Task planning method and system of building robot, robot and medium

Abstract

The invention provides a task planning method for a building robot, and the method comprises the steps: obtaining the semantic information of a to-be-constructed region, and screening a to-be-constructed object; according to the screened to-be-constructed objects, matching proper construction types and mechanical construction parameters from a preset process library, and generating a construction operation node information list; generating a path node information list according to the semantic information and the construction operation node information list; according to the path node information list, matching actions of a proper process in a preset track action library, and generating a specific construction action sequence; and according to the path node information list and the semantic information, judging whether a connecting line between adjacent path nodes has interference or not, if so, adding transfer path nodes, and if not, generating a final construction path. According to the method, it can be ensured that the construction task of each unit reaches the local optimal solution, the method can better adapt to the complex environment of a building construction site, and the safety and high efficiency of the building robot in the construction process are ensured.

Description

Technical Field

[0001] The present application relates to the technical field of construction automation, and in particular, to a task planning method, system, robot, and medium for a construction robot. Background Art

[0002] With the continuous development of the construction industry, intelligent construction has become the key focus to promote the high-quality development of the industry. As a new type of production factor, construction robots have become the key means to promote intelligent construction. However, construction robots face many challenges in practical applications. At present, there is no mature task planning technology to empower construction robots, and most of the current control models for construction robots have not evolved to the level of automated construction.

[0003] In recent years, due to the advantages of reinforcement learning such as being able to handle high-dimensional state and action spaces and being suitable for dealing with nonlinear and complex relationships, it has begun to be applied to the task planning of construction robots. However, when the reinforcement learning algorithm is applied to the construction environment, there are still many problems, such as high-dimensional data input being prone to cause the curse of dimensionality, the action space and state space being single, the reward function being sparse, and there being no successful practical application cases promoted in the construction field. Summary of the Invention

[0004] Aiming at the defects in the prior art, the purpose of the present application is to provide a task planning method, system, robot, and medium for a construction robot, to realize an efficient, reliable, and highly adaptable task planning method for construction robots, and to improve the intelligent construction level of construction robots.

[0005] In one aspect of the present application, a task planning method for a construction robot is provided, including:

[0006] Obtain the semantic information of the area to be constructed, and screen the objects to be constructed;

[0007] According to the screened objects to be constructed, match the appropriate construction types and mechanical construction parameters from a preset process library, and generate a list of construction operation node information;

[0008] Generate a list of path node information according to the semantic information and the list of construction operation node information;

[0009] According to the list of path node information, match the actions of the appropriate process in a preset trajectory action library, and generate a specific construction action sequence;

[0010] According to the list of path node information and the semantic information, judge whether there is interference in the connection between adjacent path nodes. If so, add transfer path nodes. If not, generate the final construction path.

[0011] Further, obtaining the semantic information of the area to be constructed and screening the construction objects includes:

[0012] Obtaining the image or point cloud data of the area to be constructed, extracting the information in the image or the point cloud data, and generating the semantic information of the area to be constructed;

[0013] According to the semantic information of the area to be constructed, judging the normal vector orientation of different information in the semantic information, and determining the construction objects;

[0014] Performing depth calculation on the construction objects to determine the distance between the construction objects and the building wall surface;

[0015] According to the distance, screening out the construction objects to be constructed.

[0016] Further, according to the screened construction objects to be constructed, matching appropriate construction types and mechanical construction parameters from a preset process library to generate a list of construction operation node information, including:

[0017] Obtaining the screened construction objects to be constructed and determining the dimension information of the construction area of the construction objects to be constructed;

[0018] According to the dimension information, judging whether the dimension information is greater than a set threshold. If so, dividing the construction area into multiple construction areas according to the size of the construction area. If not, fixing the dimension process rule;

[0019] Matching appropriate construction types and mechanical construction parameters from a preset process library to generate a list of construction operation node information.

[0020] Further, the list of construction operation node information includes the operation node information of multiple construction areas;

[0021] The set threshold is the size of the building robot body.

