Factory patrol and disaster relief method based on humanoid robot and related products

Abstract

The present application discloses a factory patrol and disaster relief method based on a humanoid robot and related products, the method comprising: obtaining target map data and equipment distribution data corresponding to a target factory; obtaining an initial position corresponding to a humanoid robot; generating a target patrol path based on the target map data and the initial position; controlling the humanoid robot to patrol based on the target patrol path, and performing a disaster inspection on the target factory during the patrol process to obtain target disaster data; determining n disaster relief equipment and m factory equipment within a preset distance around the humanoid robot based on the equipment distribution data; generating a target disaster relief plan based on the target disaster data, n disaster relief equipment and m factory equipment; and controlling the humanoid robot to perform disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory. The embodiment of the present application improves the safety of the factory.

Description

Technical Field

[0001] The present application relates to the field of robotics technology, and in particular to a factory patrol and disaster relief method based on a humanoid robot and related products. Background Art

[0002] In high-risk industries such as petrochemicals, production plants are generally unmanned or have fewer staff. Factory safety is mainly monitored through video surveillance systems. However, video surveillance systems are complex to install and have high maintenance costs. Once a disaster occurs, they cannot be handled immediately and are not very safe.

[0003] Therefore, how to improve the safety of the factory has become an urgent problem to be solved. Summary of the invention

[0004] The embodiments of the present application provide a factory patrol and disaster relief method based on a humanoid robot and related products. By controlling the humanoid robot to patrol the factory, it is possible to promptly discover disasters and automatically generate disaster relief plans to rescue the factory, thereby improving the safety of the factory.

[0005] In a first aspect, an embodiment of the present application provides a factory patrol and disaster relief method based on a humanoid robot, the method comprising:

[0006] Obtain target map data and equipment distribution data corresponding to the target factory;

[0007] Get the initial position corresponding to the humanoid robot;

[0008] Generate a patrol path based on the target map data and the initial position to obtain a target patrol path;

[0009] Controlling the humanoid robot to patrol based on the target patrol path, and inspecting the disaster situation of the target factory during the patrol process to obtain target disaster situation data;

[0010] Determine n disaster relief equipment and m factory equipment existing within a preset distance around the humanoid robot based on the equipment distribution data; m and n are both natural numbers;

[0011] Generate a disaster relief plan according to the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan;

[0012] The humanoid robot is controlled to perform disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory.

[0013] In a second aspect, an embodiment of the present application provides a factory patrol and disaster relief device based on a humanoid robot, the device comprising: an acquisition unit, a control unit, and a disaster relief unit, wherein:

[0014] The acquisition unit is used to acquire target map data and equipment distribution data corresponding to the target factory; and acquire an initial position corresponding to the humanoid robot;

[0015] The control unit is used to generate a patrol path based on the target map data and the initial position to obtain a target patrol path; control the humanoid robot to patrol based on the target patrol path, and perform a disaster inspection on the target factory during the patrol to obtain target disaster data; determine n disaster relief equipment and m factory equipment within a preset distance around the humanoid robot based on the equipment distribution data; m and n are both natural numbers; generate a disaster relief plan based on the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan;

[0016] The disaster relief unit is used to control the humanoid robot to carry out disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory.

[0017] In a third aspect, an embodiment of the present application provides a humanoid robot, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps of any method of the first aspect of the embodiment of the present application.

[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in any method of the first aspect of the embodiment of the present application.

[0019] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in any method of the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

[0020] The implementation of this application has the following beneficial effects:

[0021] It can be seen that the factory patrol and disaster relief method based on humanoid robots described in the present application includes: obtaining target map data and equipment distribution data corresponding to the target factory; obtaining the initial position corresponding to the humanoid robot; generating a patrol path based on the target map data and the initial position to obtain the target patrol path; controlling the humanoid robot to patrol based on the target patrol path, and conducting a disaster inspection on the target factory during the patrol to obtain target disaster data; determining the n disaster relief equipment and m factory equipment existing in a preset distance around the humanoid robot based on the equipment distribution data; generating a disaster relief plan based on the target disaster data, n disaster relief equipment and m factory equipment to obtain a target disaster relief plan; controlling the humanoid robot to perform disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory. On the one hand, the patrol path generated according to the map data and the initial position of the robot can ensure full coverage of the factory area, reduce monitoring blind spots, and improve the comprehensiveness of detection. On the other hand, by patrolling the factory along a predetermined path, the humanoid robot can promptly discover the disaster and automatically generate a disaster relief plan to rescue the factory, thereby improving the safety of the factory. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the background technology, the drawings required for use in the embodiments of the present application or the background technology will be described below.

[0023] Figure 1 This is a functional module composition block diagram of a humanoid robot provided in an embodiment of the present application;

[0024] Figure 2 is a flow chart of a factory patrol and disaster relief method based on a humanoid robot provided in an embodiment of the present application;

[0025] Figure 3 is a structural schematic diagram of a target factory provided in an embodiment of the present application;

[0026] Figure 4 is a schematic diagram of generating a patrol path provided in an embodiment of the present application;

[0027] Figure 5 This is a functional module composition block diagram of a factory patrol and disaster relief device based on a humanoid robot provided in an embodiment of the present application;

[0028] Figure 6 It is a structural schematic diagram of a humanoid robot provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0030] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0031] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0032] See also Figure 1 , Figure 1 is a functional module block diagram of a humanoid robot provided in an embodiment of the present application, such as Figure 1 As shown, the functional modules of the humanoid robot may include: a control module, a camera module, an environmental data acquisition module, a signal transceiver module, etc., which are not limited here.

[0033] The control module, as the "brain" of the humanoid robot, is responsible for coordinating the work of other modules. It processes data from various modules, makes decisions based on preset programs and algorithms, and directs the actions, behaviors, and task execution of the humanoid robot. For example, it determines the walking path and posture of the humanoid robot, controls the shooting angle and frequency of the camera module, and processes the data obtained by the environmental data acquisition module.

[0034] The camera module is equivalent to the "eyes" of the humanoid robot, which is used to obtain images and video information of the surrounding environment. This visual information can help the humanoid robot identify objects, people, obstacles, etc., and provide a basis for the robot's navigation, obstacle avoidance, target tracking, and interaction with the environment. For example, during a patrol mission, the camera module can capture scenes inside the factory to help detect abnormal situations.

[0035] Among them, the environmental data acquisition module is like the "sensory organ" of the humanoid robot, responsible for collecting various environmental parameters, such as temperature, humidity, air pressure, air quality, sound, etc. These data are very important for evaluating environmental conditions and detecting potential dangers or anomalies. For example, in a factory environment, it can detect excessive temperatures or leaks of harmful gases.

[0036] The signal transceiver module is the interface for the humanoid robot to communicate with external control sources or other devices. It can receive instructions, data and information from the remote control terminal, enabling the humanoid robot to perform tasks according to external control requirements or work in coordination with other devices. For example, it can receive the latest patrol instructions or updated environmental threshold settings sent by the factory control center.

[0037] Through the mutual collaboration between these modules, humanoid robots can perceive the environment, receive instructions, make decisions and perform corresponding actions to complete various complex tasks.

[0038] See also Figure 2 , Figure 2 : is a flow chart of a factory patrol and disaster relief method based on a humanoid robot provided in an embodiment of the present application, the method includes but is not limited to the following steps:

[0039] S201. Obtain target map data and equipment distribution data corresponding to a target factory.

[0040] In the embodiment of the present application, the target factory includes but is not limited to unmanned factories and ordinary factories; the target map data may include at least one of the following: workshop distribution data, channel and road data, terrain and topography data, obstacle data, etc., which are not limited here; the equipment distribution data may include at least one of the following: equipment name data, equipment type data, equipment location data, equipment function data, equipment specification data, etc., which are not limited here.

[0041] In a specific embodiment, the control module or signal transceiver module of the humanoid robot can be communicatively connected with the control system of the target factory, and the control module is authorized by the control system, and the control module accesses the database of the control system to obtain the target map data and equipment distribution data. Alternatively, the target map data and equipment distribution data can be manually transmitted to the control module by the staff of the target factory.

[0042] It should be explained that the control module of the humanoid robot can also periodically or intermittently update the target map data and equipment distribution data to ensure that the data stored in the control module are the latest versions. This enables the humanoid robot to better keep pace with the ever-changing factory environment, give full play to its functions, and provide more reliable support for factory production and safety.

