An intelligent assembly robot system based on inertial SLAM

Abstract

The present invention provides an intelligent assembly robot system based on inertial SLAM, belonging to the field of assembly control technology. The system comprises: a monitoring module for acquiring assembly data, assembly station data, inertial data, and assembly scene data; an analysis module for determining candidate station vectors and candidate path data for a product to be assembled; a determination module for determining a target station and target path for the product to be assembled; an assembly module for delivering the product to be assembled to a target station based on the target path, completing assembly of the product at the target station and marking it as completed; and a delivery module for delivering the assembled product to its destination. The system can flexibly select the target station and path, ensuring the efficiency and accuracy of the assembly process, improving the robot's adaptability in complex environments, increasing production efficiency, realizing intelligent assembly, and enhancing assembly efficiency, thereby rapidly responding to production needs.

Description

Technical Field

[0001] The present invention relates to the technical field of assembly control, and in particular to an intelligent assembly robot system based on inertial SLAM. Background Art

[0002] As the manufacturing industry continues to become more automated and intelligent, the use of intelligent robots on production lines is becoming increasingly widespread, especially in the assembly process. Traditional industrial assembly typically relies on fixed assembly lines and workers performing manual or semi-automated operations. The assembly process is often limited by manual labor and static production lines. Frequent changes in factors such as the relative position between workstations and load conditions result in low assembly efficiency, large errors, and poor production line adaptability. Furthermore, traditional assembly robots often lack flexibility in path planning, making it difficult to effectively adapt to environmental changes. Most systems rely on external positioning systems (such as visual positioning or ground sensors), which are susceptible to environmental interference and limited positioning accuracy.

[0003] Therefore, the present invention provides an intelligent assembly robot system based on inertial SLAM. Summary of the Invention

[0004] The present invention provides an intelligent assembly robot system based on inertial SLAM. By analyzing assembly data, assembly station data, inertial data, and assembly scenario data, the system determines candidate station vectors, candidate path data, target station, and target path for the product to be assembled. The assembly robot then delivers the product to the target station, completes product assembly, and delivers it to its destination. The system can flexibly select target stations and paths, ensuring the efficiency and accuracy of the assembly process, improving the robot's adaptability in complex environments, and increasing production efficiency. This system enables intelligent assembly, enhances assembly efficiency, and rapidly responds to production needs.

[0005] The present invention provides an intelligent assembly robot system based on inertial SLAM, comprising:

[0006] Monitoring module: Acquires assembly data of products to be assembled, obtains assembly station data of the assembly factory, collects inertial data of assembly robots, and assembly scene data;

[0007] Analysis module: Analyzes assembly data, assembly station data, inertia data, and assembly scenario data to determine candidate station vectors and candidate path data for the product to be assembled;

[0008] Determination module: Determines the target workstation and target path of the product to be assembled based on the assembly workstation data, inertia data, candidate workstation vectors of the product to be assembled, and candidate path data;

[0009] Assembly module: The assembly robot delivers the product to be assembled to the target workstation based on the target path of the product to be assembled, completes the product assembly at the target workstation and marks the assembly as complete;

[0010] Delivery module: The assembly robot delivers the assembled products to the product destination based on assembly data, inertia data, and assembly scene data.

[0011] According to the present invention, an intelligent assembly robot system based on inertial SLAM and a monitoring module are provided, comprising:

[0012] Assembly data unit: This unit acquires assembly data of products to be assembled based on the sensor network of the production line. The assembly data includes the end-of-line location of the products to be assembled, the assembly task, and the product destination.

[0013] Inertial data unit: The assembly robot's lidar and visual sensor collect assembly scene data, and the assembly robot's inertial measurement unit collects inertial data;

[0014] Assembly station data unit: The assembly station data includes assembly station sub-data of multiple assembly stations, wherein the assembly station sub-data includes station number, station location, executable assembly tasks, real-time station load, maximum station load and station yield.

[0015] According to the present invention, an intelligent assembly robot system based on inertial SLAM and an analysis module are provided, comprising:

[0016] A candidate workstation vector unit extracts assembly tasks from the assembly data of the product to be assembled, and determines a candidate workstation vector for the product to be assembled based on the extracted assembly tasks and the executable assembly tasks of each assembly station sub-data in the assembly workstation data, wherein the candidate workstation vector includes multiple assembly stations;

[0017] An obstacle data unit: determining obstacle data for each assembly station in the candidate station vector of the product to be assembled based on the production line end position in the assembly data of the product to be assembled, the station position in the assembly station sub-data of each assembly station in the candidate station vector, and the assembly scene data, wherein the obstacle data includes obstacle sub-data of multiple obstacles, and the obstacle sub-data includes obstacle position, obstacle shape, and obstacle size of the obstacle;

[0018] Detour unit: Based on the assembly scene data and each obstacle sub-data in the obstacle data, determine the detour path, detour distance and maximum detour speed of each obstacle at each assembly station in the candidate station vector of the product to be assembled;

[0019] Assembly path unit: Determines the assembly path for each assembly station in the candidate station vector of the product to be assembled based on the end-of-line position in the assembly data of the product to be assembled, the station position in the assembly station sub-data of each assembly station in the candidate station vector, and the detour paths of all obstacles in the obstacle data;

[0020] Candidate path data unit: determines candidate path data of the product to be assembled based on the assembly paths of all assembly stations in the candidate station vector of the product to be assembled.

