Robot path planning method, robot, computer device and computer readable storage medium

Abstract

Embodiments of the present application disclose a robot path arrangement method, a robot, a computer device and a computer readable storage medium. The method comprises the following steps: performing standardization processing on an acquired scene map to obtain map data, the map data being composed of at least one structured point; matching a corresponding scene material for each structured point based on the map data; and arranging a work flow according to the scene material and the structured point, and issuing the work flow to the robot. Therefore, the present application can automatically generate a work flow based on a map and matching materials, reduce additional personnel operations, and realize end-to-end automation of a closed loop from map analysis to work flow landing.

Description

Technical Field

[0001] This application belongs to the field of path navigation technology, and in particular relates to a robot path arrangement method, a robot, a computer device, and a computer-readable storage medium. Background Technology

[0002] With the large-scale application of robots in exhibition halls, scenic spots, parks, and other scenarios, the efficiency and adaptability of workflow orchestration have become core requirements. Robot workflows need to be combined with the spatial layout of the scene, connecting the action and voice execution logic of multiple points to ensure that the service is consistent and fits the scenario.

[0003] Current robot workflow orchestration technologies primarily employ a "manual input + fixed template" model, requiring manual marking of point coordinates on maps and manual classification of point types. Action and voice materials must be manually selected and matched, lacking contextual intelligent adaptation capabilities. Sequential orchestration relies on manually setting the order, without incorporating map spatial layout optimization, potentially leading to redundant robot paths and chaotic execution logic. While some solutions support simple point recognition, their recognition accuracy is low, classification is limited, and material matching and sequential orchestration still require manual intervention, resulting in insufficient intelligence. Therefore, improving the intelligence of robot path orchestration is a crucial technical problem that urgently needs to be solved by those skilled in the art.

[0004] The preceding description is intended to provide general background information and does not necessarily constitute prior art. Summary of the Invention

[0005] The purpose of this application is to provide a robot path orchestration method, a robot, a computer device, and a computer-readable storage medium that can automatically generate workflows based on maps and matching materials, reducing additional human intervention.

[0006] To achieve the above objectives: In a first aspect, embodiments of this application provide a robot path orchestration method, comprising the following steps: performing standardization processing on an acquired scene map to obtain map data, the map data consisting of at least one structured point; matching corresponding scene material to each structured point based on the map data; orchestrating a workflow according to the scene material and structured points, and sending the workflow to the robot.

[0007] In an optional embodiment of this application, the acquired scene map is subjected to standardization processing to obtain map data, including: performing a validity check on the scene map; when the validity check passes, extracting environmental features and completing coordinate calibration to generate a structured point list, the structured point list containing environmental features and at least one structured point; assigning a unique ID to the structured point list, binding it with the scene map as map data, and storing it.

[0008] In an optional embodiment of this application, the validity check of the scene map is performed, including: determining whether the scene map meets the preset verification conditions, the verification conditions including at least one of the following: the scene map has built-in coordinate parameters, the coordinate parameters do not have missing anomalies, and the scene map resolution is greater than a preset threshold; when the scene map meets all the verification conditions, the validity check is determined to be passed; when the scene map does not meet any of the verification conditions, the validity check is determined to be failed.

[0009] In an optional embodiment of this application, matching corresponding scene materials for each structured point based on map data includes: determining the robot execution requirements matched by the map data; obtaining the corresponding material repository based on the robot execution requirements; and retrieving scene materials corresponding to each structured point from the material repository.

[0010] In an optional embodiment of this application, retrieving scene materials corresponding to each structured point from the material repository includes: determining whether there is a corresponding scene material for the structured point; if there is, directly matching the corresponding scene material; if there is, obtaining the environmental keywords of the structured point, inputting the environmental keywords into a preset generation model for processing, and obtaining the corresponding customized scene material; and matching the customized scene material with the structured point.

[0011] In an optional embodiment of this application, retrieving scene materials corresponding to each structured point from the material repository includes: determining the execution device type for each structured point; and configuring the output method of the scene materials based on the execution device type.

[0012] In an optional embodiment of this application, a workflow is obtained by arranging scene materials and structured points, including: grouping scene materials to obtain multiple material groups; determining the priority of each structured point; constructing temporal dependencies based on material groups and priorities, and arranging them to obtain an initial workflow; and obtaining the final workflow after performing dynamic simulation verification on the initial workflow.

[0013] Secondly, embodiments of this application provide a robot, including a processor and a memory; the processor is used to execute a computer program stored in the memory to implement the workflow output by any one of the robot path orchestration methods as claimed in claims 1 to 7.

