Trajectory planning method, system, storage medium and welding robot for welding robot

Abstract

The present invention provides a welding robot trajectory planning method, system, storage medium and welding robot, which are applied to a positioning camera set on the welding robot; the method comprises the following steps: obtaining the positioning information of the welding robot and constructing a global map; constructing a planning path from the welding robot to a welding target based on the positioning information and the global map; and performing fault detection and identification on the welding target. The welding robot trajectory planning method, system, storage medium and welding robot of the present invention can realize accurate and reasonable planning of the welding robot trajectory planning, and are applicable to various complex environments such as indoors and outdoors.

Description

Technical Field

[0001] The present invention relates to the technical field of welding robots, and in particular to a trajectory planning method, system, storage medium and welding robot of a welding robot. Background Art

[0002] In modern manufacturing, welding technology is widely used in many fields. The emergence of welding robots has improved the efficiency and quality of welding, but traditional welding robots still have many problems when facing complex and changeable working environments. For example, in terms of positioning, due to the complex welding environment and the possible deviation of the workpiece placement, traditional welding robots are difficult to quickly and accurately locate welding points, affecting the accuracy and efficiency of welding. In terms of path planning, existing algorithms may not be able to adapt to complex indoor and outdoor environments, resulting in unreasonable movement paths of welding robots, increased energy consumption and working time, and even possible collisions with obstacles. In addition, for fault detection and identification during welding, traditional methods are often not timely and accurate enough, making it difficult to achieve effective monitoring of welding objects such as power equipment.

[0003] With the continuous improvement of the degree of industrial automation, the performance requirements for welding robots are also getting higher and higher. Therefore, developing a welding robot that can more accurately locate, more reasonably plan paths and has effective fault detection capabilities has become an urgent problem to be solved. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the object of the present invention is to provide a trajectory planning method, system, storage medium and welding robot for a welding robot, which can accurately and reasonably plan the trajectory planning of the welding robot and is suitable for various complex environments such as indoors and outdoors.

[0005] In a first aspect, the present invention provides a trajectory planning method for a welding robot, which is applied to a positioning camera provided on the welding robot; the method comprises the following steps: obtaining the positioning information of the welding robot and constructing a global map; based on the positioning information and the global map, constructing a planned path from the welding robot to a welding target; and performing fault detection and identification on the welding target.

[0006] In an implementation of the first aspect, obtaining the positioning information of the welding robot and constructing a global map includes the following steps:

[0007] Implementing laser SLAM based on the VLOAM algorithm, obtaining the positioning information and the global map through mileage measurement and scan matching of the welding robot;

[0008] Alternatively, visual SLAM is implemented based on the VINS-Fusion algorithm, and the positioning information and the global map are obtained through front-end visual information processing and back-end information fusion and optimization of the welding robot.

[0009] In an implementation of the first aspect, the mileage measurement includes the following steps:

[0010] Get the point cloud collected by the LiDAR at two moments;

[0011] Extracting feature points of the point cloud;

[0012] Based on the correspondence between the feature points at the two moments, obtaining a corresponding feature point pair;

[0013] Based on the corresponding feature point pairs, calculating the relative motion transformation matrix at the two moments;

[0014] The positional relationship of the same point cloud at the two moments is obtained based on the relative motion transformation matrix.

[0015] In an implementation of the first aspect, the scan matching includes the following steps:

[0016] Obtain the constructed map point cloud, the robot pose estimate obtained by mileage measurement at the current moment, and the point cloud collected by the lidar at the current moment;

[0017] Convert the point cloud into a map coordinate system to obtain a converted point cloud;

[0018] Based on the objective function of the best match between the converted point cloud and the map point cloud, the position and posture information of the welding robot is obtained.

[0019] In an implementation of the first aspect, the front-end visual information processing includes feature extraction and feature tracking; the feature extraction is used to extract feature points of an image captured by a camera of the welding robot; the feature tracking is used to track feature points of an image captured by an adjacent camera;

[0020] The back-end information fusion and optimization include IMU and GPS information fusion and algorithm optimization; the IMU and GPS information fusion is used to fuse the information of IMU and GPS to obtain the pose estimation information of the tracked feature points; the algorithm optimization is used to obtain real-time pose information based on the objective function and the pose estimation information.

