Obstacle Avoidance Method and System for Dual-Arm Robot

Abstract

The present invention discloses an obstacle avoidance method and system for a dual-arm robot. The present invention relates to the technical field of obstacle avoidance methods for dual-arm robots. The three-dimensional avoidance logic is determined according to the moving position of the dual-arm robot, the spatial position of the two robotic arms of the dual-arm robot, and multiple obstacle areas, thereby ensuring the accuracy of the three-dimensional avoidance logic. Therefore, the anti-interference logic between the two robotic arms is determined according to the working paths of the two robotic arms and the activity range of the two robotic arms; the multi-dimensional activity range of the dual-arm robot is determined based on the moving direction of the dual-arm robot, the shape of the dual-arm robot, and the activity range of the two robotic arms; and based on the multi-dimensional activity range of the dual-arm robot, the anti-interference logic, and the three-dimensional avoidance logic, the dual-arm robot is triggered to avoid multiple obstacles relative to each obstacle area, thereby ensuring the intelligence of the dual-arm robot's multiple obstacle avoidance relative to each obstacle area and improving the obstacle avoidance accuracy of the dual-arm robot relative to each obstacle area.

Description

Technical Field

[0001] The present invention relates to the technical field of dual-arm robot obstacle avoidance methods, and in particular to a dual-arm robot obstacle avoidance method and system. Background Art

[0002] With the development of science and technology, robots are widely used in people's lives and appear in industrial or commercial scenarios. As a type of dual-arm robot, the dual-arm robot is provided with a mobile base and two robotic arms. The mobile base and the two robotic arms have corresponding ranges of movement. In the existing technology, when the dual-arm robot finds an obstacle in front of it, the dual-arm robot only avoids it based on the range of movement of the mobile base of the dual-arm robot, realizing a single-dimensional avoidance logic, and does not fully consider the range of movement of the two robotic arms, and cannot guarantee the accuracy of the three-dimensional avoidance logic. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art, and the present invention provides an obstacle avoidance method and system for a dual-arm robot.

[0004] An embodiment of the present invention provides an obstacle avoidance method for a dual-arm robot, comprising: determining an actual moving path of the dual-arm robot based on the current position and target position of the dual-arm robot; determining the current positions of multiple obstacles based on obstacle detection of the actual moving path of the dual-arm robot, and determining multiple obstacle areas based on the current positions of the multiple obstacles and the shapes of the multiple obstacles; during the movement of the dual-arm robot, determining a three-dimensional avoidance logic based on the moving position of the dual-arm robot, the spatial positions of the two robotic arms of the dual-arm robot, and multiple obstacle areas; determining, among the two robotic arms of the dual-arm robot, an anti-interference logic between the two robotic arms based on the working paths of the two robotic arms and the activity ranges of the two robotic arms; determining a multi-dimensional activity range of the dual-arm robot based on the moving direction of the dual-arm robot, the shape of the dual-arm robot, and the activity range of the two robotic arms, and triggering multiple obstacle avoidance of the dual-arm robot relative to each obstacle area based on the multi-dimensional activity range of the dual-arm robot, the anti-interference logic, and the three-dimensional avoidance logic.

[0005] An embodiment of the present invention provides an obstacle avoidance system for a dual-arm robot. The obstacle avoidance system for the dual-arm robot is applied to the above-mentioned obstacle avoidance method for the dual-arm robot. The obstacle avoidance system for the dual-arm robot includes:

[0006] A moving path module is used to determine the actual moving path of the dual-arm robot according to the current position and target position of the dual-arm robot;

[0007] An obstacle area module, configured to determine current positions of a plurality of obstacles based on obstacle detection in an actual moving path of the dual-arm robot, and to determine a plurality of obstacle areas according to the current positions of the plurality of obstacles and the shapes of the plurality of obstacles;

[0008] A three-dimensional avoidance module is used to determine the three-dimensional avoidance logic according to the moving position of the dual-arm robot, the spatial position of the two arms of the dual-arm robot, and multiple obstacle areas during the movement of the dual-arm robot;

[0009] An anti-interference logic module is used to determine the anti-interference logic between the two robotic arms of the dual-arm robot based on the working paths and the activity ranges of the two robotic arms;

[0010] The multi-obstacle avoidance module is used to determine the multi-dimensional activity range of the dual-arm robot based on the dual-arm robot's movement direction, the dual-arm robot's shape and the activity range of the two robotic arms, and trigger the dual-arm robot's multi-obstacle avoidance relative to each obstacle area based on the dual-arm robot's multi-dimensional activity range, anti-interference logic and three-dimensional avoidance logic.