[0022] Further, according to the semantic information and the list of construction operation node information, generating a list of path node information, including:

[0023] Obtaining the list of construction operation node information and the semantic information, and generating the information of multiple path nodes corresponding to the operation node information of the multiple construction areas;

[0024] According to the information of the multiple path nodes, allocating the corresponding relationship between the operation nodes and the path nodes, generating the three-dimensional coordinate information of the multiple path nodes, and forming a list of the multiple path node information.

[0025] Further, according to the list of path node information, matching appropriate process actions in a preset trajectory action library to generate a specific construction action sequence, including:

[0026] Obtain a list of path node information, and in combination with the semantic information, the construction type, and the mechanical construction parameters, select a trajectory action that matches the list of path node information in a preset trajectory action library;

[0027] Fine-tune the construction section in the operation node information according to the trajectory action to generate a specific construction action sequence.

[0028] Further, the method for judging whether there is interference in the connection line between adjacent path nodes according to the list of path node information and the semantic information, and if so, adding a transfer path node, and if not, generating a final construction path includes:

[0029] Obtain the list of path node information and the semantic information, and convert the semantic information into map information;

[0030] According to the list of path node information and the map information, obtain the spatial relationship between two adjacent path node information and the map information;

[0031] Connect two adjacent path node information and judge whether there is interference with the map information. If so, add a transfer path node to the specific construction action sequence. If not, generate a final construction path.

[0032] In a second aspect of the present application, there is provided a task planning system for a construction robot, including:

[0033] A construction object identification module, configured to receive semantic information of a to-be-constructed area, identify a construction object, and judge the type of the construction object;

[0034] A process classification module, configured to match a suitable construction type and mechanical construction parameters from a preset process library according to the selected construction object, and generate a list of construction operation node information;

[0035] A path node module, configured to generate a list of path node information according to the semantic information and the list of construction operation node information;

[0036] A decision-making trajectory action module, configured to match an action of a suitable process in a preset trajectory action library according to the list of path node information, and generate a specific construction action sequence;

[0037] A target planning path determination module, configured to judge whether there is interference in the connection line between adjacent path nodes according to the list of path node information and the semantic information. If so, add a transfer path node. If not, generate a final construction path.

[0038] In the third aspect of the present application, a construction robot is provided, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is configured to implement the task planning method of the construction robot described above.

[0039] In the fourth aspect of the present application, a storage medium is provided. The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the task planning method of the construction robot is implemented.

[0040] Compared with the prior art, the present application has at least one of the following beneficial effects:

[0041] 1. The task planning method of the construction robot in the present application significantly improves the task planning accuracy and reliability of the construction robot, enhances the adaptability to the construction environment, improves the construction efficiency and quality, and has important practical application value and broad development prospects through technical means such as semantic information screening, path node optimization, and obstacle avoidance strategies.

[0042] 2. By receiving the semantic information of the construction area, the present application discriminates the construction objects and matches appropriate construction types and mechanical construction parameters according to the discrimination results, enabling the construction robot to better adapt to different construction environments and task requirements, improving the flexibility and adaptability of construction, and appropriately adding transfer stations during the construction process to ensure reasonable obstacle avoidance and safe postures, further enhancing the stability and safety of the construction robot in complex construction environments. By dynamically adjusting the path and transfer stations, the construction robot can effectively avoid obstacles and ensure the smooth progress of the construction process.

[0043] 3. Through the optimization of path nodes and the generation of construction action sequences, the present application can generate smooth and efficient construction paths, improve the construction efficiency, and ensure the uniformity and consistency of construction quality. During the task planning process, the information of all path nodes is concatenated to generate the final construction path, ensuring the coherence and integrity of the construction process, reducing the pause and adjustment time during the construction process, and improving the overall construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, objectives, and advantages of the present application will become more apparent:

[0045] Figure 1 It is a flowchart of a task planning method for a construction robot in an embodiment of the present application.

[0046] Figure 2 It is a flowchart for screening construction objects in an embodiment of the present application.