[0043] S202: Obtain an initial position corresponding to the humanoid robot.

[0044] In the embodiment of the present application, the initial position is the position of the humanoid robot when it is just started.

[0045] In a specific embodiment, a positioning module may be built into the humanoid robot, and the humanoid robot is positioned by the positioning module to obtain an initial position.

[0046] S203: Generate a patrol path based on the target map data and the initial position to obtain a target patrol path.

[0047] In the embodiment of the present application, the initial position can be used as the starting point of the target patrol path, and then the corresponding patrol path can be generated according to the target map data to obtain the target patrol path.

[0048] Optionally, step S203, generating a patrol path based on the target map data and the initial position to obtain a target patrol path, may include the following steps:

[0049] A1. Acquire a position with a wide field of view in the target map data to obtain a plurality of positions with a wide field of view;

[0050] A2. Generate a reference patrol path according to the initial position and the multiple positions with wide field of view to obtain multiple reference patrol paths;

[0051] A3. Determine the path characteristic data corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of path characteristic data; the path characteristic data includes any one of the following: the length of a straight path, the number of curved paths, the height of a lifting path, and the inclination of a climbing path;

[0052] A4. Determine a path distance corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of path distances;

[0053] A5. Determine a difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances and the plurality of path feature data to obtain a plurality of difficulty values;

[0054] A6. Obtain a minimum difficulty value among the multiple difficulty values, and determine that a reference patrol path corresponding to the minimum difficulty value among the multiple reference patrol paths is the target patrol path.

[0055] In the embodiment of the present application, the difficulty value represents the difficulty level of the humanoid robot when patrolling along a certain patrol path. The difficulty value may range from 0 to 100, and the larger the value, the higher the difficulty.

[0056] In a specific embodiment, the position with a wide field of view in the target map data can be first obtained to obtain multiple positions with a wide field of view. Specifically, the terrain data in the target map data can be analyzed to determine the high terrain (for example, high ground, top of a slope). The high terrain usually has a wider field of view, and the high terrain is used as the position with a wide field of view. The flat area in the target factory can also be identified, and the middle position of the flat area can be selected as the position with a wide field of view. In this way, multiple positions with a wide field of view can be obtained. Of course, personnel familiar with the target factory can also directly mark the positions they think have a wide field of view in the target map data based on their experience, thereby obtaining multiple positions with a wide field of view.

[0057] A reference patrol path is generated according to an initial position and multiple positions with a wide field of view, and multiple reference patrol paths are obtained. Specifically, a suitable path planning algorithm (for example, A-star algorithm, Dijkstra algorithm) can be selected based on target map data to perform path planning. The initial position can be set as the starting point, and a position with a wide field of view is selected from multiple positions with a wide field of view as the end point, and other positions with a wide field of view that are not selected as the end points are used as nodes in the patrol path. These conditions are input into the path planning algorithm for multiple path planning, and multiple reference patrol paths can be obtained. Then, path feature data corresponding to each reference patrol path in the multiple reference patrol paths can be determined to obtain multiple path feature data. Specifically, each reference patrol path can be firstly set as the starting point, and the path feature data corresponding to each reference patrol path can be obtained. The logical path is decomposed into smaller section units to obtain multiple section units so that different types of sections can be analyzed separately. For example, the path between each two adjacent wide-view positions in the multiple wide-view positions can be divided into a section unit; then, for each section unit in the multiple section units, it is determined whether it is a straight path. If so, its length is measured and recorded to obtain the length of the straight path in the multiple path feature data. Similarly, for each section unit in the multiple section units, the height information of different points on the section unit can be compared to determine the sections with ups and downs, and the height difference of the ups and downs path is estimated and recorded to obtain the height of the ups and downs path in the multiple path feature data. By analogy, multiple path feature data can be obtained.

[0058] For easier understanding, see Figure 3 , Figure 3is a schematic diagram of the structure of a target factory provided in an embodiment of the present application, such as Figure 3 As shown, the target factory includes but is not limited to: production workshop A, production workshop B, production workshop C, production workshop D, production workshop E and other workshops.

[0059] Among them, S is the initial position of the humanoid robot, L1, L2, L3, and L4 are all positions with wide fields of view. The humanoid robot starts patrolling from the initial position S and inspects the target factory. Every time it passes a position with a wide field of view, the humanoid robot can stay for a period of time and use the camera module to conduct disaster inspections on the surrounding production workshops. In this way, the patrol efficiency can be improved and disasters can be discovered in time. Once a disaster is discovered, the humanoid robot will execute the disaster relief plan to provide disaster relief to the production workshop, thereby improving the safety of the target factory.

[0060] Among them, the humanoid robot needs to pass through a curved path to move from the wide-view position L3 to the wide-view position L1; there is a slope section on the right side of the production workshop B, and so on. The roads or passages inside the target factory may also include other types of sections, which are not limited here.

[0061] For easier understanding, see Figure 4 , Figure 4 is a schematic diagram of generating a patrol path provided in an embodiment of the present application, such as Figure 4 As shown, a patrol path is generated based on the initial position S and the wide-view positions L1, L2, L3, and L4, where the initial position S is the starting point and the wide-view position L2 is the end point. Then, the patrol paths finally generated can be the following three: the first patrol path is: ①→②→⑤; the second patrol path is: ①→③→⑥; the third patrol path is: ①→④→⑥; then, the humanoid robot can select one of the three patrol paths as the target patrol path, patrol along the target patrol path, and protect the safety of the target factory.

[0062] Furthermore, the path distance corresponding to each reference patrol path in the multiple reference patrol paths can be determined to obtain multiple path distances. Taking the reference patrol path L1 in the multiple reference patrol paths as an example, the nodes passed by the reference patrol path L1 (i.e., positions with open fields of view) can be marked in the target map data to obtain multiple nodes, and the coordinate information of the multiple nodes can be obtained to obtain multiple coordinates. These coordinates can be based on a specific coordinate system of the map (e.g., a Cartesian coordinate system). Then, a suitable distance calculation method (e.g., Euclidean distance) can be selected according to the characteristics and accuracy requirements of the target map data. The path distance of the reference patrol path L1 can be calculated according to the multiple coordinates using the distance calculation method. By performing such calculations multiple times, multiple path distances can be obtained.

[0063] Next, the difficulty value corresponding to each reference patrol path in multiple reference patrol paths can be determined based on multiple path distances and multiple path feature data to obtain multiple difficulty values; finally, the minimum difficulty value among the multiple difficulty values ​​can be found, and the reference patrol path corresponding to the minimum difficulty value in the multiple reference patrol paths can be determined to obtain the reference patrol path with the minimum difficulty, and this reference patrol path with the minimum difficulty can be used as the target patrol path to reduce the patrol difficulty of the humanoid robot.

[0064] In this way, multiple positions with open fields of view are obtained by acquiring positions with open fields of view in target map data; multiple reference patrol paths are generated according to the initial position and the multiple positions with open fields of view; multiple path feature data corresponding to the multiple reference patrol paths are determined; multiple path distances corresponding to the multiple reference patrol paths are determined; multiple difficulty values ​​corresponding to the multiple reference patrol paths are determined according to the multiple path distances and the multiple path feature data; the minimum difficulty value among the multiple difficulty values ​​is obtained, and the reference patrol path corresponding to the minimum difficulty value among the multiple reference patrol paths is determined as the target patrol path. On the one hand, by selecting the target patrol path with the minimum difficulty value, the humanoid robot can complete the patrol task more quickly, reduce the time and energy spent on complex paths, and improve the patrol efficiency. On the other hand, a path with less difficulty usually means that the humanoid robot is less likely to encounter failures, collisions or other unexpected situations, thereby reducing the cost of repair and replacement of parts.

[0065] Optionally, step A5, determining the difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances and the plurality of path feature data to obtain a plurality of difficulty values, may include the following steps:

[0066] B1. determining a reference difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances to obtain a plurality of reference difficulty values;

[0067] B2. Determine the estimated power consumption corresponding to each path characteristic data in the plurality of path characteristic data to obtain a plurality of estimated power consumptions;

[0068] B3. Determine a first influence coefficient corresponding to each of the plurality of estimated power consumptions to obtain a plurality of first influence coefficients;

[0069] B4. Obtaining the environmental parameters corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of environmental parameters;

[0070] B5. Determine a second influence coefficient corresponding to each of the multiple environmental parameters to obtain multiple second influence coefficients;

[0071] B6. Determine the multiple difficulty values ​​according to the multiple first influence coefficients, the multiple second influence coefficients and the multiple reference difficulty values.