[0021] According to an inertial SLAM-based intelligent assembly robot system provided by the present invention, a determination module includes:

[0022] Historical assembly data unit: obtains historical assembly data of each assembly station in the candidate station vector of the product to be assembled by the assembly robot, wherein the historical assembly data includes the assembly time, station load and assembly results of multiple historical assemblies;

[0023] Path evaluation value unit: calculates the path evaluation value of each assembly station in the candidate station vector of the product to be assembled based on the assembly station data, inertia data, historical assembly data of each assembly station in the candidate station vector of the product to be assembled, and the assembly path of each assembly station in the candidate path data;

[0024] Target station unit: From the path evaluation values of all assembly stations in the candidate station vector of the product to be assembled, the assembly station corresponding to the smallest path evaluation value is selected as the target station for the product to be assembled;

[0025] Target path unit: determines the assembly path corresponding to the target workstation of the product to be assembled as the target path of the product to be assembled.

[0026] According to the present invention, an intelligent assembly robot system based on inertial SLAM, a path evaluation value unit includes:

[0027] ;

[0028] ;

[0029] ;

[0030] in, represents the path evaluation value of the i-th assembly station in the candidate station vector of the product to be assembled, represents the moving time of the product to be assembled to the i-th assembly station in its corresponding candidate station vector, represents the predicted assembly time of the product to be assembled at the i-th assembly station in its corresponding candidate station vector; is the unit time; represents the energy consumption of the i-th assembly station in the candidate station vector of the product to be assembled; is the unit energy consumption; represents the real-time station load of the i-th assembly station in the candidate station vector of the product to be assembled, represents the maximum station load of the i-th assembly station in the candidate station vector of the product to be assembled; represents the assembly load of the i-th assembly station in the candidate station vector of the product to be assembled, represents the station yield of the i-th assembly station in the candidate station vector of the product to be assembled, represents the moving distance of the product to be assembled from its corresponding end position of the production line to the position of the i-th assembly station in the candidate station vector of the product to be assembled, The moving speed in the dynamic inertial data representing the inertial data of the assembly robot, represents the moving acceleration in the dynamic inertial data of the assembly robot’s inertial data, iN1 represents the number of obstacles at the i-th assembly station in the candidate station vector of the product to be assembled, represents the maximum circumventing speed of the jth obstacle of the i-th assembly station in the candidate station vector of the product to be assembled, represents the detour path length of the jth obstacle of the i-th assembly station in the candidate station vector of the product to be assembled, 、 They represent the moving power and assembly power of the assembly robot respectively, represents the time weight, represents the energy consumption weight, represents a load penalty factor of one; is a natural constant.

[0031] According to an intelligent assembly robot system based on inertial SLAM provided by the present invention, when the assembly task performed by each assembly station is fixed, the predicted assembly time of the i-th assembly station in its corresponding candidate station vector for the product to be assembled is calculated based on the following formula:

[0032] ;

[0033] in, represents a load penalty factor of two; Represents the number of historical assemblies in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, represents the assembly time of a single product of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, represents the time decay factor, represents the assembly result of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, Indicates that the assembly failed. represents the penalty factor for assembly results, Represents the historical station load of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled; The total number of products to be assembled.

[0034] According to the present invention, an intelligent assembly robot system based on inertial SLAM is provided, and the assembly module includes:

[0035] Real-time Positioning Unit: The assembly robot moves along the target path of the product to be assembled, and the real-time position of the assembly robot is obtained based on the assembly robot's lidar and inertial measurement unit;

[0036] Detection unit: Based on the target path of the product to be assembled and the real-time position of the assembly robot, it detects whether path deviation occurs. If so, the assembly robot first moves toward the target path, returns to the target path, and then continues to move along the target path.

[0037] According to the present invention, an intelligent assembly robot system based on inertial SLAM, the assembly module further includes:

[0038] Assembly instruction unit: After the assembly robot arrives at the target workstation, the assembly model generates assembly instructions for the product to be assembled based on the assembly tasks in the assembly data of the product to be assembled and the inertia data of the assembly robot;

[0039] First marking unit: The assembly robot completes product assembly at the target workstation based on the assembly instructions of the product to be assembled, and performs an assembly completion mark on the assembled product to be assembled.