[0011] Compared with the prior art, the present invention has the following beneficial effects:

[0012] In an embodiment of the present invention, through the method in the embodiment of the present invention, during the movement of the dual-arm robot, the three-dimensional avoidance logic is determined according to the moving position of the dual-arm robot, the spatial position of the two robotic arms of the dual-arm robot and multiple obstacle areas, which is compatible with the overall consideration of the moving position of the dual-arm robot, the spatial position of the two robotic arms of the dual-arm robot and multiple obstacle areas, and ensures the accuracy of the three-dimensional avoidance logic.

[0013] Therefore, among the two robotic arms of the dual-arm robot, the anti-interference logic between the two robotic arms is determined according to the working paths of the two robotic arms and the activity range of the two robotic arms; the multi-dimensional activity range of the dual-arm robot is determined based on the moving direction of the dual-arm robot, the shape of the dual-arm robot and the activity range of the two robotic arms, and the multiple obstacle avoidance of the dual-arm robot relative to each obstacle area is triggered based on the multi-dimensional activity range of the dual-arm robot, the anti-interference logic and the three-dimensional avoidance logic, which is compatible with the overall consideration of the multi-dimensional activity range, anti-interference logic and three-dimensional avoidance logic of the dual-arm robot, ensures the intelligence of the multiple obstacle avoidance of the dual-arm robot relative to each obstacle area, and improves the obstacle avoidance accuracy of the dual-arm robot relative to each obstacle area. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 1 is a flow chart of an obstacle avoidance method for a dual-arm robot in an embodiment of the present invention;

[0015] Figure 21 is a flow chart of step S11 in the obstacle avoidance method of the dual-arm robot in an embodiment of the present invention;

[0016] Figure 3 1 is a flow chart of step S12 in the obstacle avoidance method for a dual-arm robot in an embodiment of the present invention;

[0017] Figure 4 1 is a flow chart of step S13 in the obstacle avoidance method of the dual-arm robot in an embodiment of the present invention;

[0018] Figure 5 1 is a flow chart of step S14 in the obstacle avoidance method of the dual-arm robot in an embodiment of the present invention;

[0019] Figure 6 1 is a flow chart of step S15 in the obstacle avoidance method of the dual-arm robot in an embodiment of the present invention;

[0020] Figure 7 Schematic diagram of the structure of the obstacle avoidance system of the dual-arm robot in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] See also Figures 1 to 7 , an obstacle avoidance method for a dual-arm robot, comprising:

[0023] Step S11: determining the actual moving path of the dual-arm robot according to the current position and target position of the dual-arm robot;

[0024] Step S12: determining current positions of multiple obstacles based on obstacle detection in the actual moving path of the dual-arm robot, and determining multiple obstacle areas according to the current positions of the multiple obstacles and the shapes of the multiple obstacles;

[0025] Step S13: During the movement of the dual-arm robot, a three-dimensional avoidance logic is determined according to the movement position of the dual-arm robot, the spatial positions of the two manipulator arms of the dual-arm robot, and multiple obstacle areas;

[0026] Step S14: determining, in the two robotic arms of the dual-arm robot, anti-interference logic between the two robotic arms according to the working paths and the movable ranges of the two robotic arms;

[0027] Step S15: determining the multi-dimensional range of motion of the dual-arm robot based on the movement direction of the dual-arm robot, the form of the dual-arm robot, and the range of motion of the two manipulator arms, and triggering the dual-arm robot to perform multiple obstacle avoidance operations relative to each obstacle area based on the multi-dimensional range of motion of the dual-arm robot, anti-interference logic, and three-dimensional avoidance logic;

[0028] refer to Figure 2 ,In step S11, the actual moving path of the dual-arm robot is determined according to the current position and target position of the dual-arm robot;

[0029] In the specific implementation process of the present invention, the specific steps are:

[0030] S111: collecting the current position and target position of the dual-arm robot, determining a movement space of the dual-arm robot according to the current position and target position of the dual-arm robot, and determining multiple alternative movement paths based on the movement space and previous movement paths of the dual-arm robot;

[0031] S112: Determine three optimal movement paths based on preliminary screening of the detection of multiple candidate movement paths, where matching coefficients between the three optimal movement paths and the dual-arm robot meet a preset matching coefficient threshold;

[0032] S113: Based on the detection of the three optimal moving paths, the moving distance of each optimal moving path is determined, and the actual moving path of the dual-arm robot is determined according to the moving distance of each optimal moving path, the shape of the dual-arm robot and the power of the dual-arm robot.

[0033] In an embodiment of the present application, the current position and target position of the dual-arm robot are collected, and the precise position information of the dual-arm robot is obtained through sensors or positioning systems (such as GPS, lidar, inertial navigation unit, etc.); the current position is the current coordinates or position description of the dual-arm robot, and the target position is the coordinates or position description of the end point that the dual-arm robot needs to reach; at this time, a practical example: assuming that the dual-arm robot is performing a handling task in a warehouse, its current position is the entrance of the warehouse (coordinates: X1, Y1), and the target position is a shelf in the warehouse (coordinates: X2, Y2); through GPS or lidar positioning, the dual-arm robot can accurately know its current position and target position.