[0021] In an implementation of the first aspect, constructing a planned path from the welding robot to a welding target comprises the following steps:

[0022] The A* algorithm is improved by introducing the weight δ, so that the heuristic function becomes f(n)=δg(n)+(1-ω)h(n); where g(n) is the actual cost from the starting node to the current node n, h(n) is the estimated cost from the current node n to the target node, and ω is the angular velocity;

[0023] When the environment is fixed, the improved A* algorithm is used for global path planning;

[0024] When the environment changes dynamically, the DWA algorithm is used for local path planning.

[0025] In an implementation of the first aspect, in the DWA algorithm, a dynamic window [v min ,v max ]×[ω min ,ω max ] as the speed range at the current moment, where v is the linear speed of the welding robot and ω is the angular velocity of the welding robot; for each speed in the dynamic window, an evaluation function G(v,ω)=αd is calculated. t +βκ+γd o , where α, β and γ are weight coefficients, d t is the distance from the path to the target point, κ is the curvature of the path, d o is the distance between the path and the obstacle.

[0026] In a second aspect, the present invention provides a trajectory planning system for a welding robot, which is applied to a positioning camera provided on the welding robot;

[0027] The system includes an acquisition module, a construction module and a detection and identification module;

[0028] The acquisition module is used to acquire the positioning information of the welding robot and construct a global map;

[0029] The construction module is used to construct a planned path from the welding robot to a welding target based on the positioning information and the global map;

[0030] The detection and identification module is used to perform fault detection and identification on the welding target.

[0031] In a third aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned trajectory planning method for a welding robot.

[0032] In a fourth aspect, the present invention provides a welding robot, wherein the welding robot is provided with a positioning camera; the positioning camera comprises a processor and a memory;

[0033] The memory is used to store computer programs;

[0034] The processor is used to execute the computer program stored in the memory so that the welding robot executes the above-mentioned trajectory planning method of the welding robot.

[0035] As described above, the trajectory planning method, system, storage medium and welding robot of the present invention have the following beneficial effects:

[0036] (1) By integrating information from multiple sensors such as lidar, visual cameras, IMU, and GPS, a multimodal SLAM system is constructed. By fully utilizing the advantages of each sensor, a more comprehensive and accurate environmental perception is achieved, enabling the welding robot to more adaptably cope with diverse indoor and outdoor scenes based on multi-sensor fusion.

[0037] (2) The V-LOAM algorithm is used to implement radar SLAM, and the VINS-Fusion algorithm is used to implement visual SLAM. The above collaborative operation mode effectively integrates laser and visual information, realizes high-frequency and high-precision single motion estimation, enhances the reliability of the system in complex dynamic environments, and provides more reliable position estimation for navigation tasks;

[0038] (3) In the navigation task, the A* algorithm is used for global path planning to achieve autonomous navigation in a fixed environment. In a complex dynamic environment, the DWA algorithm is used to handle local path planning tasks, taking into account obstacles under unknown environmental information, which effectively improves the stability and flexibility of the navigation system in a dynamic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Shown is a flow chart of a trajectory planning method for a welding robot according to an embodiment of the present invention;

[0040] Figure 2 Shown is a schematic diagram of a framework of a trajectory planning system for a welding robot according to an embodiment of the present invention;

[0041] Figure 3 Shown is a schematic structural diagram of a welding robot according to an embodiment of the present invention;

[0042] Figure 4 FIG. 1 is a schematic structural diagram of a positioning camera according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0044] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0045] The trajectory planning method of the welding robot of the present invention is applied to the positioning camera set on the welding robot. The positioning camera uses a three-dimensional camera system to obtain the three-dimensional shape and texture information of the welding workpiece, and determines the welding position through image processing and pattern recognition algorithms. Compared with traditional welding wire positioning and laser positioning, 3D camera positioning is more flexible and efficient: there is no need to teach programming of the weld in the pre-welding preparation stage, and a single photo can be taken to identify multiple welds during the welding process, which greatly simplifies the operation and helps users improve production flexibility and efficiency.

[0046] like Figure 1 As shown, in one embodiment, the trajectory planning method of the welding robot of the present invention includes steps S1 to S3.

[0047] Step S1, obtaining the positioning information of the welding robot and constructing a global map.