[0014] Thirdly, embodiments of this application provide a computer device, including: a processor and a memory storing a computer program, wherein when the processor runs the computer program, the steps of the above-described method are implemented.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0016] The embodiments of this application have the following beneficial effects: This application provides a robot path orchestration method, including the following steps: performing standardization processing on the acquired scene map to obtain map data, the map data consisting of at least one structured point; matching corresponding scene materials to each structured point based on the map data; orchestrating a workflow based on the scene materials and structured points, and distributing the workflow to the robot. Therefore, this application can automatically analyze the scene map and accurately identify the functional attributes and spatial relationships of points to obtain map data. It also automatically adapts scene materials using the structured points in the map data, avoiding manual repetitive annotation and material matching, significantly shortening the deployment cycle. Simultaneously, planning the robot route based on points and materials ensures that the route combines task efficiency and execution robustness, avoiding problems such as path repetition and inconsistent execution logic caused by manual intervention alone, truly achieving an end-to-end automated closed loop from map parsing to workflow implementation.

[0017] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it according to the contents of the specification, and to make the above and other objects, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit this application. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a robot path orchestration method provided in one embodiment.

[0020] Figure 2 This is a schematic block diagram of the structure of a computer device provided in one embodiment. Detailed Implementation

[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. In the following description relating to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements.

[0022] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0023] It should be understood that although the terms first, second, third, etc., may be used herein to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this document, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if," as used herein, can be interpreted as "when," "when," or "in response to determination." Furthermore, as used herein, the singular forms "a," "an," and "the" are intended to also include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprising," "including," indicate the presence of the stated feature, step, operation, element, component, item, kind, and / or group, but do not exclude the presence, occurrence, or addition of one or more other features, steps, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" as used herein are to be interpreted as inclusive, or mean any one or any combination thereof. Therefore, "A, B, or C" or "A, B, and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A, B, and C". Exceptions to this definition will only occur if the combination of elements, functions, steps, or operations is inherently mutually exclusive in some way.

[0024] It should be understood that although the steps in the flowcharts of this application's embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0025] It should be noted that step designations such as S110 and S120 are used in this document for the purpose of more clearly and concisely describing the corresponding content, and do not constitute a substantial limitation on the order. In specific implementation, those skilled in the art may execute S120 first and then S110, etc., but these should all be within the protection scope of this application.

[0026] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0027] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0028] To facilitate understanding of this application, the following explanations are provided for the terms and technical objects that may be used in this application: Scene map: A digital map containing the spatial layout of the real scene, coordinates of points, and environmental features (such as exhibits, facilities, and terrain), providing spatial basis data for robot workflow orchestration.

[0029] AI Large Model Point Recognition and Classification: By using pre-trained AI visual large models (such as multimodal large models) to perform pixel-level analysis on scene maps, points with specific functions (such as guide points, interactive points, and rest points) are automatically identified and classified and labeled according to their function types.

[0030] Action / voice material matching and synthesis: The system automatically matches suitable materials from the material library based on location type and environmental characteristics, or creates exclusive materials that fit the scene through AI generation technology (such as voice synthesis).

[0031] Intelligent orchestration: Based on the spatial layout and logical relationships of points in the scene map, points and corresponding materials are automatically linked in chronological order to form a standardized workflow that can be executed by the robot.

[0032] Existing robot workflow orchestration technologies mainly adopt a "manual input + fixed template" model: point coordinates need to be manually marked on a map, and point types need to be manually categorized; action and voice materials need to be manually selected and matched. The existing technology has the following technical drawbacks: 1. Point recognition and classification rely on manual intervention, which is inefficient and error-prone, and cannot adapt to the multi-type point recognition needs of complex scenarios; 2. Action and voice materials need to be manually matched, lacking a scenario-based intelligent adaptation mechanism, resulting in low compatibility between materials and points / environments, affecting service experience; 3. The temporal orchestration does not incorporate map spatial layout, relying solely on manual setting of the sequence, which may cause robot path repetition and inconsistent execution logic; 4. The overall orchestration process requires significant manual intervention, and manual readjustment is necessary when adapting to new scenarios, resulting in high scenario migration costs. To improve the efficiency of path orchestration for scene robots, this application proposes a robot path orchestration method. For a clear description of the method provided in this embodiment, please refer to... Figure 1 This includes steps S110 to S130.