[0047] Figure 3 Flow chart of generating construction operation node information list in an embodiment of the present application

[0048] Figure 4 Example diagram of construction operation node structure in an embodiment of the present application

[0049] Figure 5 Flow chart of generating path node information list in an embodiment of the present application

[0050] Figure 6 Flow chart of generating specific construction action sequence in an embodiment of the present application

[0051] Figure 7 Flow chart of determining final construction path in an embodiment of the present application

[0052] Figure 8 Example diagram of the method for adding transfer path nodes in an embodiment of the present application

[0053] Figure 9 Example diagram of depth calculation in an embodiment of the present application;

[0054] Figure 10 Flow chart of identifying crossbeams and bay windows in a preferred embodiment of the present invention;

[0055] Figure 11 Schematic diagram of the process of identifying a single door and window hole in a preferred embodiment of the present invention

[0056] Figure 12 Schematic diagram of the process of identifying a single bay window in a preferred embodiment of the present invention

[0057] Figure 13 Schematic diagram of the process of identifying a single crossbeam in a preferred embodiment of the present invention Detailed implementation manners

[0058] The present application will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any form. It should be noted that those of ordinary skill in the art can make several deformations and improvements without departing from the concept of the present application. These all belong to the protection scope of the present application.

[0059] Refer to Figure 1 As shown, a task planning method for a construction robot in an embodiment of the present application includes: S100, obtaining semantic information of the area to be constructed and screening the objects to be constructed;

[0060] S200. According to the selected construction objects to be constructed, match appropriate construction types and mechanical construction parameters from a preset process library, and generate a list of construction operation node information;

[0061] S300. According to the semantic information and the list of construction operation node information, generate a list of path node information;

[0062] S400. According to the list of path node information, match actions of appropriate processes in a preset trajectory action library, and generate a specific construction action sequence;

[0063] S500. According to the list of path node information and the semantic information, determine whether there is interference in the connection between adjacent path nodes. If so, add transfer path nodes. If not, generate the final construction path.

[0064] Through obtaining the semantic information of the construction area to be constructed, intelligently screening construction objects, and automatically matching appropriate construction types and mechanical parameters, the above embodiments of the present application improve the accuracy and efficiency of task planning. At the same time, it can automatically generate path node information and a specific construction action sequence, and add transfer path nodes when necessary to avoid interference, thereby ensuring the smoothness and safety of the construction process, and significantly improving the automated construction ability and operation efficiency of the construction robot.

[0065] Specifically, first, by obtaining the semantic information of the construction area to be constructed, screen out the objects that need to be constructed; then, according to the screened construction objects, search and match appropriate construction types and mechanical construction parameters in a preset process library, and generate a list of construction operation node information; then, combine the semantic information and the list of construction operation node information to generate a list of path node information; subsequently, in a preset trajectory action library, match appropriate process actions for the path nodes to form a specific construction action sequence. Finally, according to the path node information and the semantic information, determine whether there is interference in the connection between adjacent path nodes. If there is interference, add transfer path nodes to avoid conflicts; if there is no interference, directly generate the final construction path for the construction robot to execute.

[0066] Among them, the construction action sequence in the task planning of the construction robot decomposes complex tasks into an instruction set of ordered actions, ensuring that the robot can efficiently and accurately complete construction tasks, can be dynamically adjusted according to the real-time environment, supports multi-robot collaborative operation, and at the same time transforms the design intention into actual operation to achieve the effective connection between design and construction. It is the key for the construction robot to achieve automated construction and is of great significance for improving construction efficiency, quality and flexibility.

[0067] Refer to Figure 2As shown, in some possible embodiments, obtaining the semantic information of the area to be constructed and screening the objects to be constructed includes: obtaining the image or point cloud data of the area to be constructed, extracting the information in the image or point cloud data, and generating the semantic information of the area to be constructed; according to the semantic information of the area to be constructed, judging the normal vector orientation of different information in the semantic information, determining the construction object, calculating the depth of the construction object, and determining the distance between the construction object and the building wall; screening out the objects to be constructed according to the distance.

[0068] Among them, in the embodiments of the present application, the existence of holes and crossbeams in the construction object results in the non-uniformity of the semantic information corresponding to each wall, which will present quadrilaterals, hexagons or even octagons. If these polygons are not pre-processed by cropping and a set of specialized algorithms are used to handle each situation, the final code volume will be very cumbersome and redundant, and the program complexity will also increase accordingly.