[0048] Specifically, the present invention realizes high-precision positioning and environmental perception of the robot by integrating information such as laser radar and visual cameras, provides accurate positioning information for path planning, and constructs a global map to assist the navigation of the welding robot.

[0049] In one embodiment, obtaining the positioning information of the welding robot and constructing a global map includes the following steps:

[0050] 11) Implementing laser SLAM based on the VLOAM algorithm, and acquiring the positioning information and the global map through mileage measurement and scan matching of the welding robot.

[0051] The laser radar located in the positioning camera collects point cloud data of the surrounding environment at a frequency of 10 times per second. These data contain the spatial position information of objects in the environment, such as the geometric shape and distance information of obstacles and working areas. When implementing laser SLAM based on the VLOAM algorithm, the range measurement algorithm and the mapping algorithm are used to work together to generate high-frequency and high-precision single motion estimates through high-frequency, low-fidelity mileage measurement and low-frequency, high-fidelity scan matching.

[0052] Odometer measurement mainly uses the point cloud data obtained by the lidar to calculate the robot's motion information, which includes the following steps:

[0053] a) Obtain the point cloud collected by the LiDAR at two moments.

[0054] Among them, the point clouds collected by the laser radar at two adjacent times t1 and t2 are and Point Cloud Contains N1 points, point cloud Contains N2 points.

[0055] b) extracting feature points of the point cloud.

[0056] Among them, feature points are picked up based on geometric characteristics such as the curvature of the point. Points in Calculate its curvature Select the curvature greater than a certain threshold k th Assume that the extracted feature point set is Similarly, Extract feature point set

[0057] c) Based on the correspondence between the feature points at the two moments, obtaining corresponding feature point pairs.

[0058] The corresponding relationship between the feature points at two moments is found by a matching algorithm. Preferably, the feature points are matched based on descriptors, using FPFH descriptors. Calculate its FPFH descriptor Similarly, for feature points Calculate its FPFH descriptor By comparing the distances between descriptors, we can find matching point pairs. Let the corresponding point pairs after matching be in The distance includes Euclidean distance, Manhattan distance, Hamming distance, etc. By comparing the distances, it is possible to determine which descriptors are similar, thereby inferring the similarity between images. For example, in image matching, if the distance between two descriptors is less than a certain threshold, the two descriptors are considered to match, thereby inferring that the corresponding parts of the image are similar. In machine learning, the distance between descriptors can also be used for clustering analysis, classification tasks, etc. By comparing the descriptor distances between different samples, samples can be grouped or classified.

[0059] d) calculating the relative motion transformation matrix at the two moments based on the corresponding feature point pairs.

[0060] Among them, the relative motion transformation matrix from t1 to t2 is calculated by the least squares method and other methods. Let the corresponding feature point pair The number of is M, then the following equation is solved by the least squares method: The relative motion transformation matrix It contains translation and rotation information, which can represent the movement of the welding robot between these two moments.

[0061] e) obtaining the positional relationship of the same point cloud at the two moments based on the relative motion transformation matrix.

[0062] Among them, let the laser radar coordinate system be L, and the welding robot body coordinate system be B. The coordinate transformation relationship of point p in different coordinate systems is:

[0063] Therefore, in mileage measurement, the relative motion transformation matrix is ​​obtained through a series of feature extraction and matching calculations. So that for the point in the point cloud t2 at time Its corresponding position at time t1 is estimated to be

[0064] Scan matching is based on mileage measurement to further optimize map construction. It is performed at a lower frequency, once per second, to ensure higher accuracy.

[0065] In one embodiment, the scan matching comprises the following steps:

[0066] a) Obtain the constructed map point cloud, the robot pose estimate obtained by mileage measurement at the current moment, and the point cloud collected by the lidar at the current moment.

[0067] Here, let the constructed map point cloud be M, and the welding robot pose estimation obtained by mileage measurement at the current moment be T r , the point cloud collected by the laser radar at the current moment is P c .

[0068] b) converting the point cloud into a map coordinate system to obtain a converted point cloud.

[0069] Among them, the current point cloud P c Convert to map coordinate system, that is, convert point cloud

[0070] c) acquiring the position and posture information of the welding robot based on an objective function of optimal matching between the converted point cloud and the map point cloud.