[0040] According to the present invention, an intelligent assembly robot system based on inertial SLAM, a delivery module, includes:

[0041] Destination path unit: determines the destination path of the product to be assembled with the assembly completion mark based on the product destination, target workstation and assembly scene data in the assembly task of the product to be assembled with the assembly completion mark;

[0042] Second marking unit: The assembly robot delivers the products to be assembled with the assembly completion mark to the product destination along the destination path of the products to be assembled with the assembly completion mark, and performs a delivery mark on the products to be assembled at the destination.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] By analyzing assembly data, assembly station data, inertia data, and assembly scenario data, the robot determines candidate station vectors, candidate path data, target station, and target path for the product to be assembled. The robot then delivers the product to the target station, completes assembly, and delivers it to its destination. Flexible selection of target stations and paths ensures efficient and accurate assembly, improves the robot's adaptability in complex environments, enhances production efficiency, enables intelligent assembly, and improves assembly efficiency, enabling rapid response to production needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 This is a structural diagram of an intelligent assembly robot system based on inertial SLAM provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0048] Example 1:

[0049] The embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, such as Figure 1 As shown, including:

[0050] Monitoring module: Acquires assembly data of products to be assembled, obtains assembly station data of the assembly factory, collects inertial data of assembly robots, and assembly scene data;

[0051] Analysis module: Analyzes assembly data, assembly station data, inertia data, and assembly scenario data to determine candidate station vectors and candidate path data for the product to be assembled;

[0052] Determination module: Determines the target workstation and target path of the product to be assembled based on the assembly workstation data, inertia data, candidate workstation vectors of the product to be assembled, and candidate path data;

[0053] Assembly module: The assembly robot delivers the product to be assembled to the target workstation based on the target path of the product to be assembled, completes the product assembly at the target workstation and marks the assembly as complete;

[0054] Delivery module: The assembly robot delivers the assembled products to the product destination based on assembly data, inertia data, and assembly scene data.

[0055] In this embodiment, inertial data comes from the inertial measurement unit (IMU) of the assembly robot, such as acceleration, angular velocity, etc., which is used to determine the motion state of the robot.

[0056] In this embodiment, the assembly scene data includes information such as obstacles, spatial layout, and physical locations of workstations in the assembly factory.

[0057] In this embodiment, a plurality of candidate workstation vectors are determined, that is, a plurality of workstations suitable for performing the assembly task are selected according to the requirements of the assembly task.

[0058] In this embodiment, multiple candidate path data are determined, that is, path selection for the robot to reach each target workstation from the current position, to provide flexibility for the execution of the assembly task.

[0059] In this embodiment, the target workstation and path for ultimately executing the assembly task are determined based on the data generated by the analysis module.

[0060] In this embodiment, when the assembly robot arrives at the target workstation, it starts to perform the assembly task; completes the assembly task of the product to be assembled at the target workstation and performs specific process operations; after the assembly task is completed, the assembled product is marked to indicate that the product has been assembled and can enter the next stage.

[0061] In this embodiment, after the assembly task is completed, the robot delivers the assembled product to the designated destination according to the task requirements. This module combines assembly data, inertial data, and assembly scene data to determine the most appropriate delivery path and executes path planning to ensure the product arrives smoothly and accurately at the destination.

[0062] The beneficial effects of the above technical solution include: by analyzing assembly data, assembly station data, inertia data, and assembly scenario data, candidate station vectors, candidate path data, target station, and target path for the product to be assembled are determined. The assembly robot then delivers the product to the target station, completes product assembly, and delivers it to its destination. Flexible selection of target stations and paths ensures efficient and accurate assembly, improves the robot's adaptability in complex environments, enhances production efficiency, enables intelligent assembly, improves assembly efficiency, and rapidly responds to production needs.

[0063] Example 2:

[0064] The embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, a monitoring module, including:

[0065] Assembly data unit: This unit acquires assembly data of products to be assembled based on the sensor network of the production line. The assembly data includes the end-of-line location of the products to be assembled, the assembly task, and the product destination.

[0066] Inertial data unit: The assembly robot's lidar and visual sensor collect assembly scene data, and the assembly robot's inertial measurement unit collects inertial data;

[0067] Assembly station data unit: The assembly station data includes assembly station sub-data of multiple assembly stations, wherein the assembly station sub-data includes station number, station location, executable assembly tasks, real-time station load, maximum station load and station yield.