[0072] In the embodiment of the present application, the environmental parameters may include at least one of the following: ambient humidity, ambient temperature, light intensity, haze pollution level, etc., which are not limited here.

[0073] In a specific embodiment, a reference difficulty value corresponding to each reference patrol path in a plurality of reference patrol paths can be determined according to a plurality of path distances to obtain a plurality of reference difficulty values. Specifically, a mapping relationship between a preset path distance and a reference difficulty value can be pre-stored, and a plurality of reference difficulty values ​​corresponding to a plurality of path distances can be determined based on the mapping relationship, and each reference difficulty value corresponds to a path distance. Then, an estimated power consumption corresponding to each path characteristic data in a plurality of path characteristic data can be determined to obtain a plurality of estimated power consumptions. Specifically, a basic power consumption of a humanoid robot walking on a straight path can be obtained according to the hardware specifications, performance parameters and previous power consumption data of the humanoid robot. Then, for each type of path characteristic data, its influence on the basic power consumption is analyzed respectively to obtain a plurality of power consumption influencing parameters. Specifically, a mapping relationship between a preset path characteristic data and a power consumption influencing parameter can be pre-stored, and a plurality of power consumption influencing parameters corresponding to the plurality of path characteristic data can be determined based on the mapping relationship. The value range of the power consumption influencing parameter is -0.2~0.2. Multiple estimated power consumptions can be obtained by estimating based on the plurality of power consumption influencing parameters and the basic power consumption. The specific calculation formula is as follows:

[0074] Estimated power consumption a = basic power consumption × (1 + power consumption influencing parameter a);

[0075] Among them, the power consumption influencing parameter a is any one of multiple power consumption influencing parameters; the estimated power consumption a is the estimated power consumption corresponding to the power consumption influencing parameter a in multiple estimated power consumptions; by calculating multiple times through the above formula, multiple estimated power consumptions can be obtained.

[0076] Further, a first influence coefficient corresponding to each of the multiple estimated power consumptions can be determined to obtain multiple first influence coefficients. Specifically, a mapping relationship between a preset estimated power consumption and the first influence coefficient can be pre-stored, and multiple first influence coefficients corresponding to the multiple estimated power consumptions can be determined based on the mapping relationship. The value range of the first influence coefficient can be -0.3~0.3; then, the environmental parameter corresponding to each of the multiple reference patrol paths can be obtained to obtain multiple environmental parameters. Specifically, the environmental parameter can be environmental humidity. Multiple humidity sensors can be set on the multiple reference patrol paths. The humidity is detected by these multiple humidity sensors to obtain multiple environmental humidity, that is, multiple environmental parameters; then, a second influence coefficient corresponding to each of the multiple environmental parameters can be determined to obtain multiple second influence coefficients. Specifically, a mapping relationship between a preset environmental parameter and the second influence coefficient can be pre-stored, and multiple second influence coefficients corresponding to the multiple environmental parameters can be determined based on the mapping relationship. The value range of the second influence coefficient can be -0.12~0.12; finally, multiple difficulty values ​​can be determined based on multiple first influence coefficients, multiple second influence coefficients and multiple reference difficulty values. The specific calculation formula is as follows:

[0077] Difficulty value b = reference difficulty value b × (1 + first impact coefficient b) × (1 + second impact coefficient b);

[0078] Among them, the reference difficulty value b is any one of multiple reference difficulty values; the difficulty value b is the difficulty value corresponding to the reference difficulty value b in the multiple difficulty values; multiple difficulty values ​​can be obtained by calculating multiple times through the above formula.

[0079] In this way, by determining multiple reference difficulty values ​​corresponding to multiple reference patrol paths according to multiple path distances; determining multiple estimated power consumptions corresponding to multiple path characteristic data; determining multiple first influence coefficients corresponding to multiple estimated power consumptions; obtaining multiple environmental parameters corresponding to multiple reference patrol paths; determining multiple second influence coefficients corresponding to multiple environmental parameters; and determining multiple difficulty values ​​according to multiple first influence coefficients, multiple second influence coefficients and multiple reference difficulty values, on the one hand, multiple aspects such as path distance, path characteristics, power consumption, and environmental parameters are taken into consideration, and a comprehensive and detailed evaluation of the patrol path can be carried out, avoiding the one-sidedness caused by a single factor dominating the decision. On the other hand, when environmental parameters or other factors change, the difficulty value can be re-evaluated and adjusted in time, and the patrol path can be re-planned, thereby increasing the flexibility and adaptability of the humanoid robot.

[0080] S204: Control the humanoid robot to patrol based on the target patrol path, and perform a disaster inspection on the target factory during the patrol to obtain target disaster data.

[0081] In the embodiment of the present application, the target disaster data may include at least one of the following: disaster type, disaster location, disaster scale, etc., which are not limited here.

[0082] In a specific embodiment, the control module can control the humanoid robot to patrol based on the target patrol path. Every time it reaches a position with a wide field of view, it conducts a disaster inspection on the target factory until the entire target patrol path is completed to obtain the target disaster data.

[0083] Optionally, in step S204, the humanoid robot includes a camera module and an environmental data acquisition module, and the step of inspecting the target factory during the patrol to obtain target disaster data may include the following steps:

[0084] C1. When the humanoid robot reaches a first wide-view position, determine the production workshops that the humanoid robot can observe at the first wide-view position, and obtain i production workshops; the first wide-view position is any wide-view position included in the target patrol path; i is a natural number;

[0085] C2. Collecting image data of the i production workshops through the camera module to obtain i image data;

[0086] C3, performing disaster identification on the i image data to obtain i identification results; the identification results include possible existence of a disaster or no disaster;

[0087] C4, determining j identification results among the i identification results that may indicate a disaster situation; j is a natural number less than or equal to i;

[0088] C5. Determine the j production workshops corresponding to the j recognition results in the i production workshops;

[0089] C6. Obtain the distance between each of the j production workshops and the humanoid robot to obtain j distances;

[0090] C7. Sort the j production workshops according to the j distances to obtain a target viewing order; the shorter the distance, the higher the viewing order;

[0091] C8. Control the humanoid robot to enter the j production workshops in sequence according to the target viewing order, collect image data of the j production workshops through the camera module to obtain j image data, and collect environmental data of the j production workshops through the environmental data collection module to obtain j environmental data;

[0092] C9. Determine first disaster data corresponding to the first position with a wide field of view based on the j pieces of environmental data and the j pieces of image data;

[0093] C10, determining disaster data corresponding to all positions with a wide field of view except the first position with a wide field of view in the target patrol path, and obtaining at least one disaster data;

[0094] C11. Determine the target disaster data based on the at least one disaster data and the first disaster data.

[0095] In the embodiment of the present application, the production workshop may include at least one of the following: an assembly workshop, a processing workshop, a chemical workshop, etc., which are not limited here.

[0096] In a specific embodiment, when the humanoid robot reaches the first position with a wide field of view, the current position of the humanoid robot is measured by the positioning module, and the production workshops that the humanoid robot can observe at the current position can be determined according to the target map data, and i production workshops are obtained; then, the image data of the i production workshops can be collected by the camera module to obtain i image data. Specifically, the workshop positions of the i production workshops can be obtained from the target map data first to obtain i workshop positions, and i shooting directions can be obtained according to the i workshop positions and the current position. The humanoid robot is controlled to shoot in the i shooting directions respectively to obtain i image data; then, the i image data can be used to perform disaster recognition on the i image data through a preset disaster recognition model to obtain i recognition results. For example, the preset disaster recognition model can be an object classification model (i.e., a MobileNet model). The MobileNet model is a lightweight convolutional neural network model suitable for preliminary recognition of disasters (e.g., fires, floods, etc.) on devices with limited computing resources.