[0071] Among them, the scan matching algorithm finds The best match between the point cloud M and the map point cloud M. When the iterative closest point (ICP) algorithm is used for matching, let the point in the map point cloud M be m i (i=1,2,…,N m ), convert point cloud The point in is p j (j=1,2,…,N p ). By calculating the distance between points, find each p j The nearest neighbor point m in M i , construct the corresponding relationship {(p j ,m i )}.

[0072] Assume that the optimized welding robot posture obtained by scan matching is T o , then the goal of scan matching is to minimize the following objective function: in is a point in the map point cloud, is the point in the converted point cloud converted to the map coordinate system, and N is the number of points involved in the matching. For the ICP algorithm, N = N p By optimizing this objective function, a more accurate robot pose estimate can be obtained, thereby updating the map.

[0073] 12) Implementing visual SLAM based on the VINS-Fusion algorithm, acquiring the positioning information and the global map through front-end visual information processing and back-end information fusion and optimization of the welding robot.

[0074] Specifically, the camera collects images at a certain frame rate (such as 30 frames per second) to obtain visual feature information of the environment, including texture, edge, color and other information of the object, to provide data support for visual SLAM. Based on the VINS-Fusion algorithm, high-frequency and high-precision single motion estimation is achieved to obtain the accurate position and posture information of the robot in a complex environment, and at the same time, an environmental map is constructed to provide basic data for subsequent path planning, thereby enhancing the reliability and navigation accuracy of the system in a complex dynamic environment.

[0075] The front-end visual information processing includes feature extraction and feature tracking; the feature extraction is used to extract the feature points of the image captured by the camera of the welding robot; the feature tracking is used to track the feature points of the image captured by the adjacent camera. Among them, the ORB algorithm is used to extract the features of the image captured by the camera, detect the key points with significant features and calculate the descriptor; in the feature tracking stage, the feature points of two adjacent frames of images are tracked based on descriptor matching, the similarity is measured by calculating the Hamming distance and other methods, the corresponding relationship of the feature points is established, the motion information of the feature points is obtained, and the accuracy can be improved by setting the similarity threshold or multi-scale tracking and other strategies.

[0076] In one embodiment, the visual information in the feature extraction mainly comes from the image data collected by the camera. The camera collects images at a certain frame rate (for example, 30 frames per second). For each frame of the image, feature extraction is first performed. The present invention takes the ORB (Oriented FAST and Rotated BRIEF) algorithm as an example, which detects key points with significant features in the image and calculates a descriptor for each key point. Let the kth frame image be I k , whose resolution is m×n. For image I k For each pixel (x, y) in the image, the ORB algorithm is used to calculate whether it is a feature point. The specific calculation process involves operations such as comparing the grayscale values ​​of the area around the pixel. After feature extraction, from image I k The feature point set extracted from is F k ={f k,i}, where i = 1, 2, ..., N k , N k is the number of feature points extracted from the k-th frame image. k The size of depends on the content of the image and the parameter setting of the feature extraction algorithm. k-1 and I k , for the previous frame image I k-1 The feature points in the current frame image I k The tracking method used in this embodiment is based on the matching of descriptors. k-1The feature point f k-1,i ∈F k-1 , calculate its descriptor and the current frame image I k The feature point descriptor f k,i ∈F k The method for calculating the similarity can be Hamming distance, etc. Let f k-1,i The descriptor is D k-1,i , for the current frame image I k The feature point f in k,i ∈F k , whose descriptor is D k,j , then the similarity S ij The Hamming distance H(D k-1,i ,D k,j ), the higher the similarity, the smaller the Hamming distance. By comparing the similarities of all feature points, the best matching feature point f is found k,i ∈F k , so as to establish the corresponding relationship of feature points {(f k-1,i ,f k,j )}. In this way, the motion information of the feature points between two adjacent frames can be obtained. At the same time, in order to improve the accuracy and stability of tracking, some other strategies can also be adopted, such as setting a similarity threshold and only retaining the corresponding relations with similarity higher than the threshold; or adopting a multi-scale tracking method to track feature points at different scales.

