22 results about "Robot workspace" patented technology

The invention relates to the technical field of robot motion control and trajectory planning, in particular to a robot motion trajectory planning method based on AI reinforcement learning. The method comprises the following steps: acquiring robot workspace layout information and an environment sensing probe candidate list, performing probe configuration planning, calibration and planning session index generation, and generating a planning session primary key recording structure; executing multi-source observation acquisition scheduling, data alignment and three-dimensional grid modeling, and generating a reinforcement learning state basic structure; state vector construction and continuous action strategy inference are carried out, and a candidate motion track set structure is generated; and track rolling simulation, grid collision risk assessment, multi-target reward calculation, target track selection and strategy parameter updating are carried out to generate a strategy network parameter set structure. Through combination of multi-source perception and reinforcement learning, intelligent planning of the motion trail of the robot is realized, the safety and efficiency of the trail are improved, and the adaptability and robustness in a dynamic environment are enhanced.

Provided in the present invention are a virtual-robot and physical-robot collaborative safety control method and apparatus based on mixed reality and a digital twin, which method and apparatus specifically relate to the technical field of smart manufacturing and assembly involving virtual-robot and physical-robot collaboration. In the solution, the method comprises: acquiring first posture information of a physical robot, and acquiring second posture information of a virtual robot; on the basis of the first posture information and the second posture information, performing pose registration on the physical robot and the virtual robot, so as to obtain a registration result; on the basis of the registration result and a preset virtual-robot and physical-robot workspace in combination with deep reinforcement learning, performing path planning to determine a virtual-robot and physical-robot collaborative motion path; and on the basis of the virtual-robot and physical-robot collaborative motion path, executing a virtual-robot and physical-robot collaborative safety control action. The solution uses a mixed-reality device to extract virtual-robot and physical-robot posture information, implements virtual-robot and physical-robot pose registration by means of a virtual-real space mapping method, and designs a bidirectional virtual-robot and physical-robot collaborative safety interaction policy based on deep reinforcement learning, thereby effectively improving the safety and effectiveness of a virtual-robot and physical-robot interaction operation.

The utility model provides a gantry machining center capable of automatically discharging, which comprises a base, a gantry truss robot mounted on the base, a power tool apron mounted at the free end of the gantry truss robot, a machine tool table arranged below the gantry truss robot and positioned in a working area of the gantry truss robot, a plurality of mounting seats are mounted in the supporting frame, at least two supporting seats are arranged below each mounting seat, the supporting seats are arranged around a workpiece, a first telescopic mechanism is mounted at the bottom of each supporting seat, a clamping seat is mounted at the free end of each first telescopic mechanism, and lifting frames are arranged on the two sides of the supporting frame correspondingly; the bottoms of the lifting frames are slidably arranged on the supporting guide rails, and the lifting frames are driven by the linear driving module to run on the supporting guide rails. When in use, workpieces can be automatically carried during working, manual carrying is not needed, and time and labor are saved when the workpieces with large quality are machined.

The application belongs to the field of path planning, and particularly relates to a robot formation planning method based on a five-dimensional configuration space. The method comprises the following steps: workspace and configuration space definition: the workspace is a two-dimensional plane for robot movement, and the configuration space is a five-dimensional space containing formation centroid, direction and scaling factor; sampling method: random sampling in the configuration space, and ensuring that the sampling points do not cause collision between robots; distance function: defining the distance between two points in the configuration space, and eliminating the dependence on the number of robots; search: searching in the configuration space using an algorithm, and constructing a collision-free path tree; path generation: converting the path in the configuration space into the path in the robot workspace, and smoothing using a spline curve. The robot formation can pass through a narrow channel, and split and merge when encountering a large obstacle. The application provides a high-efficiency, flexible and collision-free global path planning method for a multi-robot system.

The application relates to a kind of parallel robot work space efficient solving method based on seed space segmentation and principal component analysis, the method uses Monte Carlo and stratification method as the basis of solving algorithm, through the fusion of both, in addition to the seed space segmentation for boundary space screening and the principal component analysis algorithm for the stratification direction of boundary seed space, the design of overall solving algorithm is realized.The method can segment the larger work space of parallel robot, and the rough boundary of work space is quickly and independently screened out by the ratio of reversible point and non-reversible point in the segmented space, the range of overall solving space is reduced, and the optimal stratification direction of boundary seed space is selected by the principal component analysis method, the number of stratification layers in the direction with large gradient change is increased, the feature loss is reduced, and the solving accuracy of overall work space boundary is improved.