[0097] Furthermore, it is possible to determine j identification results among the i identification results that may have disaster situations; then, the j production workshops corresponding to these j identification results can be found from the above i production workshops; then, the distance between each of the j production workshops and the humanoid robot is obtained to obtain j distances. Specifically, the positions of the j production workshops can be first obtained from the i workshop positions to obtain j positions, and the distances between these j positions and the above current positions are calculated to obtain j distances; then, the j production workshops can be sorted according to the j distances to obtain the target viewing order; the shorter the distance, the higher the viewing order.

[0098] Then, the humanoid robot can be controlled to enter j production workshops in sequence according to the target viewing order, and the image data of the j production workshops can be collected through the camera module to obtain j image data. Taking production workshop a as an example, after the humanoid robot enters production workshop a, it is first necessary to clarify the key areas that need to be monitored, the specific scenes or events to be captured, etc. For example, production workshop a contains 3 production equipment and several production lines. Then, the humanoid robot can take these 3 production equipment as key areas, collect images of them, and obtain image data a. In this way, j image data can be obtained. In addition, the environmental data of the j production workshops can be collected through the environmental data collection module to obtain j environmental data. Specifically, at least one type of environmental data to be collected can be determined first, such as temperature, humidity, air quality (including various pollutant concentrations), light intensity, noise level, etc. Then, the environmental data of the j production workshops is collected based on the at least one environmental data type through the environmental data collection module to obtain j environmental data.

[0099] Then, the first disaster data corresponding to the first wide-view position can be determined based on the j environmental data and the j image data; then, the disaster data corresponding to all wide-view positions except the first wide-view position in the target patrol path can be determined to obtain at least one disaster data. Specifically, all wide-view positions included in the target patrol path can be obtained first to obtain at least one wide-view position, and then, in the at least one wide-view position, all wide-view positions except the first wide-view position can be inspected for disasters to obtain at least one disaster data. It should be explained that the method for obtaining the at least one disaster data can be the same as the method for obtaining the first disaster data; finally, the target disaster data can be determined based on the at least one disaster data and the first disaster data. Specifically, the at least one disaster data and the first disaster data can be integrated to form a comprehensive disaster data. The disaster data can also be classified and labeled, such as labeled according to standards such as disaster type and impact degree to obtain the final target disaster data.

[0100] In this way, when the humanoid robot reaches the first position with a wide field of view, j environmental data and j image data corresponding to the first position with a wide field of view are obtained, and the first disaster data corresponding to the first position with a wide field of view is determined based on the j environmental data and the j image data; the disaster data corresponding to all positions with a wide field of view except the first position with a wide field of view in the target patrol path are determined to obtain at least one disaster data; the target disaster data is determined based on the at least one disaster data and the first disaster data. On the one hand, a detailed inspection can be carried out on the production workshops where disasters may exist in a targeted manner, thereby improving the accuracy and timeliness of disaster discovery and facilitating the rapid adoption of response measures; on the other hand, key workshops are first selected through preliminary observations at the position with a wide field of view, and then in-depth inspections are carried out in turn according to the distance, thereby avoiding blind inspections, saving the energy and working time of the humanoid robot, and improving work efficiency.

[0101] Optionally, step C9, determining the first disaster data corresponding to the first position with a wide field of view based on the j pieces of environmental data and the j pieces of image data, may include the following steps:

[0102] D1. Acquire first environmental data and first image data; the first environmental data includes first temperature data and first gas composition data; the first environmental data is any environmental data among the j environmental data; the first image data is image data corresponding to the first environmental data among the j image data;

[0103] D2. Identify the first image data to determine a fire identification result of the first production workshop corresponding to the first image data; the fire identification result includes whether a fire event has occurred or not;

[0104] D3. when the fire identification result includes the fire event, obtaining a maximum temperature value and a minimum temperature value in the first temperature data;

[0105] D4. When the maximum temperature value is greater than a preset temperature value, determining a temperature difference between the maximum temperature value and the minimum temperature value to obtain a target difference;

[0106] D5. Determine the target fire parameter corresponding to the target difference;

[0107] D6. Analyze the first gas composition data to determine the proportion of harmful gas components in the first gas composition data to obtain a first proportion;

[0108] D7. When the first ratio is greater than a preset ratio, it is determined that a harmful gas leakage event has occurred in the first production workshop;

[0109] D8. Determine disaster data corresponding to the first environmental data and the first image data based on the fire identification result, the target fire parameter, the first ratio and the harmful gas leakage event.

[0110] In the embodiment of the present application, the preset temperature value and the preset ratio can be preset or defaulted in advance; the target fire parameter can be at least one of the following: smoke concentration, fire size, heat release rate, etc., which are not limited here.

[0111] In a specific embodiment, the first environmental data and the first image data can be first obtained from the above-mentioned j environmental data and j image data; then, the first image data can be identified to determine the fire identification result of the first production workshop corresponding to the first image data. Specifically, the first image data can be identified through the above-mentioned preset disaster identification model to obtain the fire identification result; when the fire identification result is a fire event, the maximum temperature value and the minimum temperature value in the first temperature data can be obtained. When the maximum temperature value is less than or equal to the preset temperature value, it can be determined that a small fire has occurred in the first production workshop and the fire is not large.

[0112] When the maximum temperature value is greater than the preset temperature value, the temperature difference between the maximum temperature value and the minimum temperature value can be calculated. The specific calculation formula is as follows:

[0113] Target difference = maximum temperature value - minimum temperature value;

[0114] According to the above formula, the target difference can be obtained; then, the target fire parameter corresponding to the target difference can be determined. Specifically, the mapping relationship between the preset difference and the fire parameter can be pre-stored, and the target fire parameter corresponding to the target difference can be determined based on the mapping relationship; then, the first gas composition data can be analyzed to determine the proportion of harmful gas components in the first gas composition data to obtain a first proportion. Specifically, the first gas composition data can be first analyzed by a gas analysis module (for example, a gas chromatograph) to determine which gases are included in the first gas composition data, and then the total concentration of harmful gases in these gases can be determined. The total concentration of harmful gases is divided by the total concentration of all gas components, and then multiplied by 100% to obtain the first proportion.

[0115] For example, assume that carbon monoxide, carbon dioxide, oxygen, nitrogen and sulfur dioxide are detected in the first gas component data. Among them, carbon monoxide and sulfur dioxide are considered harmful gases, and the measured concentrations are: carbon monoxide concentration is 50ppm, sulfur dioxide concentration is 20ppm, carbon dioxide concentration is 800ppm, oxygen concentration is 20%, and nitrogen concentration is 78% (assuming that these concentrations can be directly calculated). First, calculate the total concentration of harmful gases: 50+20=70ppm; then, calculate the total concentration of all gas components, and convert the concentration to ppm for calculation. Assuming the total concentration is 1000ppm, then the first ratio = (70 / 1000)×100%=7%.

[0116] Among them, ppm is the abbreviation of “parts per million”, which means “parts per million” or “parts per million”.

[0117] Furthermore, when the first ratio is greater than the preset ratio, it can be determined that a harmful gas leakage incident has occurred in the first production workshop; by sorting and classifying the above-mentioned fire identification results, target fire parameters, the first ratio and the harmful gas leakage incidents, disaster data corresponding to the first environmental data and the first image data can be obtained.

[0118] In this way, by acquiring the first environmental data and the first image data; then, analyzing the first environmental data and the first image data to obtain data such as fire identification results, target fire parameters, a first ratio and harmful gas leakage events; by sorting and classifying these data, disaster data corresponding to the first environmental data and the first image data is obtained, which comprehensively considers various information such as the fire identification results of the image data, temperature data, gas composition data, etc., and can more accurately and comprehensively assess the true situation and severity of the disaster.

[0119] S205. Determine n disaster relief equipment and m factory equipment existing within a preset distance around the humanoid robot based on the equipment distribution data; m and n are both natural numbers.

[0120] In the embodiment of the present application, the preset distance may be preset in advance or default.

[0121] In a specific embodiment, the patrol position of the humanoid robot at the current moment can be obtained first, and then, based on the equipment distribution data, disaster relief equipment and factory equipment whose distance from the patrol position is less than or equal to a preset distance can be found to obtain n disaster relief equipment and m factory equipment.

[0122] S206: Generate a disaster relief plan based on the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan.

[0123] In the embodiment of the present application, a disaster relief plan for m factory equipment can be formulated based on the target disaster data and n disaster relief equipment to obtain a target disaster relief plan.