[0069] Specifically, first, obtain the image or point cloud data of the area to be constructed through an image acquisition device or a lidar system, including structures such as holes and crossbeams in the construction area. Then, use image processing technology or point cloud analysis algorithms to extract key information from the data, such as the shape, size, and position of the holes, and generate the semantic information of the area to be constructed, that is, each object or area is assigned a specific semantic label, such as "entrance hall", "ceiling", "wall surface", etc. Subsequently, according to the generated semantic information, calculate the normal vector orientation of different structures. Among them, the normal vector is a mathematical tool for describing the surface direction of an object. By analyzing the orientation of the normal vector, determine the construction object that needs to perform construction operations in the area.

[0070] The embodiments of the present application can quickly and accurately understand the overall situation and specific details of the construction area by automatically obtaining and analyzing the image or point cloud data of the area to be constructed and generating detailed semantic information, and formulate a more scientific and reasonable construction plan. At the same time, determining the construction object according to the normal vector orientation of different information in the semantic information can ensure the pertinence and effectiveness of the construction operation, avoid unnecessary misoperations and resource waste, and improve construction safety.

[0071] Among them, in the process of identifying door and window holes, a specific implementation method is to obtain the geometric contour of the construction area by obtaining the image or point cloud data of the area to be constructed, and identify the door and window holes through the geometric contour, specifically including:

[0072] S111, receiving the geometric contour of the detected wall;

[0073] S112, starting from the lower left corner of the wall, Figure 11 the red dots in the figure, set the traversal starting points in the right direction and the upward direction parallel to the wall surface, Figure 11The blue dots and red dots are the points traversing to the right and up respectively; at each starting point, a ray is emitted in the opposite direction of the wall normal vector, Figure 11 The purple arrows in the figure indicate the emitted rays.

[0074] S113, monitor the event of no collision:

[0075] The event of no collision here is deduced by reversing the collision event. For example: Given the propagation speed of the ray and the distance to be detected, the approximate time required for the ray to have a collision event can be calculated. Then it is defined that if no collision event is still monitored within this time period, it is judged as an event of no collision.

[0076] If a collision occurs and no hole is detected, return to continue traversing;

[0077] If no collision occurs, the coordinate point of no collision is used as the lower left coordinate of the hole;

[0078] S114, starting from the lower left corner of the hole, set traversing starting points in the right direction and up direction parallel to the wall, and emit rays in the opposite direction of the wall normal vector at each starting point;

[0079] S115, monitor the event of collision:

[0080] If no collision occurs, it means the hole boundary is not detected, and continue traversing;

[0081] If a collision occurs, the hole boundary is detected and appended to the semantic information of the current wall.

[0082] In some embodiments, it is further determined whether the upper and lower limits of the hole are flush with the wall. If they are flush, the hole is the area not detected in the contour recognition stage, update the wall contour information, and delete the hole information; if not flush, further detect the hole boundary and append it to the semantic information of the current wall.

[0083] On the basis of the obtained hole semantic information, further identify bay windows and crossbeams, as shown in Figure 10 and Figure 12 (a)-(e) of the figure. Specifically:

[0084] S115, receive the geometric contour of the hole;

[0085] S116, taking the lower left corner of the hole as the reference point, Figure 12 the red dots in the figure, offset a distance a to the left, and set the traversing starting point in the right direction, Figure 12 the blue dots in the figure, as the dot matrix for emitting rays (purple arrows);

[0086] S117, monitor the event of no collision:

[0087] If a collision occurs, it is detected that there is a bay window, and the traversal continues;

[0088] If no collision occurs, the bay window depth information ( Figure 12 the purple double arrow in Figure (e)) is obtained;

[0089] S118, determine whether the bay window depth information is less than the set process parameter:

[0090] If it is greater, it indicates that there is a bay window at this hole, and the relevant semantics are updated to the current wall surface.