The application provides a robot anti-collision method, device and robot control system, and the method comprises the steps of establishing a workpiece coordinate system, constructing a robot workspace under the workpiece coordinate system, obtaining a first pose of a tool end of the robot moving to a target point of the robot workspace, calculating a second pose according to the first pose and a first coordinate system transformation matrix, the second pose being a pose of the tool end in a robot base coordinate system, and the first coordinate system transformation matrix being a matrix for coordinate conversion between the workpiece coordinate system and the robot base coordinate system, calculating an enclosing box of each joint connecting rod and the tool of the robot in the robot base coordinate system according to the second pose, obtaining a plurality of base coordinate enclosing boxes, the enclosing box being a smallest cuboid wrapping the joint part, and the joint part comprising the joint connecting rod and the tool, and determining whether collision occurs between the joint parts corresponding to the plurality of base coordinate enclosing boxes by using the separate axis theorem, so that the problem of high cost of anti-collision through three-dimensional vision technology is solved.

The invention provides an intelligent robot action planning method based on reinforcement learning, and the method comprises the steps: enabling a strategy network to obtain scene features and historical actions according to the current environment observation and the last moment action; inputting the data and tiny time migration into a neural radiation field-physical dynamics coupled world model, and generating output containing physical state increment after model processing; the strategy network executes physical consistency state query on robot working space sampling points, physical state increments generated by a world model and static scene features calculated based on scene features and sampling point positions are added, predicted features are obtained and aggregated, and a physical consistency prediction state is generated; and a current action is generated accordingly. According to the method, physical consistency prediction of a future instant scene state generated by the model is taken as decision input, and physical prediction consistency is taken as a reward item cooperative training strategy, so that the strategy has excellent zero-crossing sample generalization ability in an unknown real environment, and semantic errors of simulation and reality are bridged.

The invention provides a berry lossless sorting method and system based on heterogeneous visual information space-time registration, and belongs to the technical field of agricultural product automatic lossless detection and sorting. According to the method, when berries move along with a conveying belt, image sequences are collected through a linear array hyperspectral camera and a visible light camera, and internal quality features and center coordinates are extracted. The fixed linear edge of the conveyor belt is used as a unique space reference datum, the projection distances from the feature information and the coordinates to the datum are calculated, matching is conducted through the sequence of the projection distances, and the corresponding relation between the quality and the position is established. And then, the escape distance of the berries relative to the mechanical arm movement boundary (namely the downstream boundary of the working area of the parallel robot) in the conveying direction is calculated, a grabbing priority queue is dynamically generated from near to far according to the escape distance, and the parallel robot is controlled to complete sorting according to the quality. According to the method, the problem of dislocation of asynchronous collection of heterogeneous visual information is solved through a software algorithm, and the sorting omission rate is remarkably reduced through dynamic priority scheduling.

The invention discloses a working space optimization method for a double-arm robot, and belongs to the field of robots. The method comprises the steps that D-H coordinate system parameters of the two mechanical arms are obtained; according to the D-H coordinate system parameters of the mechanical arm, the left-right limit carrying distance and the front-back limit carrying distance of the double-arm robot are calculated; and according to the left-right limit carrying distance and the front-back limit carrying distance of the double-arm robot and the actual carrying work space requirement, the base parameters of the double-arm robot are determined. The device comprises a parameter acquisition module, a calculation module and an optimization module. According to the technical scheme, the optimal mechanical arm parameters are obtained through calculation, maximization of the carrying working space of the double-arm robot is achieved, and the working efficiency and flexibility of the mechanical arm of the double-arm robot are improved.

The invention provides a differentiable simulation method and device based on three-dimensional Gaussian sputtering and electronic equipment, a three-dimensional Gaussian model is constructed based on image data and attitude data of a working area of a robot, and based on the three-dimensional Gaussian model, the internal motion state of the robot and the action output by a strategy network, the differentiable simulation of the robot is realized. And performing three-dimensional Gaussian sputtering differential rendering to generate a high-fidelity image, and training a strategy network based on the high-fidelity image and the internal motion state of the robot so as to finely adjust the three-dimensional Gaussian model based on the high-fidelity image and the real visual observation image of the robot in the process that the robot moves based on the output action of the strategy network. By the adoption of the method, strategy optimization can be more efficient, the cost and time of simulation environment construction are reduced, and therefore the applicability of the intelligent robot in the complex dynamic environment is further improved.