[0124] Optionally, step S206, generating a disaster relief plan according to the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan, may include the following steps:

[0125] E1. Determine the disaster relief equipment corresponding to the target disaster data among the n disaster relief equipment, and obtain p disaster relief equipment; p is a natural number less than or equal to n;

[0126] E2. Obtain the current position of the humanoid robot to obtain a first position;

[0127] E3. Determine the distance between each of the p disaster relief equipment and the humanoid robot according to the first position to obtain p distances;

[0128] E4. Determine the order of picking up the targets corresponding to the p distances; the closer the distance, the earlier the picking order;

[0129] E5. Obtain a target equipment value parameter corresponding to each of the m factory equipment to obtain m target equipment value parameters;

[0130] E6. assigning a rescue priority to each of the m factory equipment according to the m target equipment value parameters to obtain m rescue priorities; the larger the target equipment value parameter corresponding to the factory equipment, the higher the rescue priority;

[0131] E7. Generate the target disaster relief plan according to the target taking order and the m rescue priorities.

[0132] In an embodiment of the present application, the disaster relief equipment may include at least one of the following: a fire extinguisher, a fire hydrant, a sandbag, a fire shovel, etc., which are not limited here; the target equipment value parameter is an indicator used to quantify and characterize the value of the equipment. The larger the target equipment value parameter, the higher the value of the equipment.

[0133] In a specific embodiment, the disaster relief equipment corresponding to the target disaster data among n disaster relief equipment can be first determined to obtain p disaster relief equipment. Specifically, the disaster type existing in the target factory can be determined based on the target disaster data to obtain at least one disaster type. The mapping relationship between the preset disaster type and the disaster relief equipment can be pre-stored, and the n disaster relief equipment corresponding to the at least one disaster type can be determined based on the mapping relationship. For example, when the disaster type is a fire type, the disaster relief equipment corresponding to the fire type can be a fire extinguisher, a fire hydrant, or other equipment that can be used to extinguish a fire. Then, the humanoid robot can be positioned at the current moment to obtain a first position. The positions of p disaster relief equipment can also be obtained based on the target map data to obtain p positions. The distances between the first position and the p positions are respectively calculated to obtain p distances. The target picking order is generated based on the p distances. The closer the distance, the earlier the picking order.

[0134] Furthermore, the target equipment value parameters corresponding to each of the m factory equipment can be obtained to obtain m target equipment value parameters; a rescue priority is assigned to each of the m factory equipment according to the m target equipment value parameters to obtain m rescue priorities; the larger the target equipment value parameter corresponding to the factory equipment, the higher the rescue priority; finally, a target disaster relief plan can be generated according to the target taking order and the m rescue priorities. Specifically, the target disaster relief plan can be: first take the disaster relief equipment according to the target taking order, and after taking it, rescue the factory equipment in turn according to the m rescue priorities.

[0135] In this way, by determining the disaster relief equipment corresponding to the target disaster data in n disaster relief equipment, p disaster relief equipment are obtained; the current position of the humanoid robot is obtained to obtain the first position; the distance between each of the p disaster relief equipment and the humanoid robot is determined according to the first position to obtain p distances; the target picking order corresponding to the p distances is determined; the target equipment value parameters corresponding to each of the m factory equipment are obtained to obtain m target equipment value parameters; according to the m target equipment value parameters, rescue priorities are assigned to each of the m factory equipment to obtain m rescue priorities; a target disaster relief plan is generated according to the target picking order and the m rescue priorities. On the one hand, accurately determining the required disaster relief equipment and generating the picking order according to the size of the distance can reduce the time for searching for equipment, enable the humanoid robot to quickly obtain effective rescue tools, and speed up the pace of disaster relief operations. On the other hand, the rescue priority is assigned according to the value parameters of the factory equipment to ensure that limited rescue resources are first used for the most critical and valuable equipment to minimize economic losses.

[0136] Optionally, step E5, obtaining the target equipment value parameter corresponding to each of the m factory equipment to obtain the m target equipment value parameters, may include the following steps:

[0137] Obtaining the market price of each of the m factory equipment to obtain m market prices;

[0138] Determine a reference device value parameter corresponding to each of the m market prices to obtain m reference device value parameters;

[0139] Obtaining the first production efficiency of each of the m factory equipment to obtain m first production efficiencies;

[0140] Determine the damage coefficient of the target disaster data on each of the m factory equipment to obtain m damage coefficients;

[0141] Determine the influence factor corresponding to each of the m damage coefficients to obtain m influence factors;

[0142] Adjusting the m first production efficiencies according to the m influencing factors to obtain m second production efficiencies;

[0143] Determine the target optimization factor corresponding to each of the m second production efficiencies to obtain m target optimization factors;

[0144] The m reference device value parameters are optimized according to the m target optimization factors to obtain the m target device value parameters.

[0145] In the embodiment of the present application, the damage coefficient represents the relative degree or possibility of damage caused by the disaster to the factory equipment. It is a quantitative indicator and its value is usually between 0 and 1.

[0146] In a specific embodiment, the market price of each of the m factory equipment can be obtained first to obtain m market prices. Specifically, the equipment model of the m factory equipment can be obtained from the equipment distribution data to obtain m equipment models. The m equipment models can be used to search on equipment sales websites, e-commerce platforms or industry equipment trading platforms to obtain m market prices. Then, the reference equipment value parameters corresponding to each of the m market prices can be determined to obtain m reference equipment value parameters. Specifically, the mapping relationship between preset market prices and reference equipment value parameters can be pre-stored, and the m reference equipment value parameters corresponding to the m market prices can be determined based on the mapping relationship, and each market price corresponds to a reference equipment value parameter.

[0147] Furthermore, the first production efficiency of each of the m factory equipment can be obtained to obtain m first production efficiencies. Specifically, the production data of the m factory equipment can be obtained from the control system of the target factory, and the m first production efficiencies can be calculated based on the production data and the production efficiency calculation formula; then, the damage coefficient of the target disaster data for each of the m factory equipment can be determined to obtain m damage coefficients. Specifically, the target disaster type corresponding to the target disaster data can be obtained, and under the target disaster type, the mapping relationship between the preset factory equipment and the damage coefficient can be pre-stored, and the m damage coefficients corresponding to the m factory equipment can be determined based on the mapping relationship.

[0148] Next, the influence factor corresponding to each of the m damage coefficients can be determined to obtain m influence factors. Specifically, a mapping relationship between preset damage coefficients and influence factors can be pre-stored, and m influence factors corresponding to the m damage coefficients can be determined based on the mapping relationship. Each damage coefficient corresponds to an influence factor, and the value range of the influence factor is -1~0. Then, the m first production efficiencies can be adjusted according to the m influence factors. The specific calculation formula is as follows:

[0149] Second production efficiency c = first production efficiency c × (1 + impact factor c);

[0150] Among them, the second production efficiency c is any one of the m second production efficiencies; the first production efficiency c is the first production efficiency corresponding to the second production efficiency c in the m first production efficiencies; the above formula is calculated m times to obtain m second production efficiencies; then, the target optimization factor corresponding to each of the m second production efficiencies can be determined to obtain m target optimization factors. Specifically, the mapping relationship between the preset production efficiency and the target optimization factor can be pre-stored, and the m target optimization factors corresponding to the m production efficiencies are determined based on the mapping relationship. The value range of the target optimization factor can be -1~0; finally, the m reference equipment value parameters can be optimized according to the m target optimization factors. The specific calculation formula is as follows:

[0151] Target device value parameter d = reference device value parameter d × (1 + target optimization factor d);

[0152] Among them, the target device value parameter d is any one of the m target device value parameters; the reference device value parameter d is the reference device value parameter corresponding to the target device value parameter d in the m reference device value parameters; and the m target device value parameters are obtained by calculating m times through the above formula.

[0153] In this way, by obtaining the market price of each of the m factory equipment, m market prices are obtained; m reference equipment value parameters corresponding to the m market prices are determined; m first production efficiencies corresponding to the m factory equipment are obtained; the destruction coefficient of the target disaster data on each of the m factory equipment is determined to obtain m destruction coefficients; m influencing factors corresponding to the m destruction coefficients are determined; the m first production efficiencies are adjusted according to the m influencing factors to obtain m second production efficiencies; m target optimization factors corresponding to the m second production efficiencies are determined; the m reference equipment value parameters are optimized according to the m target optimization factors to obtain m target equipment value parameters. By determining the destruction coefficient and the influencing factor, the impact of the disaster on the production efficiency of each factory equipment can be more accurately estimated, thereby more accurately calculating the possible economic losses.