[0091] If the bay window depth information is less than the set process parameter, it indicates that there is no bay window at this hole, and further determine whether there is a cross beam, as shown in Figure 11 and Figure 13 Figures (a)-(e). Specifically:

[0092] S119, taking the lower left corner of the hole as the reference point, Figure 13 the red point in, offset a distance a to the left, and set the traversal starting point in the right direction, Figure 13 the blue point in, as the dot matrix for emitting rays (purple arrows);

[0093] S120, listen for the event of no collision:

[0094] If a collision occurs and it is detected that there is a cross beam, the traversal continues;

[0095] If no collision occurs, obtain the cross beam protrusion distance;

[0096] S121, determine whether the cross beam protrusion distance ( Figure 13 the purple double arrow in Figure (e)) is less than the set process parameter:

[0097] If it is greater, there is a cross beam at this hole, and the relevant semantics are updated to the wall surface;

[0098] If it is less, there is no cross beam at this hole, and it is just a hole.

[0099] In the above embodiments, the semantic information of the object to be recognized is further supplemented, including the filtering, merging, and cropping of holes, as well as the recognition of features such as cross beams and bay windows. This makes the semantic information more rich and detailed, and improves the accuracy of task planning.

[0100] In the above several embodiments, the ray emission function in the 3D engine is used, combined with an efficient collision detection algorithm, to quickly obtain the coordinates and normal directions of the collision points. This greatly reduces the amount of calculation and improves the efficiency of information acquisition.

[0101] Among them, the depth of the construction object is calculated to determine the distance between the construction object and the building wall surface; according to the distance, the objects to be constructed are screened out, specifically including: obtaining the construction object, dividing the surface to be planned into multiple sub-surfaces according to the door and window holes and beam information, and performing depth calculation to obtain the distance between the construction object and the building wall surface. If the distance from the building wall surface is greater than the preset threshold of the porch, it is judged as a wall surface in the system. A complete wall is judged as an ordinary wall surface or a porch wall surface due to different depth information.

[0102] Among them, as Figure 9 shown, the opposite wall distance 1 and the opposite wall distance 2 in the figure refer to the depth values of the building wall surface.

[0103] Refer to the appendix Figure 3 shown, in some possible embodiments, according to the screened objects to be constructed, suitable construction types and mechanical construction parameters are matched from a preset process library to generate a list of construction operation node information, including: obtaining the type of the object to be constructed and determining the dimension information of the construction area of the object to be constructed; according to the dimension information, judging whether the dimension information is greater than a set threshold. If so, multiple construction areas are divided according to the size of the construction area. If not, the fixed dimension process rule is adopted; suitable construction types and mechanical construction parameters are matched from the preset process library to generate a list of construction operation node information.

[0104] By intelligently matching the construction type and mechanical construction parameters, the construction efficiency is effectively improved. At the same time, according to the type and dimension information of the construction object, the most suitable construction plan is automatically selected from the preset process library, avoiding the errors and inefficiencies that may be brought by manual selection, thus ensuring the construction quality and progress.

[0105] Refer to Figure 4 shown, specifically, in the processing process, first, the type of the construction object is obtained, and its dimension information is determined. Then, according to these dimension information, the system judges whether it exceeds the set threshold. If the dimension exceeds the threshold, the system will calculate the number of construction areas to be divided according to the size of the planned area to reasonably arrange the construction. If the dimension does not exceed the threshold, the fixed dimension process rule is adopted for construction. Finally, the system will automatically match the most suitable construction type and mechanical construction parameters from the preset process library and generate a list of construction operation node information for the construction robot to execute.

[0106] Among them, for the construction area, the construction objects are actually the wall surface and the ceiling, which have become relatively independent quadrilaterals after pretreatment. For the quadrilateral, the dimension information is the length and the width.

[0107] Specifically, the ceiling has an additional ceiling height, and the wall has an attached bay depth dimension. Bay depth is a very common technical term in the construction field and is also a long-term construction indicator, belonging to the national standard information.

[0108] Making decisions on the process type based on dimensional information is a complete and self-developed mapping function. For example, when the wall width is between 700 and 1100 mm, process A is adopted; when it is between 1500 and 3000 mm, process B is adopted.

[0109] In this application, the fixed-size process rules are the preset process library. Based on this, for example: 700 - 1100 mm and 1500 - 3000 mm are both fixed sizes, that is, at this size, only 1 type of process can be corresponding. The unfixed size can be 3000 - +∞. Therefore, I think the expression here needs to be modified. It cannot be called the process size here, but should be called the size constraint of the fixed interval.