[0077] The back-end information fusion and optimization include IMU and GPS information fusion and algorithm optimization; the IMU and GPS information fusion is used to fuse the information of IMU and GPS to obtain the pose estimation information of the tracked feature points; the algorithm optimization is used to obtain real-time pose information based on the objective function and the pose estimation information. Among them, the robot pose change is predicted based on the IMU information integral, and then corrected with GPS information. An optimization algorithm is used to minimize the objective function, and the welding robot pose estimation is iteratively adjusted to converge to obtain an accurate pose. Among them, the IMU information can provide acceleration and angular velocity information of the welding robot. IMU measures acceleration and angular velocity at a higher frequency. Let the acceleration measured by the IMU at time be a t =(a t,x ,a t,y ,a t,z ), the angular velocity is ω t =(ω t,x ,ω t,y ,ω t,z ). For linear acceleration a t,x , the change in velocity Δv can be obtained by integration x, and then integrating the velocity gives the change in position Δx. At the same time, considering the angular velocity ω t,x , the change of direction can be obtained by calculating the rotation matrix. For nonlinear situations, a more complex kinematic model is required for calculation. Then, the prediction result is corrected using the position information of GPS. GPS can provide the position information of the welding robot. GPS provides position information at a lower frequency. Suppose the position provided by GPS at time p GPS,t =(p GPS,t,x ,p GPS,t,y ,p GPS,t,z ). Assume that the estimated position and posture of the welding robot after fusion is T r,t In the fusion process, it is necessary to consider the weight distribution of different sensor information. For example, when the GPS signal is good, the GPS information can be given a higher weight; when the GPS signal is poor, it mainly relies on the IMU and the front-end visual information. In order to obtain a more accurate welding robot pose estimation, an optimization algorithm is used to process the fused information. Suppose the observation value of the feature point obtained by feature point tracking in different frame images is z k,i , according to the welding robot pose estimation T r,t The calculated feature principal component analysis obtains the feature point prediction value The goal of the optimization algorithm is to minimize the following objective function: The objective function is solved by an iterative method. In each iteration, T is estimated based on the current welding robot posture. r,t Calculate the predicted value of the feature point Then calculate the value of the objective function E(T). According to the gradient information of the objective function, adjust the welding robot posture estimation T r,t , so that the value of the objective function gradually decreases. After multiple iterations, when the value of the objective function converges to a certain extent, the optimized robot posture T is obtained. o , so as to obtain the current status in real time.

[0078] Among them, it should be noted that

[0079] Step S2: constructing a planned path from the welding robot to a welding target based on the positioning information and the global map.

[0080] Specifically, according to the environment and task requirements of the welding robot, a reasonable motion path from the starting point to the target point is planned to ensure that the welding robot moves safely and efficiently in a complex environment, while improving the flexibility and adaptability of path planning.

[0081] In one embodiment, constructing a planned path from the welding robot to a welding target comprises the following steps:

[0082] 21) The A* algorithm is improved by introducing the weight δ, so that the heuristic function becomes f(n) = δg(n) + (1-ω)h(n). Where g(n) is the actual cost from the starting node to the current node n, h(n) is the estimated cost from the current node n to the target node, and ω is the angular velocity.

[0083] The weight δ is adjusted adaptively according to the robot's position. Assume that the robot's current position is (x, y) and the target position is (x g ,y g ), define the distance function When d is large, in order to improve the search efficiency, the weight of g(n) is increased, that is, δ increases; when d is small, in order to improve the path quality, the weight of h(n) is increased, that is, δ decreases. The specific adjustment method can be achieved through an exponential function.

[0084] 22) When the environment is fixed, the improved A* algorithm is used for global path planning.

[0085] When searching for a path, we start from the starting node and put it into the open list while maintaining the closed list. In each iteration, we select the node with the smallest cost function value from the open list for expansion, calculate the relevant cost and heuristic function value of the adjacent nodes, and adjust them according to the weight strategy. If the adjacent node is not in the open list, we add it. If it is already in the open list and the new calculated value is smaller, we update the information until the target node is added to the closed list, and then we backtrack to get the path.