[0033] The method provided in this application is applied to a computer device associated with a robot, the device being equipped with a processor and memory. The processor is capable of executing the method provided in this application stored in the memory. The computer device may be, for example, a mobile terminal, a server, or a cloud computing platform. Specifically, the mobile terminal may take the form of, but is not limited to, mobile phones, tablets, personal digital assistants (PDAs), mobile internet devices (MIDs), and wearable devices (such as smartwatches). Applications (such as apps) on the mobile terminal can process and display relevant information and perform corresponding operations, thereby improving efficiency.

[0034] The communication technologies associated with robots can be wireless communication technologies, specifically including but not limited to: Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (W-CDMA), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Bluetooth, Wireless, Fidelity, WiFi (such as IEEE 802.11a, IEEE 802.11b, IEEE 802.11g and / or IEEE 802.11n standards), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging and short messages, and any other suitable communication protocols, even those protocols that have not yet been developed.

[0035] Step S110: Perform standardization processing on the acquired scene map to obtain map data, which consists of at least one structured point.

[0036] In one embodiment, the acquired scene map is subjected to standardization processing to obtain map data, including: performing a validity check on the scene map; when the validity check passes, extracting environmental features and completing coordinate calibration to generate a structured point list, the structured point list containing environmental features and at least one structured point; assigning a unique ID to the structured point list, binding it with the scene map as map data, and storing it.

[0037] In one embodiment, users can upload a map of the scene to be processed for subsequent automatic processing, providing basic data support for the generation of subsequent workflows. For example, users can upload a digital map corresponding to the target scene through the system's preset upload interface by dragging or clicking. The digital map is a true reconstruction of the target scene, containing core information such as scene spatial layout and various signs. This application does not impose specific restrictions on the acquisition method.

[0038] It is understandable that the acquired map data may be diverse and differ from one another, and requires standardized processing to ensure recognition accuracy and compatibility with subsequent steps. Therefore, the uploaded map data needs to be uniformly converted into a standard format, referred to as map data in this application.

[0039] The first step in the standardization process is to perform a validity check to ensure that the map meets the initial processing requirements.

[0040] In one embodiment, performing a validity check on the scene map includes: determining whether the scene map meets preset verification conditions, the verification conditions including at least one of the following: the scene map has built-in coordinate parameters, the coordinate parameters do not have missing anomalies, and the scene map resolution is greater than a preset threshold; when the scene map meets all verification conditions, the validity check is determined to be passed; when the scene map does not meet any one of the verification conditions, the validity check is determined to be failed.

[0041] In one implementation, as described above, the validity check mainly determines whether the scene map meets the check criteria. These criteria include at least one of the following: the scene map has built-in coordinate parameters; the coordinate parameters do not have missing anomalies; and the scene map resolution is greater than a preset threshold.

[0042] Specifically, the scene map contains built-in coordinate parameters, which are used to determine the integrity of the map data and ensure that subsequent coordinate calibration has a basis. These coordinate parameters include not only the map's latitude and longitude reference, but also key metadata such as projection method and coordinate system type, as well as auxiliary parameters that can be used to achieve precise spatial positioning, such as the scale. It can also determine whether these coordinate parameters conform to industry-standard coordinate references, ensuring that subsequent coordinate calibration work can be carried out normally.

[0043] The next step is to verify the validity of the map data, that is, whether the coordinate parameters are missing, abnormal, or contradictory. This can be achieved through multi-dimensional cross-validation to ensure logical consistency. If the coordinate information is missing or the file is damaged or incomplete, the verification conditions are not met.

[0044] Finally, this also includes checking whether the map resolution meets the minimum clarity requirements of the application scenario, i.e., whether the scene map resolution is greater than a preset threshold. This check ensures that the map has sufficient pixel accuracy in the subsequent point recognition and environmental feature extraction stages, guaranteeing that the processing can proceed.

[0045] The above three verification conditions can be combined arbitrarily, but at least one of them must be met. The map validity test is considered passed only if all verification conditions are met; if any condition is not met, the validity test is considered failed.

[0046] Furthermore, for maps that fail the inspection, the system can immediately provide users with specific reasons for the non-compliance and offer targeted correction guidance. It can also activate a self-repair function to determine if it has the capability for automatic coordinate completion or resolution enhancement. If so, it performs intelligent repair and secondary verification; otherwise, it guides users to re-upload or manually supplement key parameters. This design ensures that map data quality remains controllable and reliable, providing strong support for subsequent processing.