[0034] Based on the current position and target position, combined with the environmental map or real-time sensor data, the spatial range occupied by the dual-arm robot during the movement is determined. This spatial range should take into account the size, shape and movement trajectory of the dual-arm robot; at this time, in the warehouse environment, the movement space of the dual-arm robot is a rectangular area on a two-dimensional plane, which starts from the current position and ends at the target position, and takes into account the width and length of the dual-arm robot, as well as the turning radius; for example, if the width of the dual-arm robot is W, the length is L, and the turning radius is R, then the movement space is a rectangle with a width of W+2*R (taking into account turning) and a length of the shortest distance from the current position to the target position.

[0035] Using a path planning algorithm, multiple movement paths are generated within a determined movement space. These paths should take into account the dual-arm robot's shape, size, movement speed, steering ability and other constraints. At the same time, the dual-arm robot's previous movement path data is combined with machine learning or heuristic search algorithms to optimize path selection. At this point, in the warehouse, the dual-arm robot has multiple paths from its current position to its target position.

[0036] For example, it chooses to move directly in a straight line (if there is no obstacle) or to go around an obstacle. Based on its previous movement experience, the dual-arm robot knows which paths are feasible and which paths encounter obstacles. Suppose the dual-arm robot has successfully walked two paths in the past: one is to move in a straight line (path A) and the other is to go around an obstacle (path B). In this task, the dual-arm robot also considers a new path (path C), which combines parts of path A and path B to find a better solution. Therefore, the dual-arm robot will generate these three alternative paths and prepare for further evaluation and selection.

[0037] Furthermore, three better movement paths are determined based on preliminary screening of the detection of multiple candidate movement paths, and the matching coefficients between the three better movement paths and the dual-arm robot meet a preset matching coefficient threshold.

[0038] At this time, the dual-arm robot will detect multiple alternative movement paths generated previously, which usually includes detecting potential obstacles on the path, calculating the path length, evaluating the path curvature, etc. The purpose of the detection is to evaluate the feasibility, safety and efficiency of each path. At this time, the dual-arm robot will use sensors (such as lidar, cameras) to detect obstacles on the path in real time, and combine the environmental map and path planning algorithm to calculate the length and curvature of the path. In addition, the dual-arm robot also considers other factors, such as lighting conditions on the path, ground material, etc. These factors will affect the movement speed and stability of the dual-arm robot.

[0039] After inspecting all alternative paths, the dual-arm robot will conduct a preliminary screening of the paths based on preset screening criteria; the screening criteria include the path's safety (such as avoiding collisions with obstacles), efficiency (such as path length and time required), and the dual-arm robot's adaptability (such as considering the dual-arm robot's shape, size, and mobility); at this time, the dual-arm robot will score each path, with higher scores indicating better paths; the scoring is based on the weighted sum of multiple indicators, such as the inverse of the path length (the shorter the higher the score), the inverse of the number of obstacles on the path (the fewer the higher the score), and the inverse of the path curvature (the smoother the path, the higher the score); then, the dual-arm robot will select the paths with the highest scores as candidates.

[0040] After preliminary screening, the dual-arm robot will select three better movement paths from the candidate paths. These paths must not only perform well in terms of safety, efficiency and adaptability, but also be highly matched with the current state and task requirements of the dual-arm robot. To quantify this degree of matching, the dual-arm robot will calculate the matching coefficient for each path. At this time, the matching coefficient is a comprehensive score that takes into account the degree of matching between the path and multiple factors such as the dual-arm robot's shape, size, movement speed, power, task priority, etc. For example, if a path requires the dual-arm robot to turn frequently, and the dual-arm robot's steering ability is limited, then the matching coefficient of this path will be lower. On the contrary, if a path is straight and short, and the dual-arm robot has sufficient power, then the matching coefficient of this path will be higher. To ensure that the selected path is feasible and efficient, the dual-arm robot will set a preset matching coefficient threshold. Only paths with a matching coefficient higher than this threshold will be selected as the better movement path.