[0068] In this embodiment, the production line will go through multiple processes from the input of raw materials to the completion of preliminary processing. When the product completes the main production steps preset by the production line, it will reach the end of the line. The end position of the production line refers to the last position of the product to be assembled on the production line.

[0069] In this embodiment, the assembly task includes the specific steps, required tools, and operation sequence for product assembly. This information helps the robot complete efficient and accurate assembly.

[0070] In this embodiment, the product destination refers to the target location where the product needs to be transported after assembly, ensuring that the assembled product can be correctly delivered to the warehouse or other designated area. The product destination can be a specific shelf location in the warehouse or the location of a visiting guest.

[0071] In this embodiment, lidar and vision sensors are used to collect assembly scene data to help the robot identify objects or obstacles in the environment, and an inertial measurement unit (IMU) is used to obtain the robot's motion information (such as position, speed, and posture) during the assembly process.

[0072] In this embodiment, the workstation number represents a unique identifier for each assembly workstation, which is used to distinguish different workstations; the workstation location represents the physical location of the workstation, which helps determine the assembly path; and the executable assembly task represents the type of assembly task that each workstation can perform, which helps to select a suitable workstation to handle different assembly requirements.

[0073] In this embodiment, the real-time workstation load represents the current workload of the corresponding assembly workstation, reflecting the busyness of the workstation.

[0074] In this embodiment, the maximum station load indicates the maximum load that the assembly station can bear. Excessive load may affect assembly quality or speed.

[0075] In this embodiment, the workstation yield is an indicator to measure the assembly quality of the workstation. Workstations with low yield may need to be avoided or optimized to improve overall production efficiency and product quality.

[0076] The beneficial effects of the above technical solution are: obtaining assembly data of the product to be assembled, obtaining assembly station data of the assembly factory, collecting inertial data of the assembly robot and assembly scene data, which can provide data basis for determining candidate station vectors and candidate path data of the product to be assembled.

[0077] Example 3:

[0078] The embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, an analysis module, including:

[0079] A candidate workstation vector unit extracts assembly tasks from the assembly data of the product to be assembled, and determines a candidate workstation vector for the product to be assembled based on the extracted assembly tasks and the executable assembly tasks of each assembly station sub-data in the assembly workstation data, wherein the candidate workstation vector includes multiple assembly stations;

[0080] An obstacle data unit: determining obstacle data for each assembly station in the candidate station vector of the product to be assembled based on the production line end position in the assembly data of the product to be assembled, the station position in the assembly station sub-data of each assembly station in the candidate station vector, and the assembly scene data, wherein the obstacle data includes obstacle sub-data of multiple obstacles, and the obstacle sub-data includes obstacle position, obstacle shape, and obstacle size of the obstacle;

[0081] Detour unit: Based on the assembly scene data and each obstacle sub-data in the obstacle data, determine the detour path, detour distance and maximum detour speed of each obstacle at each assembly station in the candidate station vector of the product to be assembled;

[0082] Assembly path unit: Determines the assembly path for each assembly station in the candidate station vector of the product to be assembled based on the end-of-line position in the assembly data of the product to be assembled, the station position in the assembly station sub-data of each assembly station in the candidate station vector, and the detour paths of all obstacles in the obstacle data;

[0083] Candidate path data unit: determines candidate path data of the product to be assembled based on the assembly paths of all assembly stations in the candidate station vector of the product to be assembled.

[0084] In this embodiment, by extracting the assembly tasks of the products to be assembled, multiple assembly stations that can perform the tasks are screened out from multiple assembly stations based on the executable assembly tasks of each assembly station, and finally a candidate station vector is generated, which contains multiple suitable assembly stations for subsequent selection.

[0085] In this embodiment, the executable assembly tasks represent specific assembly operations that can be completed at each workstation.

[0086] In this embodiment, possible obstacles on the path from the production line end position to the workstation position in the candidate workstation vector are identified based on the production line end position, the workstation position in the candidate workstation vector, and assembly scene data.

[0087] In this embodiment, the obstacle sub-data represents detailed information of each obstacle, including its position, size, and shape.

[0088] In this embodiment, a detour path and related detour parameters are determined based on the assembly scene data and obstacle data, including a detour distance and a maximum detour speed, wherein the detour path represents the path selection to avoid obstacles; the detour distance represents the additional distance required from the current path to the path after avoiding obstacles; and the maximum detour speed is the maximum speed allowed for the robot during the detour to ensure safety.

[0089] In this embodiment, a final assembly path for the product to be assembled is determined based on the end position of the production line, the positions of the candidate workstations, and the detour path of the obstacles. The final assembly path is a feasible route for the robot after considering the obstacles.

[0090] In this embodiment, the assembly paths of all candidate workstations are integrated to determine candidate path data for the robot to select for performing the assembly task.