[0047] In one implementation, after the validity check passes, environmental features are extracted and coordinate calibration is completed to generate a structured list of points. Coordinate calibration can be initiated first, converting scene map data into a standardized format to eliminate data redundancy, correct coordinate deviations, and ensure the accuracy of subsequent point identification. Coordinate calibration can be based on an industry-standard coordinate system. Users can manually input latitude and longitude information to assist calibration according to the actual geographical location of the target scene. The system corrects deviations and performs unified conversion on the coordinate data of the original map, controlling the coordinate calibration error within a range that meets the requirements for accurate robot positioning. After calibration, a coordinate mapping lookup table is generated and stored in the storage unit for future reference and adjustment.

[0048] This allows for the identification of environmental features on the map, which are then used to generate a structured list of points. This list includes environmental features and at least one structured point. Environmental features refer to core features on the map such as walls, passageways, obstacles, and functional markers. Structured points are categorized into three types: operational points, auxiliary points, and key nodes. Each type of point is labeled with a unique ID, precise coordinates, semantic tags, and a functional description. Operational points correspond to specific operational tasks, such as cleaning, transporting, or inspection. Auxiliary points focus on parking, charging, or obstacle avoidance. Key points serve as pathway turning points, area boundaries, or system interaction hubs.

[0049] To differentiate the attribute differences of various features, irrelevant text, redundant lines, and other invalid information in the map can be removed before feature extraction. After extraction, various environmental features are classified and labeled, clearly defining attributes such as passable areas, impassable areas, and obstacle areas. The processed map data is then converted into a unified standardized format, including core content such as coordinate information, feature types, attribute labels, and text descriptions. The semantic features, attribute classifications, and spatial coordinates corresponding to all structured points are stored in a single form, resulting in a list of structured points. This list of structured points serves as the sole data source for subsequent intelligent recognition and material matching. Each point can be assigned a unique ID and stored after being bound to the scene map as map data. This operation ensures data consistency and traceability, guaranteeing that each point maintains a unified ID mapping across the map, recognition model, material library, and workflow. Before subsequent matching, the map data can be displayed to the user, clearly explaining the calibration parameters and feature extraction list. After user confirmation, the next implementation step can proceed.

[0050] Step S120: Based on the map data, match the corresponding scene material for each structured point.

[0051] In one embodiment, based on map data, matching corresponding scene materials for each structured point includes: determining the robot execution requirements matched by the map data; obtaining the corresponding material repository based on the robot execution requirements; and retrieving scene materials corresponding to each structured point from the material repository.

[0052] In one embodiment, existing technologies rely on manual experience or simple rule-based matching for point matching, lacking semantic understanding and dynamic adaptation capabilities. In this embodiment, however, semantically driven, precise matching is achieved through a dual understanding of map environment features and structured point semantic labels.

[0053] To address the technical challenges of matching materials using existing technologies, we can break it down into two key questions: What materials to match? And how to match them? The former focuses on the semantic alignment between material type and location function, while the latter emphasizes the automation and robustness of the matching process. To address these two technical challenges, this application proposes a material matching mechanism based on map data and structured locations.

[0054] Under this mechanism, the first step is to determine the robot execution requirements matched with the map data. These execution requirements define how the associated robot will display or process scene materials during its operation. For example, in an exhibition scenario, the robot needs multilingual narration, real-time interaction, and dynamic path adjustment capabilities; while in a warehouse scenario, the robot needs to emphasize precise handling, obstacle avoidance response, and status voice broadcasting. Therefore, determining the execution requirements requires considering the scene type, user role, and service objectives to accurately locate the necessary materials—the material repository. The material repository pre-sets standardized action sequences, voice prompts, visual feedback, and other multimodal scene materials according to scene type. All scene materials are indexed using both semantic tags and functional attributes, ensuring multidimensional semantic matching based on location classification (e.g., "guide points"), environmental keywords (e.g., "historical exhibits"), and execution requirements (e.g., "multilingual narration") during retrieval.

[0055] Once the material repository is determined, the semantic labels and environmental context features of each structured point can be extracted from the map data, and corresponding multimodal material combinations can be matched for each structured point.

[0056] In one embodiment, retrieving scene materials corresponding to each structured point from the material repository includes: determining whether a corresponding scene material exists for each structured point; if it exists, directly matching the corresponding scene material; if it does not exist, obtaining the environmental keywords of the structured point, inputting the environmental keywords into a preset generation model for processing, and obtaining the corresponding customized scene material; and matching the customized scene material with the structured point.