[0110] Among them, the preset process library includes the corresponding relationship between processes and construction materials, as well as the matching construction parameters, specifically including the moving speed of the spray gun, the distance between the spray gun and the wall, the effective width of the spray pattern, etc. For example, in a certain construction scenario, process A will be executed, then the material α (material brand, model, etc.) applicable to process A recorded in the process library should be used, and the corresponding parameters should be extracted.

[0111] In some specific embodiments, the construction operation node information list includes the operation node information of multiple construction areas. The set threshold is the size of the construction robot body.

[0112] Specifically, for the construction robot body, there are the length and width of the body chassis, the height of the body, the safe rotation diameter when the robotic arm extends to the maximum posture, and the safe rotation diameter when the construction robot is in the contracted posture. These values will all be used as the threshold reference for subsequent decisions. For example, the body width of a normal adult is 40 - 50 cm, then 40 cm is used as the common sense threshold. A normal adult cannot pass horizontally through a narrow aisle with a width less than 40 cm, and it is judged that "it is necessary to pass sideways".

[0113] When generating the construction operation node information list, it is necessary to create a list of the operation node information for each construction area according to each construction area. Or use each construction area to calculate several operation node information and then integrate them into a list. For example, the list of operation node information has 100 nodes, and nodes 1 - 20 belong to area A, 20 - 50 belong to area B, and so on. Corresponding labels will be set in the operation nodes for marking.

[0114] Refer to Figure 5As shown, in some possible embodiments, a list of path node information is generated according to semantic information and a list of construction operation node information, including: obtaining the list of construction operation node information and semantic information, and generating information on multiple path nodes corresponding to the operation node information of multiple construction areas; according to the information on multiple path nodes, allocating the corresponding relationship between operation nodes and path nodes, generating three-dimensional coordinate information of multiple path nodes, and forming a list of multiple path node information.

[0115] Specifically, obtain the list of construction operation nodes, set several path node information in combination with information such as mechanical type, process type, and mechanical parameters, allocate the corresponding relationship between operation nodes and path nodes, set the three-dimensional coordinate information of path nodes, and form a list of path node information.

[0116] Refer to Figure 6 As shown, in some possible embodiments, according to the list of path node information, within a preset trajectory action library, actions of a suitable process are matched to generate a specific construction action sequence, including: obtaining the list of path node information, combining semantic information, construction type, and the mechanical construction parameters, and selecting trajectory actions matching the list of path node information within the preset trajectory action library; according to the trajectory actions, fine-tuning the construction intervals in the operation node information to generate a specific construction action sequence.

[0117] A specific implementation manner is, for example: currently, there is a wall surface in a construction area, the width of the wall surface is 4m, the height is 3m, and the bay depth is 4m and 5m respectively. These data are the wall surface semantic information that needs to be input from the interface initially.

[0118] The current safety rotation diameter of the construction robot when the robotic arm reaches the maximum posture is 2.6m. This value is the threshold. It is judged that 2.6m is less than the depth of 4m of the current construction area, and the construction robot is allowed to construct in this area, and the construction type is process A.

[0119] Under process A, the amplitude of the construction robot's single operation is 1m. Therefore, 4 operation nodes need to be set to completely cover the 4m-wide wall surface, and the relative positions of the path nodes (the distance from the leftmost side of the wall surface) are 0.5m, 1.5m, 2.5m, and 3.5m in sequence.

[0120] For each path node, the trajectory actions of process A from 0 to 3m are executed in sequence. The relative position information included in the above four path nodes, for the required process A and the starting point (0 to 3m), form a list with 4 node information.

[0121] Of course, the above are only some specific implementation manners of the present application. In other embodiments, other parameter values and construction action sequences may also be used.

[0122] Reference Figure 7 As shown, in some possible embodiments, based on the path node information list and semantic information, it is determined whether there is interference in the connection between adjacent path nodes. If so, transfer path nodes are added. If not, the final construction path is generated, including: obtaining the path node information list and semantic information, and converting the semantic information into map information; based on the path node information list and map information, obtaining the spatial relationship between two adjacent path node information and the map information; connecting two adjacent path node information, and determining whether there is interference with the map information. If so, transfer path nodes are added to the specific construction action sequence. If not, the final construction path is generated.

