currently expected to be necessary for creating the machining program and adds the marking content specified by the user via prompts.

[0043] The marking conditions shall be those already set in the numerical control device, machining program, etc. If they are not set, the condition analysis unit 2 may prompt the user to set the marking conditions. The marking conditions are not particularly limited. Examples of the marking conditions include character size, character spacing, line spacing, font, marked character string (including special characters), etc. The marking conditions may include information on the workpiece. Examples of the workpiece information include the shape of the workpiece, the surface to be marked, the starting position of the marking, etc.

[0044] The marking content is multimodal. The reception unit 1 receives the marking content in various forms such as natural language, image data, modeling data, etc.

[0045] The method of condition analysis varies depending on the data format of the marking content. When the marking content is "only natural language", the marking content is text. The reception unit may receive natural language in the form of structured text input into a template (described later), or may receive text in a style that interacts with humans (the voice spoken by the user) as natural language. In the form of interacting with humans, a chat screen, etc. is displayed, and instructions from the user such as "Create a machining program that marks the manufacturing date as today's date", "Create a machining program that marks the manufacturing number 'LLM112'" are received.

[0046] [Correction based on Rule 91, 28.04.2026] When the user's instruction is "natural language and image data", the marking content is represented by an image. FIG. 7 is an example of image data showing the marking content. In this example, the characters "AB" are specified as the marking content. The condition analysis unit 2 may directly pass the image data to the prompt creation unit 3. The condition analysis unit 2 may convert the marking content into text using technologies such as character matching (OCV), character recognition (OCR), etc. The condition analysis unit 2 determines the marked character string, the font of the character string, etc., and passes it to the prompt creation unit 3 as the text-converted marking content.

[0047] When the user's instruction is "natural language and modeling data", the marking content is expressed as a combination of points, lines, polygons, etc. FIG. 8 is an example of modeling data showing the marking content. Similar to image data, the marking content of the modeling data can also be texturized using character collation or character recognition.

[0048] The prompt creation unit 3 creates a prompt based on the analysis result of the condition analysis unit 2, the marking content input by the user, and the user's instruction. The content of the prompt varies depending on the type of data.

[0049] When the user's instruction is "natural language only", a prompt including the current marking condition and the marking content is created. FIG. 9 is an example of a prompt. The prompt in FIG. 9 includes the current marking condition and the marking content instructed by the user. The current marking condition is text such as "The character size is ○○, the character spacing is ○○, the line spacing is ○○, and the font is ○○. The work is ○○" and "The device to be used is ○○". The marking content instructed by the user is text such as "Please create a processing program to mark today's date as the manufacturing date" and "Please create a processing program to mark the manufacturing number 'LLM112'".

[0050] When the user's instruction is "natural language + image data", the prompt creation unit adds the image data to the prompt created in the case of natural language only, and creates a prompt such as "Please read the character string of the image data and mark the same character string" as shown in FIG. 10. Also, the marking content determined using character collation or character recognition may be added to the prompt. For example, "Use the font ○○" and "The character string is ○○".

[0051] [Correction based on Rule 91 28.04.2026] When the user's instruction is "natural language + modeling data", the prompt creation unit 3 adds the modeling data to the prompt created for natural language only and creates a prompt such as "Read the string in the modeling data and mark the same string". It may also add the content of the marking determined using character matching or character recognition to the prompt. For example, "Use the font ○○", "The string is ○○", etc.

[0052] The prompt generation unit 3 may add example processing programs to the prompts. Examples include the processing program being modified this time, or processing programs created in the past. Past processing programs are, for example, processing programs with similar marking content. The language model performs in-context learning by reading example programs, which may improve the accuracy of program creation.

[0053] The transmission unit 4 sends the generated prompt to the AI ​​server. The language model creates a processing program based on the prompt. However, since the language model is based on statistics, the same processing program will not be created every time. Therefore, it may be possible to create multiple processing programs and select the one that works as expected.

[0054] The receiving unit 5 receives the machining program created by the language model. The program update unit 7 stores the machining program received by the receiving unit 5 in the storage unit 8. The program update unit 7 determines whether the machining program operates as expected. It also selects the better of the machining programs that operate as expected. One example of a selection method is simulation. Based on the simulation results, it is determined whether the marking content is as expected. Alternatively, simulation may be used to select a machining program that speeds up the cycle time.

[0055] Marking is a process that involves imprinting patterns on the surface of a workpiece using a drill or laser. In marking, quality control information such as serial numbers and manufacturing dates may be engraved onto the product. Because quality control information changes frequently, creating a new machining program using CAD / CAM each time a change occurs is cumbersome. Functions exist that allow for the creation of marking machining programs without using CAD / CAM. However, these functions create machining programs from predetermined characters and motion patterns, lacking machining flexibility and leaving room for improvement in cycle time.