[0091] The beneficial effects of the above technical solution are: analyzing assembly data, assembly station data inertia data and assembly scene data, determining candidate station vectors and candidate path data for the products to be assembled, and providing high-quality data support for determining the target station and target path, ensuring that the robot can complete the task quickly and accurately, and improving the system's flexibility, adaptability and efficiency.

[0092] Example 4:

[0093] An embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, wherein the determination module includes:

[0094] Historical assembly data unit: obtains historical assembly data of each assembly station in the candidate station vector of the product to be assembled by the assembly robot, wherein the historical assembly data includes the assembly time, station load and assembly results of multiple historical assemblies;

[0095] Path evaluation value unit: calculates the path evaluation value of each assembly station in the candidate station vector of the product to be assembled based on the assembly station data, inertia data, historical assembly data of each assembly station in the candidate station vector of the product to be assembled, and the assembly path of each assembly station in the candidate path data;

[0096] Target station unit: From the path evaluation values of all assembly stations in the candidate station vector of the product to be assembled, the assembly station corresponding to the smallest path evaluation value is selected as the target station for the product to be assembled;

[0097] Target path unit: determines the assembly path corresponding to the target workstation of the product to be assembled as the target path of the product to be assembled.

[0098] In this embodiment, historical assembly data of each workstation in the candidate workstation vector of the product to be assembled by the assembly robot is obtained, including the assembly time, workstation load and assembly results of all historical assemblies of each assembly workstation.

[0099] In this embodiment, the assembly time represents the time required for each assembly at the assembly station; the station load represents the workload of the assembly station in performing the historical assembly task; and the assembly result represents the result of each assembly at the assembly station, including assembly success and assembly failure.

[0100] In this embodiment, based on the path evaluation value of each workstation, the unit selects the assembly workstation with the smallest path evaluation value as the target workstation for the product to be assembled. Selecting the workstation with the lowest evaluation value means higher efficiency in reaching the assembly workstation and in the assembly process.

[0101] In this embodiment, the corresponding assembly path is determined based on the selected target workstation as the target path for the product to be assembled. The target path is the path that the robot takes from the end of the production line of the product to be assembled to the target workstation to ensure the smooth completion of the assembly task.

[0102] The beneficial effects of the above technical solution are as follows: based on assembly station data, inertia data, candidate station vectors of the product to be assembled, and candidate path data, the target station and target path of the product to be assembled are determined, which can realize intelligent and personalized path planning, avoid station overload, improve assembly efficiency, enhance the flexibility and adaptability of the overall production line, and adapt to complex and changing assembly needs.

[0103] Example 5:

[0104] An embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, a path evaluation value unit, including:

[0105] ;

[0106] ;

[0107] ;

[0108] in, represents the path evaluation value of the i-th assembly station in the candidate station vector of the product to be assembled, represents the moving time of the product to be assembled to the i-th assembly station in its corresponding candidate station vector, represents the predicted assembly time of the product to be assembled at the i-th assembly station in its corresponding candidate station vector (which can be determined by combining the historical assembly data of the assembly station); is the unit time; represents the energy consumption of the i-th assembly station in the candidate station vector of the product to be assembled; is the unit energy consumption; represents the real-time station load of the i-th assembly station in the candidate station vector of the product to be assembled (the station load when the product to be assembled has not arrived), represents the maximum station load of the i-th assembly station in the candidate station vector of the product to be assembled; represents the assembly load of the i-th assembly station in the candidate station vector of the product to be assembled, represents the station yield of the i-th assembly station in the candidate station vector of the product to be assembled, represents the moving distance of the product to be assembled from its corresponding end position of the production line to the position of the i-th assembly station in the candidate station vector of the product to be assembled, The moving speed in the dynamic inertial data representing the inertial data of the assembly robot, represents the moving acceleration in the dynamic inertial data of the assembly robot’s inertial data, iN1 represents the number of obstacles at the i-th assembly station in the candidate station vector of the product to be assembled, represents the maximum circumventing speed of the jth obstacle of the i-th assembly station in the candidate station vector of the product to be assembled, represents the detour path length of the jth obstacle of the i-th assembly station in the candidate station vector of the product to be assembled, 、 They represent the moving power and assembly power of the assembly robot respectively, represents the time weight, represents the energy consumption weight, represents a load penalty factor of one; is a natural constant. 、 The unit is the same and can be kilowatt-hour;

[0109] When the assembly tasks performed by each assembly station are fixed (wherein each assembly station can be equipped with multiple assembly robots), the predicted assembly time of the i-th assembly station in its corresponding candidate station vector for the product to be assembled is calculated based on the following formula:

[0110] ;

[0111] in, represents a load penalty factor of two; Represents the number of historical assemblies in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, represents the assembly time of a single product of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, represents the time decay factor, represents the assembly result of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, Indicates that the assembly failed. represents the penalty factor for assembly results, Represents the historical station load of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled; The total number of products to be assembled.