[0086] In one embodiment, the search starts from the starting node, and the starting node is placed in an open list OpenList, which stores the nodes to be expanded. At the same time, a closed list ClosedList is maintained to store the nodes that have been expanded. In each iteration, the node n with the smallest f(n) value is selected from the OpenList as the current node for expansion. For the adjacent nodes of the current node, g(m) = g(n) + c(n, m) is calculated, where c(n, m) is the cost of moving from node n to node m. At the same time, h(m) is calculated, and then f(m) is calculated according to the adaptive dynamic weight strategy. If node m is not in the OpenList, it is added to the OpenList; if node m is already in the OpenList and the newly calculated f(m) is less than the original value, the relevant information of node m is updated. When the target node is added to the ClosedList, the search ends, and backtracking is started from the target node to obtain the path from the starting node to the target node.

[0087] 23) When the environment changes dynamically, the DWA algorithm is used for local path planning.

[0088] In the DWA algorithm, a dynamic window is defined in the robot speed space to limit the current possible speed range. According to the current position of the robot and environmental information, the robot trajectory at each speed in the dynamic window is calculated, and an evaluation function that comprehensively considers the path safety, smoothness, and degree of proximity to the target point is used for evaluation. The speed with the largest evaluation function value is selected as the current speed. After moving the robot, the dynamic window and evaluation function are recalculated at the new position, and the optimal speed is continuously selected to achieve local path planning.

[0089] In one embodiment, in the DWA algorithm, a dynamic window [v min ,v max ]×[ω min ,ω max ] as the speed range at the current moment, where v is the linear speed of the welding robot and ω is the angular velocity of the welding robot. In the dynamic window, based on the current position of the robot and environmental information, calculate the trajectory of the welding robot at each possible speed. Suppose the current position of the welding robot is (x, y, θ) (θ is the orientation angle of the robot). For the speed, the position after time can be calculated by the kinematic equation. For each speed in the dynamic window, calculate an evaluation function G(v, ω). This evaluation function takes into account multiple factors, such as the safety of the path (whether it will collide with obstacles), the smoothness of the path, and whether the path is close to the target point. Suppose the distance from the path to the target point is d t , the curvature of the path is κ, and the distance to the obstacle is d o , then the evaluation function can be defined as G(v,ω)=αd t +βκ+γd o , where α, β and γ are weight coefficients used to adjust the importance of different factors. The speed with the largest evaluation function value is selected as the current speed of the welding robot, and then the welding robot is moved according to this speed, and the dynamic window and evaluation function are recalculated at the new position, and the optimal speed is continued to be selected to achieve local path planning.

[0090] Step S3: performing fault detection and identification on the welding target.

[0091] Specifically, fault detection and identification are performed on objects such as power equipment during the welding process to achieve effective monitoring of welding operations and ensure welding quality and safe operation of equipment. The process includes the following three stages: (1) real-time capture of power equipment images; (2) use of CNN to identify faults; and (3) convey fault information to the operator through images and fault location.

[0092] In one embodiment, images captured using a camera are provided to a trained CNN network. Comparing with the features extracted from the training images, the CNN network identifies faults and places the captured images into their respective categories. This process is repeated for each image taken at intervals of 5 seconds, and the information is passed to a graphical user interface (GUI), thereby enabling real-time monitoring of power equipment. CNN is a deep learning algorithm used to analyze and classify images. CNN has been shown to effectively reduce the number of parameters without losing model quality. In the present invention, the CNN network uses a 25-layer CNN architecture AlexNet. The fault classification program is implemented in the MATLAB platform. The weighted matrix in the MATLAB program is run on a RaspberryPi4ModelB with 4GBRam to obtain real-time verification and fault identification.

[0093] The protection scope of the trajectory planning method for a welding robot described in an embodiment of the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0094] An embodiment of the present invention also provides a trajectory planning system for a welding robot, and the trajectory planning system of the welding robot can implement the trajectory planning method of the welding robot described in the present invention. However, the implementation device of the trajectory planning system of the welding robot described in the present invention includes but is not limited to the structure of the trajectory planning system of the welding robot listed in this embodiment. All structural deformations and replacements of the prior art made according to the principles of the present invention are included in the protection scope of the present invention.

[0095] The trajectory planning system of the welding robot of the present invention is applied to the positioning camera provided on the welding robot. Figure 2 As shown, in one embodiment, the trajectory planning system of the welding robot of the present invention includes an acquisition module 21 , a construction module 22 and a detection and identification module 23 .