[0154] Optionally, the method may further include the following steps:

[0155] F1. Obtain a target position corresponding to a first disaster relief device; the first disaster relief device is any one of the p disaster relief devices; the first disaster relief device includes a fire extinguisher; the first disaster relief device is placed in a preset enclosed space, and the preset enclosed space includes a protective glass door and an opening button;

[0156] F2. determining a target distance between the target position and the first position;

[0157] F3. Determine a first moving speed corresponding to the target distance;

[0158] F4, controlling the humanoid robot to move toward the target position at the first moving speed, and identifying data of the target obstacle in real time during the movement to obtain the target obstacle data;

[0159] F5. Controlling the humanoid robot to avoid the target obstacle according to the target obstacle data, so as to safely reach the target position;

[0160] F6. Searching for the first disaster relief equipment through the built-in visual sensor of the humanoid robot. When the first disaster relief equipment is found, detecting the preset enclosed space based on the preset button image data through the visual sensor to determine the opening position corresponding to the opening button;

[0161] F7. Control the humanoid robot to press the opening button with a preset force according to the opening position to open the protective glass door, and after opening the protective glass door, control the humanoid robot to take the first disaster relief equipment with a preset taking operation;

[0162] F8. Determine the time taken by the humanoid robot to pick up the first disaster relief equipment, and obtain a first time taken;

[0163] F9. Determine a first adjustment parameter corresponding to the first consuming time;

[0164] F10. Adjust the first moving speed according to the first adjustment parameter to obtain a second moving speed;

[0165] F11. Control the humanoid robot to carry the first disaster relief equipment and return to the first position at the second moving speed.

[0166] In an embodiment of the present application, the target obstacle data may include at least one of the following: obstacle position data, obstacle size data, etc., which are not limited here; the preset button image data and the preset force can be preset or defaulted in advance.

[0167] In a specific embodiment, the target position of the first disaster relief equipment can be obtained from the equipment distribution data; then, the target distance between the target position and the first position can be calculated according to the target map data; then, the first moving speed can be determined according to the target distance. Specifically, a mapping relationship between a preset distance and a moving speed can be pre-stored, and the first moving speed corresponding to the target distance can be determined based on the mapping relationship; then, the humanoid robot can be controlled to move toward the target position at the first moving speed, and the data of the target obstacle can be identified in real time during the movement to obtain the target obstacle data. Specifically, when the humanoid robot moves, data of the surrounding environment can be collected through a camera module and an environmental data collection module, and the collected raw data can be processed and analyzed in real time to identify objects that may constitute obstacles and obtain target obstacle data.

[0168] Furthermore, the humanoid robot can be controlled to avoid the target obstacle according to the target obstacle data to safely reach the target location. Specifically, the distance and size of the target obstacle can be determined according to the target obstacle data, and the moving direction and speed of the humanoid robot can be dynamically adjusted. If the target obstacle is small, the humanoid robot can choose to cross the target obstacle. If the target obstacle is large, the humanoid robot can choose to bypass the target obstacle to safely reach the target location. Then, the first disaster relief equipment can be searched through the built-in visual sensor of the humanoid robot. When the first disaster relief equipment is found, the visual sensor can be used to search for the first disaster relief equipment based on the preset button image data. The preset enclosed space is detected to determine the opening position corresponding to the opening button. Specifically, the appearance characteristics, shape, color and other image data of the first disaster relief equipment can be entered into the robot's database in advance as a basis for identification. Then, the humanoid robot is controlled to move near the target position, and the surrounding images are continuously collected through the visual sensor, and the collected images are compared with the entered image data of the first disaster relief equipment, so as to find the first disaster relief equipment, that is, the preset enclosed space. Then, the image of the preset enclosed space can be collected by the visual sensor, and matched with the preset button image data. If the match is successful, it is confirmed as the opening button.

[0169] Then, according to the opening position, the humanoid robot is controlled to press the opening button with a preset force to open the protective glass door. After the protective glass door is opened, the humanoid robot is controlled to take the first disaster relief equipment with a preset taking operation; then, the time consumed by the humanoid robot to take the first disaster relief equipment can be determined to obtain the first consuming time. Specifically, the time when the humanoid robot first reaches the first position can be taken as the starting time, and the time when the first disaster relief equipment is taken as the ending time. The first consuming time can be obtained by subtracting the starting time from the ending time; then, the first adjustment parameter corresponding to the first consuming time can be determined. Specifically, a mapping relationship between a preset consuming time and an adjustment parameter can be pre-stored, and the first adjustment parameter corresponding to the first consuming time is determined based on the mapping relationship. The value range of the first adjustment parameter can be 0~0.5; then, the first moving speed can be adjusted according to the first adjustment parameter. The specific calculation formula is as follows:

[0170] Second moving speed = first moving speed × (1 + first adjustment parameter);

[0171] The second moving speed can be obtained according to the above formula; finally, the humanoid robot can be controlled to carry the first disaster relief equipment and return to the first position at the second moving speed.

[0172] In this way, by adjusting the round-trip movement speed according to the target distance and the first consuming time, it is possible to reduce time waste, quickly obtain the required disaster relief equipment, and improve the efficiency of the disaster relief operation. In addition, the humanoid robot uses visual sensors to accurately find and open the preset enclosed space of the disaster relief equipment, and pick up the disaster relief equipment with the preset picking operation, ensuring the accuracy and reliability of the operation.

[0173] S207: Control the humanoid robot to provide disaster relief to the target factory according to the target disaster relief plan to protect production safety of the target factory.

[0174] In an embodiment of the present application, the humanoid robot is controlled to perform operations such as fire fighting and disaster relief, rescuing factory equipment, and shutting down dangerous equipment according to the target disaster relief plan. For example, a fire extinguisher is used to put out open flames, demolition tools are used to open rescue channels, and leaking pipeline valves are closed, etc., to protect the production safety of the target factory.

[0175] It can be seen that the factory patrol and disaster relief method based on humanoid robots described in the present application includes: obtaining target map data and equipment distribution data corresponding to the target factory; obtaining the initial position corresponding to the humanoid robot; generating a patrol path based on the target map data and the initial position to obtain the target patrol path; controlling the humanoid robot to patrol based on the target patrol path, and conducting a disaster inspection on the target factory during the patrol to obtain target disaster data; determining the n disaster relief equipment and m factory equipment existing in a preset distance around the humanoid robot based on the equipment distribution data; generating a disaster relief plan based on the target disaster data, n disaster relief equipment and m factory equipment to obtain a target disaster relief plan; controlling the humanoid robot to perform disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory. On the one hand, the patrol path generated according to the map data and the initial position of the robot can ensure full coverage of the factory area, reduce monitoring blind spots, and improve the comprehensiveness of detection. On the other hand, by patrolling the factory along a predetermined path, the humanoid robot can promptly discover the disaster and automatically generate a disaster relief plan to rescue the factory, thereby improving the safety of the factory.

[0176] See also Figure 5 , Figure 5 1 is a functional module block diagram of a humanoid robot-based factory patrol and disaster relief device provided in an embodiment of the present application. The humanoid robot-based factory patrol and disaster relief device 500 includes: an acquisition unit 501, a control unit 502, and a disaster relief unit 503, wherein:

[0177] The acquisition unit 501 is used to acquire the target map data and equipment distribution data corresponding to the target factory; and acquire the initial position corresponding to the humanoid robot;

[0178] The control unit 502 is used to generate a patrol path based on the target map data and the initial position to obtain a target patrol path; control the humanoid robot to patrol based on the target patrol path, and perform a disaster inspection on the target factory during the patrol to obtain target disaster data; determine n disaster relief equipment and m factory equipment within a preset distance around the humanoid robot based on the equipment distribution data; m and n are both natural numbers; generate a disaster relief plan based on the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan;

[0179] The disaster relief unit 503 is used to control the humanoid robot to perform disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory.