[0123] Specifically, first, the path node information list and relevant semantic information are obtained, the semantic information is converted into map information, and the spatial relationship is analyzed; then, based on the path node information list and map information, the spatial relationship between two adjacent path nodes and map elements is determined; then, the two adjacent path nodes are connected, and it is checked whether this connection interferes with obstacles or other non-traversable areas in the map. If interference is detected, one or more transfer path nodes are automatically inserted into the construction action sequence to avoid obstacles. If there is no interference, the final construction path is directly generated.

[0124] Reference Figure 8 As shown, by intelligently analyzing the spatial relationship between path nodes and map information, potential interference problems in the construction path are effectively avoided. When interference is detected, transfer path nodes are automatically added to optimize the construction path, ensuring the smooth progress of construction, improving construction efficiency, and reducing the risk of construction delays and increased costs caused by improper path planning.

[0125] Based on the same technical concept, in another embodiment of the present application, a task planning system for a construction robot is provided, including: a construction object identification module, configured to receive the semantic information of the area to be constructed, identify the construction object, and determine the type of the construction object; a process classification module, configured to match the appropriate construction type and mechanical construction parameters from a preset process library according to the selected construction object, and generate a list of construction operation node information; a path node module, configured to generate a list of path node information according to the semantic information and the list of construction operation node information; a decision-making trajectory action module, configured to match the actions of the appropriate process in a preset trajectory action library according to the list of path node information, and generate a specific construction action sequence; a target planning path determination module, configured to determine whether there is interference in the connection between adjacent path nodes according to the list of path node information and the semantic information. If so, transfer path nodes are added. If not, the final construction path is generated.

[0126] Specifically, first, the construction object identification module receives the semantic information of the area to be constructed, identifies the construction object, determines whether the object is a ceiling, a wall, a porch, etc., and transmits the identification result to the process classification module.

[0127] Among them, the data obtained is the semantic information of the area to be constructed, including images, point cloud data, etc.

[0128] Next, the process classification module matches the appropriate construction type and mechanical construction parameters from the preset process library according to the type of the construction object. The matched construction type and parameters are transmitted to the path node module. Among them, the data obtained is the type of the construction object.

[0129] Then, the path node module obtains the construction type and mechanical construction parameters, and the wall semantic information, sets the path nodes to ensure the rationality and integrity of the construction path, and transmits the path node information to the decision trajectory action module.

[0130] Next, the decision trajectory action module obtains the path node information and the construction type, matches the appropriate actions of the current process in the preset trajectory action library, generates a specific construction action sequence, and transmits the generated construction action sequence to the target planning path determination module.

[0131] Finally, the target planning path determination module obtains the construction action sequence and the path node information, concatenates the information of all path nodes, appropriately adds transfer stations to ensure reasonable obstacle avoidance and safe postures, and generates the final construction path. The final construction path is transmitted to the construction robot execution module.

[0132] Through the collaborative work of the above modules and their sub-modules, the construction robot task planning system of the present application can achieve efficient and accurate task planning and path generation, ensuring the safety and efficiency of the construction robot in a complex construction environment.

[0133] It should be noted that the modules in the construction robot task planning system provided in the above embodiments of the present application correspond to the steps of the construction robot task planning method in any of the above embodiments. Those skilled in the art can refer to the step features of the construction robot task planning method to implement the corresponding modules in the construction robot task planning system, which will not be elaborated here.

[0134] In another embodiment of the present application, a construction robot is provided, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the construction robot task planning method.

[0135] In another embodiment of the present application, a storage medium is provided. The storage medium is a computer-readable storage medium, and a computer program is stored thereon. When the computer program is executed by a processor, it implements the task planning method of a construction robot.

[0136] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0138] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0140] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0141] The specific embodiments of the present application have been described above. It should be understood that the present application is not limited to the above specific embodiments, and those skilled in the art can make various deformations or modifications within the scope of the claims, which does not affect the essence of the present application. The above preferred features can be used in any combination without conflict.