[0057] In one implementation, the matching method can dynamically evaluate the matching confidence between structured point labels and material feature vectors through semantic similarity calculation and context-aware algorithms. This allows for the matching of one or more scene materials with the highest confidence for each structured point, ensuring that the matching result conforms to both functional semantics and environmental context. For example, in a museum exhibition, the structured point might consist of multiple exhibition points, each with a corresponding exhibit as its semantic feature. The material repository pre-stores multilingual explanation scripts, high-definition close-up animations of cultural relics, and interactive question-and-answer audio for each exhibit. When the semantic feature of a structured point is identified as "Bronze Age - Shang and Zhou Dynasties," the corresponding script, animations, and question-and-answer audio are automatically matched.

[0058] In other words, the resource repository can categorize various resources according to their type, such as four main categories: voice resources, motion resources, visual resources, and interactive resources. Each category of resources is labeled with its applicable scenarios, triggering conditions, and collaborative logic. Overall, resources are the content used to control and demonstrate the robot. Essentially, they are the concrete expression of the robot's behavioral instructions, carrying the triple goals of functional semantics, interactive logic, and user experience.

[0059] The audio materials mainly cover four categories: multilingual narration scripts, environmental prompts, status announcements, and interactive responses. All are stored in a standardized, preset format and labeled with language, speaking speed, emotional tone (e.g., solemn, friendly, lively), and applicable location type. Their triggering logic is strongly coupled with the semantic tags of the structured locations—for example, locations labeled "Guide Point - Historical Exhibits" automatically load narration scripts with historical background descriptions, while locations labeled "Emergency Exit - Safety Exit" prioritize matching short, highly recognizable directional prompts.

[0060] Action-based materials mainly encompass three categories: posture adjustment commands, interactive action templates, and emergency obstacle avoidance behaviors. These are used to control the robot to perform corresponding tasks at different stages of its operation. For example, posture adjustment commands are typically used when the robot arrives at a structured location, fine-tuning its posture to its initial state to facilitate subsequent work. For instance, upon arrival at a location, it might adjust to the optimal angle for explanation and simultaneously activate eye LED lighting effects to enhance audience focus. Interactive action templates cover high-frequency guiding actions such as pointing, waving, and nodding. These are used for natural coordination and rhythmic guidance when displaying other scene materials at structured locations. For example, they support automatic semantic combination based on location—such as simultaneously executing pointing actions when presenting audio materials to guide the audience's attention to the core area of ​​the exhibit; and nodding actions to reinforce semantic emphasis at key information points. Emergency obstacle avoidance behaviors are automatically triggered when the robot encounters sudden obstacles during its movement. These actions may include emergency braking, audio-visual warnings, and path replanning to ensure the safety of personnel and equipment.

[0061] Visual materials focus on dynamic image expression, including but not limited to close-up motion effects of objects, ambient light and shadow mapping, spatial guidance animations, and AR overlay layers, all of which are precisely labeled according to point spatial coordinates, viewpoint parameters, and triggering timing. Furthermore, the labeling system for visual materials also incorporates a timeline alignment mechanism to ensure strict synchronization between the start and end frames of motion effects, the rhythm of light and shadow changes, the semantic units of voice narration, and the beat of action execution. This achieves millisecond-level collaborative alignment of multimodal materials in the time dimension, thereby constructing an immersive interactive experience integrating "sound, form, light, and motion."

[0062] Interactive materials encompass multimodal interaction protocols and context-aware responses, used to build two-way dialogue capabilities between the robot and the user. This can be understood as endowing the robot with a cognitive loop of "understanding—judgment—response," the core of which lies in the dynamic integration of user intent, environmental context, and historical interaction memory. For example, when a user stands at the "Shang and Zhou bronzes" point and asks, "Why are the eyes of the taotie pattern on this bronze artifact so large?", the robot can acquire and analyze this speech, retrieve information from its cultural relics knowledge base, generate corresponding scene materials, respond to the user's question, and enhance the user's experience.

[0063] As can be seen, the scene materials are diverse. During matching, for the same structured point, at least one of the four types of materials can be matched simultaneously, ultimately forming a complete response mode of "voice explanation + pointing action + close-up of cultural relic + interactive Q&A". Voice materials, action materials, and visual materials are the voice, action, and visual feedback output by the robot. Interactive materials are interactive devices that may exist at the structured point, such as touch screens, sensor buttons, or AR recognition areas. That is, controlling the robot requires establishing communication with interactive devices and triggering response logic to achieve a closed loop of "perception-response-interaction".