[0180] Optionally, in the aspect of generating a patrol path based on the target map data and the initial position to obtain a target patrol path, the control unit 502 is specifically configured to:

[0181] Acquire a position with a wide field of view in the target map data to obtain a plurality of positions with a wide field of view;

[0182] Generate a reference patrol path according to the initial position and the multiple positions with wide field of view to obtain multiple reference patrol paths;

[0183] Determine the path characteristic data corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of path characteristic data; the path characteristic data includes any one of the following: the length of a straight path, the number of curved paths, the height of a lifting path, and the inclination of a climbing path;

[0184] Determine a path distance corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of path distances;

[0185] Determine a difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances and the plurality of path feature data, to obtain a plurality of difficulty values;

[0186] The minimum difficulty value among the multiple difficulty values ​​is obtained, and a reference patrol path corresponding to the minimum difficulty value among the multiple reference patrol paths is determined as the target patrol path.

[0187] Optionally, in the aspect of determining the difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances and the plurality of path feature data to obtain the plurality of difficulty values, the control unit 502 is specifically configured to:

[0188] Determine a reference difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances to obtain a plurality of reference difficulty values;

[0189] Determine the estimated power consumption corresponding to each path characteristic data in the plurality of path characteristic data to obtain a plurality of estimated power consumptions;

[0190] Determine a first influence coefficient corresponding to each of the plurality of estimated power consumptions to obtain a plurality of first influence coefficients;

[0191] Acquire an environmental parameter corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of environmental parameters;

[0192] Determine a second influence coefficient corresponding to each of the multiple environmental parameters to obtain multiple second influence coefficients;

[0193] The plurality of difficulty values ​​are determined according to the plurality of first influence coefficients, the plurality of second influence coefficients, and the plurality of reference difficulty values.

[0194] Optionally, the humanoid robot includes a camera module and an environmental data acquisition module. In terms of inspecting the target factory for disaster conditions during the patrol process and obtaining target disaster data, the control unit 502 is specifically used to:

[0195] When the humanoid robot reaches a first wide-view position, determining the production workshops that the humanoid robot can observe at the first wide-view position, and obtaining i production workshops; the first wide-view position is any wide-view position included in the target patrol path; i is a natural number;

[0196] Collecting image data of the i production workshops through the camera module to obtain i image data;

[0197] Performing disaster identification on the i image data to obtain i identification results; the identification results include possible existence of a disaster or no disaster;

[0198] Determine j identification results among the i identification results that may have disaster situations; j is a natural number less than or equal to i;

[0199] Determine the j production workshops corresponding to the j identification results in the i production workshops;

[0200] Obtaining the distance between each of the j production workshops and the humanoid robot to obtain j distances;

[0201] The j production workshops are sorted according to the j distances to obtain a target viewing order; the shorter the distance, the higher the viewing order;

[0202] Controlling the humanoid robot to enter the j production workshops in sequence according to the target viewing order, collecting image data of the j production workshops through the camera module to obtain j image data, and collecting environmental data of the j production workshops through the environmental data collection module to obtain j environmental data;

[0203] Determine first disaster data corresponding to the first position with a wide field of view based on the j pieces of environmental data and the j pieces of image data;

[0204] Determine disaster data corresponding to all positions with a wide field of view in the target patrol path except the first position with a wide field of view, and obtain at least one disaster data;

[0205] The target disaster data is determined based on the at least one disaster data and the first disaster data.

[0206] Optionally, in determining the first disaster data corresponding to the first position with a wide field of view based on the j pieces of environmental data and the j pieces of image data, the control unit 502 is specifically configured to:

[0207] Acquire first environmental data and first image data; the first environmental data includes first temperature data and first gas composition data; the first environmental data is any environmental data among the j environmental data; the first image data is image data corresponding to the first environmental data among the j image data;

[0208] Recognize the first image data to determine a fire recognition result of the first production workshop corresponding to the first image data; the fire recognition result includes whether a fire event occurs or not;

[0209] When the fire identification result includes the fire event, obtaining a maximum temperature value and a minimum temperature value in the first temperature data;

[0210] When the maximum temperature value is greater than a preset temperature value, determining a temperature difference between the maximum temperature value and the minimum temperature value to obtain a target difference;

[0211] Determining a target fire parameter corresponding to the target difference;

[0212] Analyzing the first gas composition data, determining the proportion of harmful gas components in the first gas composition data, and obtaining a first proportion;

[0213] When the first ratio is greater than a preset ratio, it is determined that a harmful gas leakage event has occurred in the first production workshop;

[0214] Disaster data corresponding to the first environmental data and the first image data are determined based on the fire identification result, the target fire parameter, the first ratio, and the harmful gas leakage event.

[0215] Optionally, in the aspect of generating a disaster relief plan according to the target disaster situation data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan, the control unit 502 is specifically used to:

[0216] Determine the disaster relief equipment corresponding to the target disaster data among the n disaster relief equipment, and obtain p disaster relief equipment; p is a natural number less than or equal to n;

[0217] Obtaining the current position of the humanoid robot to obtain a first position;

[0218] Determine the distance between each of the p disaster relief equipment and the humanoid robot according to the first position to obtain p distances;

[0219] Determine the order of picking up the targets corresponding to the p distances; the closer the distance, the earlier the picking order;

[0220] Obtaining a target equipment value parameter corresponding to each of the m factory equipment to obtain m target equipment value parameters;

[0221] Assigning a rescue priority to each of the m plant equipment according to the m target equipment value parameters to obtain m rescue priorities; the larger the target equipment value parameter corresponding to the plant equipment, the higher the rescue priority;

[0222] The target disaster relief plan is generated according to the target taking order and the m rescue priorities.

[0223] Optionally, the humanoid robot-based factory patrol and disaster relief device 500 is also specifically used for:

[0224] Obtain a target position corresponding to a first disaster relief device; the first disaster relief device is any one of the p disaster relief devices; the first disaster relief device includes a fire extinguisher; the first disaster relief device is placed in a preset enclosed space, and the preset enclosed space includes a protective glass door and an opening button;

[0225] determining a target distance between the target position and the first position;

[0226] Determine a first moving speed corresponding to the target distance;

[0227] Controlling the humanoid robot to move toward the target position at the first moving speed, and identifying data of a target obstacle in real time during the movement to obtain the target obstacle data;

[0228] Controlling the humanoid robot to avoid the target obstacle according to the target obstacle data so as to safely reach the target location;

[0229] Searching for the first disaster relief equipment through the built-in visual sensor of the humanoid robot, and when the first disaster relief equipment is found, detecting the preset enclosed space based on the preset button image data through the visual sensor to determine the opening position corresponding to the opening button;

[0230] According to the opening position, the humanoid robot is controlled to press the opening button with a preset force to open the protective glass door, and after the protective glass door is opened, the humanoid robot is controlled to take the first disaster relief equipment with a preset taking operation;

[0231] Determining the time taken by the humanoid robot to pick up the first disaster relief equipment to obtain a first taken time;

[0232] Determine a first adjustment parameter corresponding to the first consuming time;

[0233] Adjust the first moving speed according to the first adjustment parameter to obtain a second moving speed;

[0234] The humanoid robot is controlled to carry the first disaster relief equipment and return to the first position at the second moving speed.

[0235] In a specific implementation, the humanoid robot-based factory patrol and disaster relief device 500 described in the embodiment of the present invention can also execute other implementations described in the humanoid robot-based factory patrol and disaster relief method provided in the above-mentioned embodiment of the present invention, which will not be repeated here.

[0236] See also Figure 6 , Figure 6 This is a structural schematic diagram of a humanoid robot provided in an embodiment of the present application, wherein the humanoid robot includes a processor, a memory, a communication interface, and one or more programs, wherein the processor, the memory, and the communication interface are interconnected via a bus, wherein the one or more programs are stored in the memory and are configured to be executed by the processor, wherein the one or more programs include instructions for executing other implementations described in the humanoid robot-based factory patrol and disaster relief method provided in the above-mentioned embodiment of the present invention, which will not be repeated here.

[0237] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0238] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0239] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0240] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0241] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0242] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the above-mentioned methods in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0243] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A factory patrol and disaster relief method based on a humanoid robot, characterized in that: The method comprises: Obtain target map data and equipment distribution data corresponding to the target factory; Get the initial position corresponding to the humanoid robot; Generate a patrol path based on the target map data and the initial position to obtain a target patrol path; Controlling the humanoid robot to patrol based on the target patrol path, and inspecting the disaster situation of the target factory during the patrol process to obtain target disaster situation data; Determine n disaster relief equipment and m factory equipment existing within a preset distance around the humanoid robot based on the equipment distribution data; m and n are both natural numbers; Generate a disaster relief plan according to the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan; The humanoid robot is controlled to perform disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory.