[0112] In this embodiment, the historical assembly data of the assembly station used in the calculation is the same as the assembly process (task) of the current assembly robot for the current product to be assembled;

[0113] In this embodiment, When the assembly result of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled is assembly failure, the value is 1.

[0114] In this embodiment, the time weight Indicates movement time and assembly time The importance of the path evaluation value is greater than 0 and less than 1. 、 、 The units are the same and can both be hours;

[0115] In this embodiment, the energy consumption weight Indicates energy consumption The importance of the path evaluation value is greater than 0 and less than 1 (can be greater than 0 and less than 0.3).

[0116] In this embodiment, Represents the assembly success rate of the i-th assembly station in the candidate station vector of the product to be assembled in the historical assembly data, and the assembly result penalty factor The value is greater than 0 and less than 1.

[0117] In this embodiment, the time decay factor The value is greater than 0 and less than 1.

[0118] In this embodiment, the moving time and assembly time The shorter the path, the greater the evaluation value. The smaller it is, the higher the path efficiency is.

[0119] In this embodiment, the real-time station load or historical station load The higher, or The larger the factor, the higher the path cost. , Load penalty factor 2 Penalty, load penalty factor 、 The values of are all greater than 0 and less than 0.3.

[0120] In this embodiment, the station yield The higher it is, the smaller the exponential term is, and the path cost is reduced, which meets the optimization goal of high yield.

[0121] In this embodiment, represents the energy consumption of the i-th assembly station in the candidate station vector of the product to be assembled, that is, the energy consumption required for assembling the product to be assembled at the i-th assembly station in its corresponding candidate station vector;

[0122] The beneficial effect of the above technical solution is: calculating the path evaluation value of each assembly station in the candidate station vector of the product to be assembled can provide high-quality data support for the target station and realize intelligent and personalized path planning.

[0123] Example 6:

[0124] The embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, an assembly module, including:

[0125] Real-time Positioning Unit: The assembly robot moves along the target path of the product to be assembled, and the real-time position of the assembly robot is obtained based on the assembly robot's lidar and inertial measurement unit;

[0126] Detection unit: Based on the target path of the product to be assembled and the real-time position of the assembly robot, it detects whether path deviation occurs. If so, the assembly robot first moves toward the target path, returns to the target path, and then continues to move along the target path.

[0127] In this embodiment, the assembly robot's current position is acquired in real time, ensuring it consistently moves along the predetermined target path. The robot's real-time position is acquired using a lidar (LiDAR) and an inertial measurement unit (IMU). The lidar provides real-time environmental scanning information, helping to determine the robot's positional relationship with surrounding objects. The IMU provides data such as the robot's acceleration and angular velocity to assist with positioning and path tracking.

[0128] In this embodiment, the lidar constructs an environmental map in real time by emitting laser beams and receiving reflected signals, helping the robot identify surrounding obstacles and accurately locate itself. The inertial measurement unit (IMU) measures the robot's acceleration and angular velocity to infer its position and direction, thus making up for the positioning needs when environmental information is insufficient.

[0129] In this embodiment, the assembly robot detects whether it has deviated from its target path. If the robot deviates from the target path (for example, due to obstacles, environmental factors, or system errors), the detection unit immediately identifies the deviation and instructs the robot to recalibrate toward the target path. After the adjustment, the robot continues along the target path, ensuring the completion of the assembly task.

[0130] The beneficial effects of the above technical solution are as follows: the assembly robot delivers the products to be assembled to the target workstation based on the target path of the products to be assembled, which can avoid the impact of path errors on assembly accuracy and efficiency, achieve flexible response to obstacles, environmental changes or errors, improve the robot's adaptability in complex environments, enhance the stability and accuracy of the system's automated production line, and improve production efficiency and quality.

[0131] Example 7:

[0132] The embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, the assembly module also includes:

[0133] Assembly instruction unit: After the assembly robot arrives at the target workstation, the assembly model generates assembly instructions for the product to be assembled based on the assembly tasks in the assembly data of the product to be assembled and the inertia data of the assembly robot;

[0134] First marking unit: The assembly robot completes product assembly at the target workstation based on the assembly instructions of the product to be assembled, and performs an assembly completion mark on the assembled product to be assembled.

[0135] In this embodiment, when the assembly robot arrives at the target workstation, the assembly model generates assembly instructions for the product to be assembled based on the assembly task and inertial data. The assembly task describes the specific assembly operations (such as installing components, connecting circuits, adjusting, and testing), while the inertial data helps ensure the assembly robot's movements are precise and stable. By combining this information, the assembly instruction unit generates detailed operational instructions to guide the robot to complete the assembly work at the target workstation.