[0064] Furthermore, it is understandable that not all structured locations can be directly matched with pre-set materials. Some locations may not be able to match scene materials due to special functions or dynamic environmental changes. The embodiments described above are all for the case where structured locations can match scene materials. When they cannot be matched, adapted scene materials can be dynamically generated based on the generative model for supplementation and replacement. Correspondingly, it is understandable that, as mentioned above, structured locations are extracted through environmental features, including corresponding semantic features, i.e., environmental keywords. For example, keywords such as "Shang and Zhou bronzes," "interactive light and shadow area," and "multilingual guide" serve as input prompts for the generative model, driving the multimodal generative model to output matched customized scene materials. In addition, environmental keywords can not only be explicit labels directly represented by structured locations, but can also be combined with implicit context such as map spatial relationships, lighting conditions, and pedestrian density for enhanced modeling. Furthermore, users can also manually input the specific functional requirements of the location, using these requirements and environmental keywords as joint prompts for the generative model, thereby guiding the model to output more accurate and scene-adaptive customized scene materials. Understandably, customized scene assets and original scene assets are merely different names; they are identical in content. Therefore, after generation, customized scene assets can be verified to ensure they strictly align with the original asset library standards in terms of format, duration, and semantic consistency. Once verified, the customized scene assets are treated as scene assets and matched with structured points to ensure all structured points have corresponding adapted assets, thus guaranteeing the robot workflow can be executed completely in any scenario.

[0065] After the material matching or synthesis is completed, the materials are bound to the unique identifiers of the corresponding functional points through the preset association logic. Each point can be bound to several adapted materials, and users can manually replace materials and adjust the execution parameters of materials without having to manually perform association operations point by point, thus improving implementation efficiency.

[0066] In one embodiment, retrieving scene materials corresponding to each structured point from the material repository includes: determining the execution device type for each structured point; and configuring the output method of the scene materials based on the execution device type.

[0067] In one implementation, as mentioned above, the scene materials ultimately need to be executed by a robot. However, executing scene materials sometimes requires not only a robot but also corresponding equipment. For example, in a museum scene, when a structured point is labeled "AR Artifact Restoration Interactive Platform," this point not only needs to match the robot's action sequence and voice prompts but also needs to be linked with the AR interactive device's start command and screen rendering parameters. In this case, the AR interactive device is the execution device associated with the robot. Conversely, in a warehouse transportation scene, when a structured point is labeled "Intelligent Sorting Workstation," not only a robot but also a sorting robotic arm, a visual recognition module, and a conveyor control system are needed. These execution devices are all strongly associated with this point. Therefore, before generating the subsequent workflow, the execution mode of the scene materials and the required type of execution equipment can be predicted and bound, and the output method of the scene materials can be configured based on the type of execution equipment. This configuration and output method includes adapting communication protocols, command formats, trigger timing, and fault-tolerant retry mechanisms to ensure zero-error command interaction between the scene materials and the execution equipment. Meanwhile, the system automatically checks the online status and protocol compatibility of the executing device. If the device is offline or the protocol is incompatible, the abnormal point is marked in real time and a backup plan is triggered, such as starting the preset local cache material or switching to a lightweight interactive mode, to ensure that the workflow can still maintain basic navigation functions when the device is abnormal.

[0068] Step S130: Obtain the workflow based on the scene materials and structured points, and send the workflow to the robot.

[0069] In one implementation, this step is used to generate a standardized robot workflow file, which can be exported to the robot execution system to realize the implementation of the workflow. Specifically, the system automatically generates a standardized workflow file that conforms to industry standards based on the time-series orchestrated workflow data. The file contains complete execution information, including robot movement path coordinates, function point execution order, dwell time, and material association information, ensuring that the robot can accurately execute the workflow.

[0070] In one embodiment, a workflow is obtained by arranging scene materials and structured points, including: grouping scene materials to obtain multiple material groups; determining the priority of each structured point; constructing temporal dependencies based on material groups and priorities, and arranging them to obtain an initial workflow; and obtaining the final workflow after performing dynamic simulation verification on the initial workflow.

[0071] In one implementation, the workflow is the core instruction set controlling the robot's execution. Essentially, it connects all structured points in the adapted scene materials according to specific requirements. The shortcoming of existing technologies lies in their navigation-oriented approach, which simply connects points linearly to ensure the shortest route, neglecting the semantic meaning of each point and its collaborative relationship with the equipment. This problem is particularly prominent in exhibition hall scenarios. Therefore, this application introduces a dual constraint mechanism in the point-connection logic, combining semantic understanding of scene materials with spatial arrangement of structured points, to generate the workflow.