The invention relates to a device (10) for processing the front and / or rear bumper (28) of a passenger car, comprising a robot (12) which has gripping elements (22, 24, 26) for the bumper (28), comprising a processing station (14) arranged in the working area of ​​the robot (12) for processing the bumper (28), and a detection device (18) which is configured to detect the position of the bumper (28) at least in the area of ​​the at least one processing station (14). The detection device (18) is connected to a controller (16) of the robot (12), wherein the robot (12) is configured to hold the bumper (28) in the processing station (14) during processing and the controller (16) is configured to control the robot (12) to achieve a target position of the bumper (28) in the at least one processing station (14) depending on the signals from the detection device (18).

The application discloses a kind of hybrid flexible robot workspace analysis method and system, the method comprises the following steps: considering minimizing rope length error, constructing the forward kinematics model of hybrid flexible robot, obtaining the end pose of hybrid flexible robot;The inverse kinematics model of hybrid flexible robot is constructed, and the generalized joint angle control quantity of hybrid flexible robot is obtained;The statics analysis model of hybrid flexible robot is constructed, and the static force balance condition of robot is obtained;The end pose of hybrid flexible robot, the generalized joint angle control quantity and static force balance condition are combined, and the workspace of hybrid flexible robot is constructed, and the working state of hybrid flexible robot is controlled according to workspace.It can quickly and accurately complete the task that needs large space movement and local fine operation simultaneously.The application can be widely applied to the technical field of hybrid flexible robot as a kind of hybrid flexible robot workspace analysis method and system.

The present application relates to a kind of flexible cloth manipulation methods based on key point detection, comprising the following steps: step 1, construct dual-arm robot workspace, the workspace includes top depth camera, the cloth state and surrounding environment in the manipulation area are captured by the real-time capture of visual data by the depth camera;Step 2, define multiple action primitives;Step 3, decode multiple action primitives using VIT-Transformer, generate operation strategy, output the action that needs to be executed;Step 4, key point detection and identification are carried out to cloth by Swin-Transformer, select operation strategy, generate action instruction;Step 5, action instruction is sent to dual-arm robot, and action instruction is executed by dual-arm robot.Compared with prior art, the present application can improve operation efficiency, improve adaptability, improve accuracy, reduce implementation cost and the advantages such as wide application.

The invention relates to the technical field of robot navigation, and discloses a path planning method based on fusion of a zero neural network and a resistance network. According to the method, a robot working area is dispersed into a grid map, the mapping relation between the grid map and a time-varying conical curved surface resistance network is established, obstacle grids correspond to disconnection resistors, and a starting point and an ending point correspond to a current injection point and a current outflow point respectively; constructing a fast zero neural network based on the resistance network mathematical model, and solving time-varying potential of each node; starting from the starting point by the robot, screening candidate nodes at the current grid position, determining the next step of moving the grid by combining the time-varying potential and the direction deviation of the candidate nodes, and repeating until the robot arrives at the terminal point. The real-time performance and the dynamic adaptability of path planning are improved, the local optimal solution is effectively avoided, the path planning is stable and reliable, and the method is suitable for various complex operation scenes.

The invention relates to a robot remote cooperative control method, device and equipment based on the Internet of Things and a medium thereof. The method comprises the following steps: based on environment data and initial attitude data of a robot working area, performing feature extraction on the environment data to obtain environment feature data; based on the environment feature data and the initial attitude data, performing path feasibility analysis through a multi-layer sensing network model to obtain path planning data; according to the path planning data, carrying out collaboration evaluation on key nodes in the path to obtain collaboration evaluation data; performing multi-robot coordination processing through a preset decision fusion mechanism based on the coordination evaluation data to obtain a coordination control scheme; according to the coordination control scheme, a coordination control instruction of each robot is generated, each robot is controlled to execute a remote coordination action, a real-time data correction instruction can be combined, the adaptive capacity and stability of the robot in a complex dynamic environment are enhanced, and the reliability and flexibility of remote coordination control of the robot are improved.

The embodiment of the invention provides a snakelike mechanical arm working space boundary solving and trajectory planning method based on unified kinematics and related equipment, and belongs to the technical field of robots. The method comprises the steps of establishing a unified kinematics model based on a geometric shape of a central axis; based on a unified kinematic model, a system equation used for boundary solving is established by being combined with constraint conditions of the snakelike mechanical arm; on the basis of the system equation, the augmented operability is defined, and the mapping relation of outer boundary points of the working space in the Cartesian space and the joint parameter space is determined by optimizing the augmented operability; and according to the mapping relation and the unified kinematics model, calculation is conducted, and the working space boundary of the snake-shaped mechanical arm is obtained. According to the embodiment of the invention, by establishing a unified coordinate transformation and parameterization representation form, the pose relation among the joints, the connecting rods and the flexible bodies of the mechanical arms of different structures can be described, and unified modeling and solving of the snakelike mechanical arms of multiple structure types are achieved.