[0136] In this embodiment, the inertial data represents the acceleration, angular velocity, and other data of the assembly robot when performing a task, which helps to accurately control the movement of the robot.

[0137] In this embodiment, when the assembly robot completes the task according to the generated assembly instructions, it marks the assembled product to indicate that the product is ready to be sent to the product destination.

[0138] The beneficial effects of the above technical solution are: completing the assembly of the products to be assembled at the target workstation and marking the assembly completion, which can improve the automation level and production efficiency of the production line, enhance the stability and accuracy of the system's automated production line, and flexibly respond to complex assembly tasks.

[0139] Example 8:

[0140] The embodiment of the present invention provides an intelligent assembly robot system based on inertial SLAM, a delivery module, including:

[0141] Destination path unit: determines the destination path of the product to be assembled with the assembly completion mark based on the product destination, target workstation and assembly scene data in the assembly task of the product to be assembled with the assembly completion mark;

[0142] Second marking unit: The assembly robot delivers the products to be assembled with the assembly completion mark to the product destination along the destination path of the products to be assembled with the assembly completion mark, and performs a delivery mark on the products to be assembled at the destination.

[0143] In this embodiment, the optimal transportation route is determined based on the product destination, target workstation, and assembly scenario data in the assembly task for the product to be assembled, which is marked as completed. This route takes into account the transportation efficiency of the product after assembly, obstacles along the route, transportation distance, and other environmental factors in the scenario.

[0144] In this embodiment, after the assembly robot delivers the product with the assembly completion mark to the destination, the second marking unit adds a delivery mark to the product that has been delivered to the destination. This mark indicates that the product has completed the entire assembly process and arrived at the designated location.

[0145] The beneficial effects of the above technical solution are as follows: the assembly robot delivers the assembled products to the product destination based on assembly data, inertia data and assembly scene data, which can realize the intelligent and automated transportation of the products to be assembled, improve transportation efficiency, increase the automation level of the production line, and improve overall production efficiency and logistics accuracy.

[0146] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0147] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent assembly robot system based on inertial SLAM, characterized in that: include: Monitoring module: Acquires assembly data of products to be assembled, obtains assembly station data of the assembly factory, collects inertial data of assembly robots, and assembly scene data; Analysis module: Analyzes assembly data, assembly station data, inertia data, and assembly scenario data to determine candidate station vectors and candidate path data for the product to be assembled; Determination module: Determines the target workstation and target path of the product to be assembled based on the assembly workstation data, inertia data, candidate workstation vectors of the product to be assembled, and candidate path data; Assembly module: The assembly robot delivers the product to be assembled to the target workstation based on the target path of the product to be assembled, completes the product assembly at the target workstation and marks the assembly as complete; Delivery module: The assembly robot delivers the assembled products to their destination based on assembly data, inertia data, and assembly scene data; Monitoring module, including: Assembly data unit: This unit acquires assembly data of products to be assembled based on the sensor network of the production line. The assembly data includes the end-of-line location of the products to be assembled, the assembly task, and the product destination. Inertial data unit: The assembly robot's lidar and visual sensor collect assembly scene data, and the assembly robot's inertial measurement unit collects inertial data; Assembly station data unit: The assembly station data includes assembly station sub-data of multiple assembly stations, wherein the assembly station sub-data includes station number, station location, executable assembly tasks, real-time station load, maximum station load, and station yield; Analysis modules, including: A candidate workstation vector unit extracts assembly tasks from the assembly data of the product to be assembled, and determines a candidate workstation vector for the product to be assembled based on the extracted assembly tasks and the executable assembly tasks of each assembly station sub-data in the assembly workstation data, wherein the candidate workstation vector includes multiple assembly stations; An obstacle data unit: determining obstacle data for each assembly station in the candidate station vector of the product to be assembled based on the production line end position in the assembly data of the product to be assembled, the station position in the assembly station sub-data of each assembly station in the candidate station vector, and the assembly scene data, wherein the obstacle data includes obstacle sub-data of multiple obstacles, and the obstacle sub-data includes obstacle position, obstacle shape, and obstacle size of the obstacle; Detour unit: Based on the assembly scene data and each obstacle sub-data in the obstacle data, determine the detour path, detour distance and maximum detour speed of each obstacle at each assembly station in the candidate station vector of the product to be assembled; Assembly path unit: Determines the assembly path for each assembly station in the candidate station vector of the product to be assembled based on the end-of-line position in the assembly data of the product to be assembled, the station position in the assembly station sub-data of each assembly station in the candidate station vector, and the detour paths of all obstacles in the obstacle data; Candidate path data unit: determines candidate path data of the product to be assembled based on the assembly paths of all assembly stations in the candidate station vector of the product to be assembled; Identify modules, including: Historical assembly data unit: obtains historical assembly data of each assembly station in the candidate station vector of the product to be assembled by the assembly robot, wherein the historical assembly data includes the assembly time, historical station load and assembly results of multiple historical assemblies; Path evaluation value unit: calculates the path evaluation value of each assembly station in the candidate station vector of the product to be assembled based on the assembly station data, inertia data, historical assembly data of each assembly station in the candidate station vector of the product to be assembled, and the assembly path of each assembly station in the candidate path data; Target station unit: From the path evaluation values of all assembly stations in the candidate station vector of the product to be assembled, the assembly station corresponding to the smallest path evaluation value is selected as the target station for the product to be assembled; Target path unit: determines the assembly path corresponding to the target workstation of the product to be assembled as the target path of the product to be assembled.