[0072] In one embodiment, this application first performs semantic analysis on the scene materials, such as extracting their core intent (e.g., "explanation," "interactive guidance," "emergency reminder") or emotional tendency (e.g., solemn, friendly, lively), and then groups the scene materials into multiple material groups. The grouping criteria can be arbitrarily set according to actual needs. For example, in an exhibition hall scene, material groups can be divided into three themes: "historical narrative," "technological interaction," and "humanistic care"; for a warehouse scene, material groups can be constructed according to "inbound scheduling," "in-warehouse inspection," and "outbound verification." This elevates the connection of points that were originally anchored by physical space to a semantic-spatial dual-dimensional collaborative mechanism.

[0073] Furthermore, the priority of each structured point can be determined. The specific definition of this priority can also be arbitrarily defined according to actual needs. For example, in an exhibition hall, the historical exhibit guide point has a higher priority than the rest point, while the interactive experience point is given a higher execution weight due to its high participation rate; in a warehousing scenario, the safety alarm point has a higher priority than the regular inspection point, and the outbound verification point is set as the highest priority because it involves the logistics closed loop.

[0074] Subsequently, based on the grouping and priority of materials, a temporal dependency relationship is constructed, ensuring that the workflow not only satisfies spatial accessibility but also guarantees both semantic coherence and task urgency. For example, when historical narrative materials are bound to high-priority guide points, they will be automatically executed first, avoiding interruption of the narrative logic by low-priority rest points. Simultaneously, during workflow orchestration, the semantic consistency between adjacent points can be dynamically verified. If a "Technology Interaction" group material is immediately followed by a "Historical Narrative" group material, a semantic buffering mechanism is triggered, automatically inserting transitional guiding voice and a soft-start action to ensure a natural and smooth narrative rhythm.

[0075] In addition, workflow orchestration must also consider the robot's physical accessibility. That is, it is necessary to combine the map layout to calculate the physical accessible path between points, read the pre-processed standardized map data, obtain the traversable area, obstacle distribution and coordinate information of all functional points, and calculate the optimal path from the start point to the end point of the operation through the preset path planning logic. This path can avoid obstacles, reduce unnecessary detours, and reserve emergency passages to deal with emergencies during the operation.

[0076] Simultaneously, following the principle of "critical points first, then auxiliary points; near points first, then far points," and combined with the user's preset task priorities, the execution order and dwell time for each functional point can be set. The dwell time can be customized according to the functional requirements of each point. The materials bound to each point are embedded into corresponding timing nodes, clearly defining the execution timing of the materials and ensuring that the robot can execute the corresponding materials in an orderly manner after arriving at the point. After the timing arrangement is completed, the system automatically simulates the complete execution process of the workflow, displaying the robot's movement path, point execution order, and material execution effects in real time. Users can intuitively view and manually adjust the timing parameters to ensure that the workflow meets actual operational needs.

[0077] Based on the above approach, each structured point and its corresponding scene materials are integrated into a standardized workflow file, covering path coordinates, execution sequence, material binding, and priority strategies. This file can be directly deployed to the robot. Specifically, a dedicated export interface is provided, allowing users to export workflow files to local storage or upload them directly to the robot execution system. The files are encrypted to prevent data leakage. Simultaneously, the system automatically backs up the generated workflow files, allowing users to retrieve, modify, and reuse historical workflow files at any time, reducing repetitive workload. After the workflow file is output, the system generates a detailed output report, clearly specifying the file format, compatible robot types, and deployment points. Users can import and deploy the workflow file without additional configuration of the robot execution system, ensuring efficient workflow implementation.

[0078] Correspondingly, this application also provides a robot, which includes a processor and a memory. The processor is used to execute a computer program stored in the memory to implement the workflow output by the robot path orchestration method described above.

[0079] This application provides a robot path orchestration method, including the following steps: performing standardization processing on the acquired scene map to obtain map data, the map data consisting of at least one structured point; matching corresponding scene materials to each structured point based on the map data; orchestrating a workflow based on the scene materials and structured points, and distributing the workflow to the robot. Therefore, this application can automatically analyze the scene map and accurately identify the functional attributes and spatial relationships of points to obtain map data. It also automatically adapts scene materials using the structured points in the map data, avoiding manual repetitive annotation and material matching, significantly shortening the deployment cycle. Simultaneously, planning the robot route based on points and materials ensures that the route combines task efficiency and execution robustness, avoiding problems such as path repetition and inconsistent execution logic caused by manual intervention alone, truly achieving an end-to-end automated closed loop from map parsing to workflow implementation.