Embodiments of the present application disclose a path planning method, related equipment for the path planning method, and a robot, relating to the technical field of path planning, the path planning method includes acquiring first distance information between two adjacent obstacles, determining obstacle cluster type information based on the first distance information, and determining a travel path based on the obstacle cluster type information. The path planning method provided by embodiments of the present application can classify obstacle cluster types in the robot's work area based on the distance between obstacles, for example, it can classify the robot's work area into areas with a high density of obstacles, and when determining a travel path based on this, the obstacle cluster type information can be used as a basis to avoid areas with a high density of obstacles as much as possible, thereby improving the robot's travel efficiency, further improving the robot's work efficiency, and improving the user experience.

The invention discloses a conveying method and system for automatic paint spraying of a robot. Comprising the steps that a workpiece is positioned in the advancing direction, the workpiece moves towards an automatic paint spraying station in a multi-stage speed reduction mode till the workpiece reaches the automatic paint spraying station, and the workpiece is fixed on the automatic paint spraying station; detecting the position of the workpiece, and adjusting the position of the workpiece until the position of the workpiece meets the requirement; the height of the workpiece is adjusted in a self-adaptive mode, after the workpiece is positioned in the advancing direction, the height of the paint spraying working face of the workpiece is obtained, and the height of the workpiece is adjusted till the height meets the requirement; after workpiece information identification and transmission and workpiece height adjustment are completed, the workpiece information of the workpiece is read, and the integrity and validity of the workpiece information are verified until the integrity and validity of the workpiece information meet requirements; and after automatic paint spraying and workpiece information identification and transmission are completed, the paint spraying quality and the operation state are monitored in real time till the paint spraying quality and the operation state meet the requirements. The problems that the workpiece positioning precision is insufficient, the working space of the robot is limited, and automation of the production process is interrupted can be solved.

Workplace system (100) comprising at least one work surface (104, 200, 602), at least one positioning module (106, 400, 500, 606) for arrangement between the at least one work surface (104, 200, 602) and at least one manipulation object (108, 110, 300, 600, 700, 844), a first locking arrangement effective between the at least one work surface (104, 200, 602) and the at least one positioning module (106, 400, 500, 606), and a second locking arrangement effective between the at least one positioning module (106, 400, 500, 606) and the at least one manipulation object (108, 110, 300, 600, 700, 844), characterized in that the first locking arrangement The first resting section (114, 204, 402, 502) has periodically corresponding first resting sections and the second resting arrangement has periodically corresponding second resting sections (302, 404, 406, 504, 816, 816, 854).

The application relates to the technical field of three-dimensional environment reconstruction, in particular to a three-dimensional environment reconstruction system for a marking robot working area, which comprises a point cloud attribute calculation module, which performs ground scanning on the marking robot working area, obtains original point clouds, performs coordinate system alignment on the original point clouds, traverses all data points and calculates the principal curvature value and the principal direction value of each data point. In the application, after the ground scanning on the marking robot working area and the acquisition of the original point clouds, the coordinate system alignment and the principal curvature and the principal direction calculation of each data point are performed, on the basis of the establishment of the curvature attribute point cloud with local geometric attributes, the data points with higher principal curvature values are screened by adopting a set threshold value, and the sensitive response capability to the fine ground marking traces is improved; the feature point cluster established in combination with the spatial proximity and the geometric direction consistency constructs the discrete marking feature expression with stronger spatial continuity and more accurate shape recognition.

The invention relates to an automatic wood sorting facility comprising a cross conveyor (3) with a conveying plane (Z0) on which sawn pieces of wood (2) are advanced, an inspection zone (5) comprising a camera (70), a camera-carrying arrangement (6) by which the camera (70) is arranged in such a manner that the pieces of wood (2) advanced on the cross conveyor (3) can be inspected by means of the camera, a robot gripper zone (8) comprising a robot (RB1) equipped with a robot gripper tool (11) by which a piece of wood (2) can be gripped and handled, a robot-carrying arrangement (9) by which the robot (RB1) is arranged in such a manner that said conveying plane (Z0) extends below the working area of the robot (RB1) with the robot gripper tool (11), a robot control unit (40) to which the camera (70) is connected, and which robot control unit further comprises a vision system (30) by which the piece of wood (2) can be identified, an image processing unit (50) and a robot control (60) configured to form a tracking system with a defined tracking area (TR1) of the length of the conveying plane (Z0), wherein the robot (RB1) is configured, with instruction from the vision system (30), to grip and handle an identified piece of wood (2) within the tracking area (TR1).