2. The intelligent assembly robot system based on inertial SLAM according to claim 1, characterized in that: Path evaluation value unit, including: ； ； ； in, represents the path evaluation value of the i-th assembly station in the candidate station vector of the product to be assembled, represents the moving time of the product to be assembled to the i-th assembly station in its corresponding candidate station vector, represents the predicted assembly time of the product to be assembled at the i-th assembly station in its corresponding candidate station vector; is the unit time; represents the energy consumption of the i-th assembly station in the candidate station vector of the product to be assembled; is the unit energy consumption; represents the real-time station load of the i-th assembly station in the candidate station vector of the product to be assembled, represents the maximum station load of the i-th assembly station in the candidate station vector of the product to be assembled; represents the assembly load of the i-th assembly station in the candidate station vector of the product to be assembled, represents the station yield of the i-th assembly station in the candidate station vector of the product to be assembled, represents the moving distance of the product to be assembled from its corresponding end position of the production line to the position of the i-th assembly station in the candidate station vector of the product to be assembled, The moving speed in the dynamic inertial data representing the inertial data of the assembly robot, represents the moving acceleration in the dynamic inertial data of the assembly robot’s inertial data, iN1 represents the number of obstacles at the i-th assembly station in the candidate station vector of the product to be assembled, represents the maximum circumventing speed of the jth obstacle of the i-th assembly station in the candidate station vector of the product to be assembled, represents the detour path length of the jth obstacle of the i-th assembly station in the candidate station vector of the product to be assembled, 、 They represent the moving power and assembly power of the assembly robot respectively, represents the time weight, Energy consumption weight represents a load penalty factor of one; is a natural constant.

3. The intelligent assembly robot system based on inertial SLAM according to claim 2, characterized in that: When the assembly tasks performed by each assembly station are fixed, the predicted assembly time of the i-th assembly station in its corresponding candidate station vector for the product to be assembled is calculated based on the following formula: ； in, represents a load penalty factor of two; Represents the number of historical assemblies in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, represents the assembly time of a single product of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, represents the time decay factor, represents the assembly result of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled, Indicates that the assembly failed. represents the penalty factor for assembly results, Represents the historical station load of the kth historical assembly in the historical assembly data of the i-th assembly station in the candidate station vector of the product to be assembled; The total number of products to be assembled.

4. The intelligent assembly robot system based on inertial SLAM according to claim 1, characterized in that: Assembly module, including: Real-time Positioning Unit: The assembly robot moves along the target path of the product to be assembled, and the real-time position of the assembly robot is obtained based on the assembly robot's lidar and inertial measurement unit; Detection unit: Based on the target path of the product to be assembled and the real-time position of the assembly robot, it detects whether path deviation occurs. If so, the assembly robot first moves toward the target path, returns to the target path, and then continues to move along the target path.

5. The intelligent assembly robot system based on inertial SLAM according to claim 3, characterized in that: Assembly module, also includes: Assembly instruction unit: After the assembly robot arrives at the target workstation, the assembly model generates assembly instructions for the product to be assembled based on the assembly tasks in the assembly data of the product to be assembled and the inertia data of the assembly robot; First marking unit: The assembly robot completes product assembly at the target workstation based on the assembly instructions of the product to be assembled, and performs an assembly completion mark on the assembled product to be assembled.

6. The intelligent assembly robot system based on inertial SLAM according to claim 1, characterized in that: Delivery module, including: Destination path unit: determines the destination path of the product to be assembled with the assembly completion mark based on the product destination, target workstation and assembly scene data in the assembly task of the product to be assembled with the assembly completion mark; Second marking unit: The assembly robot delivers the products to be assembled with the assembly completion mark to the product destination along the destination path of the products to be assembled with the assembly completion mark, and performs a delivery mark on the products to be assembled at the destination.

Images