2. The method according to claim 1, characterized in that The generating of the patrol path based on the target map data and the initial position to obtain the target patrol path includes: Acquire a position with a wide field of view in the target map data to obtain a plurality of positions with a wide field of view; Generate a reference patrol path according to the initial position and the multiple positions with wide field of view to obtain multiple reference patrol paths; Determine the path characteristic data corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of path characteristic data; the path characteristic data includes any one of the following: the length of a straight path, the number of curved paths, the height of a lifting path, and the inclination of a climbing path; Determine a path distance corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of path distances; Determine a difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances and the plurality of path feature data, to obtain a plurality of difficulty values; The minimum difficulty value among the multiple difficulty values ​​is obtained, and a reference patrol path corresponding to the minimum difficulty value among the multiple reference patrol paths is determined as the target patrol path.

3. The method according to claim 2, characterized in that The determining, according to the multiple path distances and the multiple path feature data, a difficulty value corresponding to each reference patrol path in the multiple reference patrol paths to obtain multiple difficulty values ​​comprises: Determine a reference difficulty value corresponding to each reference patrol path in the plurality of reference patrol paths according to the plurality of path distances to obtain a plurality of reference difficulty values; Determine the estimated power consumption corresponding to each path characteristic data in the plurality of path characteristic data to obtain a plurality of estimated power consumptions; Determine a first influence coefficient corresponding to each of the plurality of estimated power consumptions to obtain a plurality of first influence coefficients; Acquire an environmental parameter corresponding to each reference patrol path in the plurality of reference patrol paths to obtain a plurality of environmental parameters; Determine a second influence coefficient corresponding to each of the multiple environmental parameters to obtain multiple second influence coefficients; The plurality of difficulty values ​​are determined according to the plurality of first influence coefficients, the plurality of second influence coefficients, and the plurality of reference difficulty values.

4. The method according to claim 2 or 3, characterized in that The humanoid robot includes a camera module and an environmental data acquisition module. During the patrol process, the target factory is inspected for disaster conditions to obtain target disaster data, including: When the humanoid robot reaches a first wide-view position, determining the production workshops that the humanoid robot can observe at the first wide-view position, and obtaining i production workshops; the first wide-view position is any wide-view position included in the target patrol path; i is a natural number; Collecting image data of the i production workshops through the camera module to obtain i image data; Performing disaster identification on the i image data to obtain i identification results; the identification results include possible existence of a disaster or no disaster; Determine j identification results among the i identification results that may have disaster situations; j is a natural number less than or equal to i; Determine the j production workshops corresponding to the j identification results in the i production workshops; Obtaining the distance between each of the j production workshops and the humanoid robot to obtain j distances; The j production workshops are sorted according to the j distances to obtain a target viewing order; the shorter the distance, the higher the viewing order; Controlling the humanoid robot to enter the j production workshops in sequence according to the target viewing order, collecting image data of the j production workshops through the camera module to obtain j image data, and collecting environmental data of the j production workshops through the environmental data collection module to obtain j environmental data; Determine first disaster data corresponding to the first position with a wide field of view based on the j pieces of environmental data and the j pieces of image data; Determine disaster data corresponding to all positions with a wide field of view except the first position with a wide field of view in the target patrol path, and obtain at least one disaster data; The target disaster data is determined based on the at least one disaster data and the first disaster data.

5. The method according to claim 4, characterized in that The determining, based on the j pieces of environmental data and the j pieces of image data, first disaster data corresponding to the first position with a wide field of view includes: Acquire first environmental data and first image data; the first environmental data includes first temperature data and first gas composition data; the first environmental data is any environmental data among the j environmental data; the first image data is image data corresponding to the first environmental data among the j image data; Recognize the first image data to determine a fire recognition result of the first production workshop corresponding to the first image data; the fire recognition result includes whether a fire event occurs or not; When the fire identification result includes the fire event, obtaining a maximum temperature value and a minimum temperature value in the first temperature data; When the maximum temperature value is greater than a preset temperature value, determining a temperature difference between the maximum temperature value and the minimum temperature value to obtain a target difference; Determining a target fire parameter corresponding to the target difference; Analyzing the first gas composition data, determining the proportion of harmful gas components in the first gas composition data, and obtaining a first proportion; When the first ratio is greater than a preset ratio, it is determined that a harmful gas leakage event has occurred in the first production workshop; Disaster data corresponding to the first environmental data and the first image data are determined based on the fire identification result, the target fire parameter, the first ratio, and the harmful gas leakage event.

6. The method according to any one of claims 1 to 3, characterized in that: The generating of a disaster relief plan according to the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan includes: Determine the disaster relief equipment corresponding to the target disaster data among the n disaster relief equipment, and obtain p disaster relief equipment; p is a natural number less than or equal to n; Obtaining the current position of the humanoid robot to obtain a first position; Determine the distance between each of the p disaster relief equipment and the humanoid robot according to the first position to obtain p distances; Determine the order of picking up the targets corresponding to the p distances; the closer the distance, the earlier the picking order; Obtaining a target equipment value parameter corresponding to each of the m factory equipment to obtain m target equipment value parameters; Assigning a rescue priority to each of the m plant equipment according to the m target equipment value parameters, to obtain m rescue priorities; the larger the target equipment value parameter corresponding to the plant equipment, the higher the rescue priority; The target disaster relief plan is generated according to the target taking order and the m rescue priorities.

7. The method according to claim 6, characterized in that The method further comprises: Obtain a target position corresponding to a first disaster relief device; the first disaster relief device is any one of the p disaster relief devices; the first disaster relief device includes a fire extinguisher; the first disaster relief device is placed in a preset enclosed space, and the preset enclosed space includes a protective glass door and an opening button; determining a target distance between the target position and the first position; Determine a first moving speed corresponding to the target distance; Controlling the humanoid robot to move toward the target position at the first moving speed, and identifying data of a target obstacle in real time during the movement to obtain the target obstacle data; Controlling the humanoid robot to avoid the target obstacle according to the target obstacle data so as to safely reach the target location; Searching for the first disaster relief equipment through the built-in visual sensor of the humanoid robot, and when the first disaster relief equipment is found, detecting the preset enclosed space based on the preset button image data through the visual sensor to determine the opening position corresponding to the opening button; According to the opening position, the humanoid robot is controlled to press the opening button with a preset force to open the protective glass door, and after the protective glass door is opened, the humanoid robot is controlled to take the first disaster relief equipment with a preset taking operation; Determining the time taken by the humanoid robot to pick up the first disaster relief equipment to obtain a first taken time; Determine a first adjustment parameter corresponding to the first consuming time; Adjust the first moving speed according to the first adjustment parameter to obtain a second moving speed; The humanoid robot is controlled to carry the first disaster relief equipment and return to the first position at the second moving speed.

8. A factory patrol and disaster relief device based on a humanoid robot, characterized in that: The device comprises: an acquisition unit, a control unit, and a disaster relief unit, wherein: The acquisition unit is used to acquire target map data and equipment distribution data corresponding to the target factory; and acquire an initial position corresponding to the humanoid robot; The control unit is used to generate a patrol path based on the target map data and the initial position to obtain a target patrol path; control the humanoid robot to patrol based on the target patrol path, and perform a disaster inspection on the target factory during the patrol to obtain target disaster data; determine n disaster relief equipment and m factory equipment within a preset distance around the humanoid robot based on the equipment distribution data; m and n are both natural numbers; generate a disaster relief plan based on the target disaster data, the n disaster relief equipment and the m factory equipment to obtain a target disaster relief plan; The disaster relief unit is used to control the humanoid robot to carry out disaster relief on the target factory according to the target disaster relief plan to protect the production safety of the target factory.

9. A humanoid robot, characterized in that: include: A processor and a memory, wherein the memory is used to store one or more programs and is configured to be executed by the processor, wherein the program includes instructions for executing the steps in the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program for electronic data exchange is stored, wherein the computer program enables a computer to execute the method according to any one of claims 1 to 7.

Images