Claims

1. A task planning method for a construction robot, characterized in that: include: Obtain semantic information of the area to be constructed and filter the objects to be constructed; According to the screened objects to be constructed, suitable construction types and mechanical construction parameters are matched from a preset process library to generate a list of construction operation node information; Generate a path node information list according to the semantic information and the construction operation node information list; According to the path node information list, in the preset trajectory action library, the action of the appropriate process is matched to generate a specific construction action sequence; According to the path node information list and the semantic information, it is determined whether the connection between adjacent path nodes interferes with each other. If so, a transit path node is added; if not, a final construction path is generated.

2. A task planning method for a construction robot according to claim 1, characterized in that: The obtaining of semantic information of the area to be constructed and screening of construction objects includes: Acquire an image or point cloud data of the area to be constructed, extract information from the image or the point cloud data, and generate semantic information of the area to be constructed; According to the semantic information of the area to be constructed, the normal vector orientations of different information in the semantic information are judged to determine the construction object; Calculating the depth of the construction object to determine the distance between the construction object and the building wall; According to the distance, the objects to be constructed are screened out.

3. The task planning method of a construction robot according to claim 1, characterized in that: According to the screened objects to be constructed, suitable construction types and mechanical construction parameters are matched from a preset process library to generate a construction operation node information list, including: Acquire the screened object to be constructed, and determine the size information of the construction area of ​​the object to be constructed; According to the size information, determine whether the size information is greater than a set threshold, if so, divide the construction area into multiple construction areas according to the size of the construction area, if not, fix the size process rule; Match the appropriate construction type and mechanical construction parameters from the preset process library to generate a list of construction operation node information.

4. A task planning method for a construction robot according to claim 3, characterized in that: The construction operation node information list includes operation node information of multiple construction areas; The set threshold is the size of the construction robot body.

5. A task planning method for a construction robot according to claim 4, characterized in that: The generating a path node information list according to the semantic information and the construction operation node information list includes: Acquire the construction operation node information list and the semantic information, and generate information of multiple path nodes corresponding to the operation node information of the multiple construction areas; According to the information of the multiple path nodes, the corresponding relationship between the operation nodes and the path nodes is assigned, the three-dimensional coordinate information of the multiple path nodes is generated, and a list of the multiple path node information is formed.

6. The task planning method of a construction robot according to claim 1, characterized in that: According to the path node information list, in the preset trajectory action library, matching the action of the appropriate process to generate a specific construction action sequence includes: Obtain a list of path node information, and select a trajectory action that matches the list of path node information in a preset trajectory action library in combination with the semantic information, the construction type, and the mechanical construction parameters; According to the trajectory action, the construction interval in the operation node information is fine-tuned to generate a specific construction action sequence.

7. The task planning method of a construction robot according to claim 1, characterized in that: The step of determining whether there is interference between the lines of adjacent path nodes according to the path node information list and the semantic information, and if so, adding a transit path node, and if not, generating a final construction path, includes: Acquire the path node information list and the semantic information, and convert the semantic information into map information; According to the path node information list and the map information, a spatial relationship between two adjacent path node information and the map information is obtained; The information of two adjacent path nodes is connected to determine whether there is interference with the map information. If so, a transit path node is added to the specific construction action sequence. If not, a final construction path is generated.

8. A task planning system for a construction robot, characterized in that: include: The construction object identification module is used to receive the semantic information of the area to be constructed, identify the construction object, and determine the type of the construction object; The process classification module is used to match the appropriate construction type and mechanical construction parameters from the preset process library according to the screened construction objects, and generate a list of construction operation node information; A path node module, used to generate a path node information list according to the semantic information and the construction operation node information list; A decision trajectory action module is used to match the actions of the appropriate process in a preset trajectory action library according to the path node information list to generate a specific construction action sequence; The target planning path determination module is used to determine whether there is interference between the connections between adjacent path nodes based on the path node information list and the semantic information. If so, a transit path node is added; if not, a final construction path is generated.

9. A construction robot, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the task planning method for a construction robot as claimed in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the task planning method for a construction robot as described in any one of claims 1 to 7 is implemented.

Images