[0056] In the program creation device 100 of the second embodiment, since the machining program is created using generated AI, the marking machining program can be created without using CAD / CAM. Furthermore, since the machining program is not created from a predetermined operation pattern, the machining program becomes more flexible, and the cycle time may be improved.

[0057] [Third Embodiment] In the third embodiment, the program creation device 100 has a condition analysis unit 2 that identifies the parts of the machining process to be changed and creates a machining program for only the identified parts to be changed. The parts of the machining process to be changed are mainly patterns such as strings of characters. The workpiece shape and the position of the machining surface are often not changed. The parts that do not need to be changed are left as they are, and only the parts that need to be changed are selected. Note that the program creation device 100 of the third embodiment has substantially the same configuration as the program creation device of the first embodiment, so only the differences will be explained. In addition, although marking machining is explained as an example in the third embodiment, it can be applied to other machining processes as well.

[0058] [Correction based on Rule 91 28.04.2026] The receiving unit 1 of the third embodiment uses a template. Figure 11 is an example of a template. The template allows specifying the processing program to be changed and the change in marking content. In the template of Figure 11, the processing program to be changed is "1", and the marking content is specified to be changed from the character "B" to the character "C".

[0059] The condition analysis unit 2 obtains the processing program before the change and identifies the parts of the processing program that have been changed. Figure 12 shows an example of the parts that have been changed. In the example in Figure 12, the program “G01 X100 Q100; G01 Y0.1 Q0; G01 X0. Q100;” has been identified. The parts that have been changed are the parts that include the marking process for the character “B”. Note that the processing paths are not independent for each character, so parts of adjacent characters may also be included in the parts that have been changed as needed.

[0060] The prompt creation unit 3 creates a prompt based on the changes read by the condition analysis unit 2 and the markings instructed by the user. Figure 13 shows an example of a prompt. In the prompt in Figure 13, the natural language message "Please update the machining program," the fragment of the machining program to be changed, and the marking content are defined. The marking content includes information about the characters to be changed ("Before change "B"", "After change "C"") and the marking conditions. The marking conditions include the type of machining "marking machining" and the type of machine tool "laser machining machine".

[0061] In the example described above, the marking content was received as text, but the marking content may also be received as image data or modeling data. The condition analysis unit 2 can determine the marking content using techniques such as character matching and character recognition. Based on the analysis results, the condition analysis unit 2 identifies the parts of the processing program that need to be changed.

[0062] Furthermore, the content of markings received as text can be reflected in the modeling data. Generating AI has technologies that convert text into modeling data (for example, TexttoCAD). Such technologies can be used to reflect user instructions in text into image data and modeling data.

[0063] The receiving unit 5 receives the processing program created by the language model. An example of a processing program is shown in Figure 14. The changes shown are only a part of the processing program. When an existing processing program is updated with the processing program created by the language model, the string to be marked is changed from "AB" to "AC".

[0064] The output unit 6 uses the processing program created by the language model to replace a part of the existing processing program and create a new processing program.

[0065] The program update unit 7 stores the created processing program in the storage unit 8. The created processing program includes fragments of the program created by the language model and new processing programs that replace parts of existing processing programs.

[0066] As described above, the program creation device 100 of the third embodiment identifies the parts of the marking content that need to be changed and causes the generation AI to create a machining program for only the identified parts. This shortens the length of the generated machining program and improves accuracy. In addition, limiting the parts to be changed reduces the load on the generation AI. Limiting the parts to be changed also allows for quick verification of the machining program.

[0067] The hardware configuration of the program creation device 100 to which this disclosure is applied will be described below. Figure 15 is a hardware configuration diagram of the program creation device 100. As shown in Figure 15, the program creation device 100 includes a CPU 111 that controls the program creation device 100 as a whole, a ROM 112 that records programs and data, and a RAM 113 for temporarily expanding data. The CPU 111 reads the system program recorded in the ROM 112 via the bus and determines the consistency between the blocks and the tools.

[0068] The non-volatile memory 114 is backed up, for example, by a battery (not shown), so that its stored state is maintained even when the power to the program creation device 100 is turned off. The non-volatile memory 114 stores various data, such as programs read from the external device 120 via interfaces 115, 118, and 119, and operation inputs entered via the input device 20. The non-volatile memory 114 may also store programs and data for executing the program creation device 100 of this embodiment.