[0080] Figure 2An internal structural diagram of a computer device in one embodiment is shown. This computer device can specifically be a robot or an execution terminal for the method of this application. Figure 2 As shown, the device includes: a processor 310 and a memory 311 storing a computer program; wherein, Figure 2 The processor 310 shown in the diagram does not indicate that there is only one processor 310, but only indicates the positional relationship of the processor 310 relative to other devices. In practical applications, there can be one or more processors 310; similarly, Figure 2 The memory 311 illustrated herein has the same meaning, that is, it is only used to indicate the positional relationship of memory 311 relative to other devices. In practical applications, there can be one or more memories 311. When the processor 310 runs the computer program, the method applied to the above-mentioned device is implemented.

[0081] The device may also include at least one network interface 312. The various components of the device are coupled together via a bus system 313. It is understood that the bus system 313 is used to implement communication between these components. In addition to a data bus, the bus system 313 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 2 The general designated all buses as Bus System 313.

[0082] The memory 311 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 311 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0083] The memory 311 in this embodiment of the invention is used to store various types of data to support the operation of the device. Examples of this data include: any computer programs used to operate on the device, such as operating systems and applications; contact data; phonebook data; messages; pictures; videos, etc. The operating system includes various system programs, such as the framework layer, core library layer, driver layer, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications, such as media players, browsers, etc., used to implement various application services. Here, the program implementing the method of this embodiment of the invention can be included in the application.

[0084] Based on the same inventive concept as the foregoing embodiments, this embodiment also provides a computer-readable storage medium storing a computer program. The computer-readable storage medium can be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; it can also be various devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc. When the computer program stored in the computer-readable storage medium is run by a processor, it implements the above method. For the specific steps implemented when the computer program is executed by the processor, please refer to [link to relevant documentation]. Figure 1 The description of the illustrated embodiments will not be repeated here.

[0085] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0086] In this document, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, which includes not only the elements listed but also other elements not expressly listed.

[0087] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A robot path arrangement method, characterized in that, Includes the following steps: The acquired scene map is standardized to obtain map data, which consists of at least one structured point. Based on the map data, match corresponding scene materials for each structured point; The workflow is generated based on the scene materials and the structured points, and then the workflow is sent to the robot.

2. The robot path arrangement method as described in claim 1, characterized in that, The standardization process performed on the acquired scene map to obtain map data includes: Perform a validity check on the scene map; Once the validity check is passed, environmental features are extracted and coordinate calibration is completed to generate a structured point list, which includes environmental features and at least one structured point. The structured point list is assigned a unique ID and bound to the scene map as map data and stored.

3. The robot path arrangement method as described in claim 2, characterized in that, The validity check performed on the scene map includes: Determine whether the scene map meets the preset verification conditions. The verification conditions include at least one of the following: the scene map has built-in coordinate parameters, the coordinate parameters do not have missing anomalies, and the scene map resolution is greater than a preset threshold. When the scene map meets all the verification conditions, the validity check is deemed to have passed. If the scene map does not meet any of the verification conditions, the validity check is deemed to have failed.

4. The robot path arrangement method as described in claim 1, characterized in that, The step of matching corresponding scene materials for each structured point based on the map data includes: Determine the robot execution requirements that match the map data; Obtain the corresponding material repository based on the robot's execution requirements; Retrieve the scene material corresponding to each of the structured points from the material repository.

5. The robot path arrangement method as described in claim 4, characterized in that, The step of retrieving the scene material corresponding to each of the structured points from the material repository includes: Determine whether the structured point has a corresponding scene material; If it exists, then directly match the corresponding scene material; If it does not exist, then obtain the environmental keywords of the structured point, input the environmental keywords into the preset generation model for processing, and obtain the corresponding customized scene material; match the customized scene material with the structured point.

6. The robot path arrangement method as described in claim 4, characterized in that, The step of retrieving the scene material corresponding to each of the structured points from the material repository includes: Determine the execution device type for each of the structured points; Configure the output method of the scene material based on the type of execution device.

7. The robot path arrangement method as described in claim 1, characterized in that, The workflow obtained by arranging the scene materials and structured points includes: The scene materials are grouped to obtain multiple material groups; Determine the priority of each of the structured points; Based on the material group and the priority, a temporal dependency relationship is constructed, and an initial workflow is orchestrated. The workflow is obtained after dynamic simulation verification of the initial workflow.

8. A robot, characterized in that, Including processor and memory; The processor is used to execute the computer program stored in the memory to implement the workflow output by the robot path orchestration method as described in any one of claims 1 to 7.

9. A computer device, characterized in that, Including processor and memory; The processor is used to execute a computer program stored in the memory to implement the robot path orchestration method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images