[0096] The acquisition module 21 is used to acquire the positioning information of the welding robot and construct a global map.

[0097] The construction module 22 is connected to the acquisition module 21 and is used to construct a planned path from the welding robot to the welding target based on the positioning information and the global map.

[0098] The detection and identification module 23 is connected to the construction module 22 and is used for performing fault detection and identification on the welding target.

[0099] The structures and principles of the acquisition module 21, the construction module 22 and the detection and recognition module 23 correspond one to one with the steps in the above-mentioned method for adaptively recognizing electroencephalogram signals, and therefore will not be described in detail here.

[0100] In the several embodiments provided by the present invention, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0101] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention. For example, the functional modules / units in the various embodiments of the present invention may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0102] Those skilled in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0103] The embodiment of the present invention also provides a computer-readable storage medium. A person of ordinary skill in the art can understand that all or part of the steps in the method for adaptively identifying EEG signals in the above embodiment can be completed by instructing a processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state disk (SSD)), etc.

[0104] The embodiment of the present invention also provides a welding robot. Figure 3 As shown, in one embodiment, the welding robot is provided with a positioning camera 1, a top-view camera 2, and an overall camera 3. The positioning camera 1 is located at the end of the mechanical arm of the welding robot, and it can follow the mechanical arm to complete omnidirectional movement. The top-view camera 2 is located on the forearm of the welding robot arm, and it completes the top-view photography, scanning and video recording of the welded object. The overall camera 3 is located on the pillar of the welding robot, and it completes the overall photography, scanning and video recording of the welded object.

[0105] The positioning camera includes a processor and a memory.

[0106] The memory is used to store computer programs.

[0107] The memory includes: ROM, RAM, disk, USB flash drive, memory card or CD and other media that can store program codes.

[0108] The processor is connected to the memory and is used to execute the computer program stored in the memory so that the welding robot executes the above-mentioned method for adaptively identifying EEG signals.

[0109] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0110] like Figure 4 As shown, the positioning camera of the present invention is in the form of a general computing device. The components of the positioning camera may include but are not limited to: one or more processors or processing units 41, a memory 42, and a bus 43 connecting different system components (including the memory 42 and the processing unit 41).

[0111] Bus 43 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0112] The positioning camera typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the positioning camera, including volatile and non-volatile media, removable and non-removable media.

[0113] The memory 42 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 421 and / or cache memory 422. The positioning camera may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 424 may be used to read and write non-removable, non-volatile magnetic media ( Figure 4 not shown, usually called a "hard drive"). Although Figure 4Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 43 via one or more data medium interfaces. Memory 42 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of various embodiments of the present invention.

[0114] A program / utility 424 having a set (at least one) of program modules 4241 may be stored, for example, in the memory 42, such program modules 4241 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 4241 generally perform the functions and / or methods of the embodiments described herein.

[0115] The positioning camera may also communicate with one or more external devices (e.g., keyboards, pointing devices, displays, etc.), one or more devices that enable a user to interact with the positioning camera, and / or any device that enables the positioning camera to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via input / output (I / O) interface 44. Furthermore, the positioning camera may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 45. Figure 4 As shown, the network adapter 45 communicates with other modules of the positioning camera via the bus 43. It should be understood that although not shown in the figure, other hardware and / or software modules can be used in conjunction with the positioning camera, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0116] In summary, the trajectory planning method, system, storage medium and welding robot of the welding robot of the present invention construct a multimodal SLAM system by integrating information of various sensors such as laser radar, visual camera, IMU, GPS, etc., and fully utilize the advantages of each sensor to achieve more comprehensive and accurate environmental perception, so that the welding robot can more adaptably cope with diversified indoor and outdoor scenes on the basis of multi-sensor fusion; the V-LOAM algorithm is used to realize radar SLAM, and the VINS-Fusion algorithm is used to realize visual SLAM. The above-mentioned collaborative operation mode effectively integrates laser and visual information, realizes high-frequency and high-precision single motion estimation, enhances the reliability of the system in complex dynamic environments, and provides more reliable position estimation for navigation tasks; in navigation tasks, the A* algorithm is used for global path planning to realize autonomous navigation of fixed and unchanging environments; in complex dynamic environments, the local path planning tasks are processed by the DWA algorithm, and obstacles under unknown environmental information are considered, which effectively improves the stability and flexibility of the navigation system in dynamic environments. Therefore, the present invention effectively overcomes the various shortcomings in the prior art and has a high industrial utilization value.