[0069] Interface 115 is an interface for connecting the program creation device 100 to an external device 120 such as an adapter. Programs and various parameters are read from the external device 120. Interface 118 is an interface for connecting the program creation device 100 to a display device 30 such as a liquid crystal display. The display device 30 displays data read into memory, data obtained as a result of executing programs, etc. Interface 119 is an interface for connecting the program creation device 100 to an input device 20 such as a keyboard or pointing device. The input device 20 passes commands, data, etc. based on operator operations to the CPU 111 via interface 119.

[0070] While this disclosure has been described in detail, it is not limited to the individual embodiments described above. These embodiments can be added, replaced, modified, partially deleted, etc., in any way that does not depart from the gist of this disclosure or from the spirit of this disclosure derived from the claims and their equivalents. These embodiments can also be implemented in combination. For example, the order of operations and processes in the embodiments described above are given as examples only and are not limited thereto.

[0071] The following are additional notes regarding embodiments of the present disclosure. (Note 1) The program creation device (100) includes a receiving unit (1) that receives processing content in a multimodal format, a condition analysis unit (2) that analyzes processing content and processing conditions according to the format of the processing content, a prompt creation unit (3) that creates a prompt including processing content, processing conditions, and a request to create a processing program, a transmitting unit (4) that transmits the prompt to a generating AI, a receiving unit (5) that receives a processing program created by the generating AI, and an output unit (6) that outputs the processing program created by the generating AI. (Note 2) The output unit (6) includes a program update unit (7) that stores the processing program created by the generating AI in a storage unit. (Note 3) The receiving unit (1) receives a request to change the processing content of an existing processing program, and the condition analysis unit (2) identifies the parts of the processing content to be changed in the existing processing program. (Note 4) The prompt creation unit (3) creates a prompt that includes the parts of the processing program to be changed. (Note 5) The receiving unit (5) receives a processing program that changes the modified parts to the processing content instructed by the user, and the output unit (6) replaces the modified parts with the received processing program to create a new processing program. (Note 6) The multimodal format is at least one of natural language, image data, and modeling data. (Note 7) The condition analysis unit (2) converts the processing content expressed in image data or modeling data into text. (Note 8) The prompt creation unit (3) includes processing programs created in the past as example sentences in the prompts. (Note 9) The computer-readable storage medium (112, 113, 114) stores instructions for one or more processors (111) to execute processing, which include receiving processing details in a multimodal format, analyzing marking details and marking conditions according to the format of the processing details, creating a prompt including processing details, processing conditions and a request for the creation of a processing program, transmitting the prompt to a generating AI, receiving the processing program created by the generating AI, and outputting the processing program created by the generating AI.

[0072] 100 Program creation device 1 Reception unit 2 Condition analysis unit 3 Prompt creation unit 4 Transmission unit 5 Receiving unit 6 Output unit 7 Program update unit 8 Storage unit 111 CPU 112 ROM 113 RAM 114 Non-volatile memory

Claims

1. A program creation device comprising: a receiving unit that receives processing details in a multimodal format; a condition analysis unit that analyzes processing details and processing conditions according to the format of the processing details; a prompt creation unit that creates a prompt including processing details, processing conditions, and a request for the creation of a processing program; a transmission unit that transmits the prompt to a generation AI; a receiving unit that receives a processing program created by the generation AI; and an output unit that outputs the processing program created by the generation AI.

2. The program creation apparatus according to claim 1, wherein the output unit includes a program update unit that stores the processing program created by the generating AI in the storage unit.

3. The program creation apparatus according to claim 1, wherein the receiving unit receives a request to change the processing content of an existing processing program, and the condition analysis unit identifies the parts of the existing processing program that need to be changed.

4. The program creation device according to claim 3, wherein the prompt creation unit creates prompts that include changes in the processing program.

5. The program creation device according to claim 3, wherein the receiving unit receives a processing program that changes the modified portion to the processing content instructed by the user, and the output unit replaces the modified portion with the received processing program to create a new processing program.

6. The program creation apparatus according to claim 1, wherein the multimodal format is at least one of natural language, image data, and modeling data.

7. The program creation device according to claim 1, wherein the condition analysis unit converts the processing content expressed in image data or modeling data into text.

8. The program creation device according to claim 1, wherein the prompt creation unit includes a previously created machining program as an example in the prompt.

9. A computer-readable storage medium that stores instructions for one or more processors to execute a process, which includes receiving processing details in a multimodal format, analyzing the processing details and processing conditions according to the format of the processing details, creating a prompt including the processing details, processing conditions and a request for the creation of a processing program, transmitting the prompt to a generating AI, receiving the processing program created by the generating AI, and outputting the processing program created by the generating AI.

Images