[0117] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A trajectory planning method for a welding robot, characterized in that: A positioning camera provided on the welding robot; The method comprises the following steps: Obtaining positioning information of the welding robot and constructing a global map; Based on the positioning information and the global map, construct a planned path from the welding robot to a welding target; Fault detection and identification are performed on the welding target.

2. The trajectory planning method of a welding robot according to claim 1, characterized in that: Obtaining the positioning information of the welding robot and constructing a global map includes the following steps: Implementing laser SLAM based on the VLOAM algorithm, obtaining the positioning information and the global map through mileage measurement and scan matching of the welding robot; Alternatively, visual SLAM is implemented based on the VINS-Fusion algorithm, and the positioning information and the global map are obtained through front-end visual information processing and back-end information fusion and optimization of the welding robot.

3. The trajectory planning method of a welding robot according to claim 2, characterized in that: The mileage measurement includes the following steps: Get the point cloud collected by the LiDAR at two moments; Extracting feature points of the point cloud; Based on the correspondence between the feature points at the two moments, obtaining a corresponding feature point pair; Based on the corresponding feature point pairs, calculating the relative motion transformation matrix at the two moments; The positional relationship of the same point cloud at the two moments is obtained based on the relative motion transformation matrix.

4. The trajectory planning method of a welding robot according to claim 2, characterized in that: The scan matching comprises the following steps: Obtain the constructed map point cloud, the robot pose estimate obtained by mileage measurement at the current moment, and the point cloud collected by the lidar at the current moment; Convert the point cloud into a map coordinate system to obtain a converted point cloud; Based on the objective function of the best match between the converted point cloud and the map point cloud, the position and posture information of the welding robot is obtained.

5. The trajectory planning method of a welding robot according to claim 1, characterized in that: The front-end visual information processing includes feature extraction and feature tracking; the feature extraction is used to extract feature points of the image captured by the camera of the welding robot; the feature tracking is used to track the feature points of the image captured by adjacent cameras; The back-end information fusion and optimization include IMU and GPS information fusion and algorithm optimization; the IMU and GPS information fusion is used to fuse the information of IMU and GPS to obtain the position and posture estimation information of the tracked feature points; The algorithm optimization is used to obtain real-time posture information based on the objective function and the posture estimation information.

6. The trajectory planning method of a welding robot according to claim 1, characterized in that: Constructing the planning path of the welding robot to the welding target includes the following steps: The A* algorithm is improved by introducing the weight δ, so that the heuristic function becomes f(n)=δg(n)+(1-ω)h(n); where g(n) is the actual cost from the starting node to the current node n, h(n) is the estimated cost from the current node n to the target node, and ω is the angular velocity; When the environment is fixed, the improved A* algorithm is used for global path planning; When the environment changes dynamically, the DWA algorithm is used for local path planning.

7. The trajectory planning method of a welding robot according to claim 6, characterized in that: In the DWA algorithm, a dynamic window [v min ,v max ]×[ω min ,ω max ] as the speed range at the current moment, where v is the linear speed of the welding robot and ω is the angular velocity of the welding robot; for each speed in the dynamic window, an evaluation function G(v,ω)=αd is calculated. t +βκ+γd o , where α, β and γ are weight coefficients, d t is the distance from the path to the target point, κ is the curvature of the path, d o is the distance between the path and the obstacle.

8. A trajectory planning system for a welding robot, characterized in that: A positioning camera provided on the welding robot; The system includes an acquisition module, a construction module and a detection and identification module; The acquisition module is used to acquire the positioning information of the welding robot and construct a global map; The construction module is used to construct a planned path from the welding robot to a welding target based on the positioning information and the global map; The detection and identification module is used to perform fault detection and identification on the welding target.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the trajectory planning method of the welding robot according to any one of claims 1 to 7 is implemented.

10. A welding robot, characterized in that: The welding robot is provided with a positioning camera; the positioning camera comprises a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory, so that the welding robot executes the trajectory planning method of the welding robot according to any one of claims 1 to 7.

Images