[0041] Specifically, the dual-arm robot has generated multiple alternative movement paths; in step S112, the dual-arm robot begins to detect and screen these paths; the dual-arm robot uses lidar and cameras to detect obstacles on the path in real time, and calculates the length and curvature of each path in combination with the environment map; for example, the dual-arm robot finds that there is a large obstacle on path A that needs to be bypassed, while path B is longer but relatively smooth and has no obstacles; based on the detection results, the dual-arm robot gives each path a score; path A has a lower score because it needs to bypass obstacles; path B is long but has a lower score. Safe and smooth, with a medium score; path C (a newly discovered short and straight path) scored the highest because it is short and has no obstacles; after preliminary screening, the dual-arm robot selected the three paths with the highest scores (assuming they are paths B, C and another alternative path D) as candidates; then, the dual-arm robot calculated the matching coefficient of each path, taking into account factors such as the dual-arm robot's shape, size, moving speed, power and task priority; finally, the dual-arm robot determined paths C, B and D as the better moving paths because their matching coefficients were all higher than the preset threshold.

[0042] Therefore, based on the detection of three optimal moving paths, the moving distance of each optimal moving path is determined, and the actual moving path of the dual-arm robot is determined according to the moving distance of each optimal moving path, the shape of the dual-arm robot and the power of the dual-arm robot. This is compatible with the overall consideration of the moving distance of each optimal moving path, the shape of the dual-arm robot and the power of the dual-arm robot, ensuring the accuracy of the actual moving path of the dual-arm robot.

[0043] At this time, the dual-arm robot will calculate the actual moving distance of each path based on the three best moving paths determined previously. This is usually an accurate measurement of the path length, taking into account the straight and curved parts of the path, as well as the detour around obstacles; at this time, the dual-arm robot uses the distance calculation function in the path planning algorithm, combined with the environmental map and sensor data, to accurately measure the length of each path; for curved parts and detour around obstacles.

[0044] When determining the actual movement path, the dual-arm robot needs to comprehensively consider its own shape and power. Shape factors include the size, weight, and range of motion of the dual-arm robot, which will affect the movement efficiency and stability of the dual-arm robot on different paths. Power factors involve the remaining power and expected power consumption of the dual-arm robot, which will affect the endurance and reliability of the dual-arm robot during the task execution. At this time, the dual-arm robot will assign an additional weight or score to each path based on the shape and power factors. For example, if a path requires the dual-arm robot to move or turn frequently, and the shape or power of the dual-arm robot is not suitable for such high-intensity movement, then the weight of this path will be reduced. On the contrary, if a path is relatively flat and does not require much movement, then the weight of this path will be increased.

[0045] After comprehensively considering factors such as moving distance, shape and power, the dual-arm robot will select an optimal path as the actual moving path. This path should be the best balance in terms of safety, efficiency and adaptability, and be able to meet the current state and task requirements of the dual-arm robot; at this time, the dual-arm robot will use decision-making methods such as weighted sum or sorting to determine the optimal path; the weighted sum method involves adding the weights of factors such as moving distance, shape and power to obtain a comprehensive score; then, the dual-arm robot will select the path with the highest score as the actual moving path; the sorting method involves sorting the paths according to one or more key factors (such as moving distance or shape adaptability), and then selecting the path that ranks first.

[0046] refer to Figure 3 In step S12, the current positions of the plurality of obstacles are determined based on obstacle detection in the actual moving path of the dual-arm robot, and the plurality of obstacle areas are determined according to the current positions of the plurality of obstacles and the shapes of the plurality of obstacles;

[0047] In the specific implementation process of the present invention, the specific steps are:

[0048] S121: collecting an actual moving path of the dual-arm robot, marking a plurality of obstacles in the actual moving path based on the detection of the actual moving path of the dual-arm robot, and determining current positions of the plurality of obstacles; the current positions of the plurality of obstacles include horizontal positions and height positions of the plurality of obstacles;

[0049] S122: The dual-arm robot is equipped with a radar, and collects corresponding point cloud sets based on the radar's detection of each obstacle, and determines the shape and type of the obstacle based on the recognition of each point cloud set;

[0050] S123: Determine the activity impact range corresponding to the obstacle based on the shape and type of the obstacle, and determine multiple obstacle areas based on the horizontal positions and height positions of the multiple obstacles and the activity impact ranges corresponding to the obstacles;

[0051] In an embodiment of the present application, the dual-arm robot will use its built-in navigation system and sensors (such as GPS, inertial navigation unit, lidar, etc.) to collect its movement path in the actual environment in real time. These sensors will provide key information such as the position, speed and direction of the dual-arm robot, thereby allowing the dual-arm robot to build and update a map of its current movement path; at this time, the dual-arm robot will use SLAM (Simultaneous Localization and Mapping) technology to build an environmental map in real time and update the map according to its own position; in addition, the dual-arm robot also uses auxiliary means such as visual sensors or RFID tags to improve positioning accuracy.

[0052] While collecting the moving path, the dual-arm robot uses its sensors (such as lidar, cameras, etc.) to detect obstacles on the path. These sensors emit signals (such as laser beams or light) to the surrounding environment and receive reflected signals to identify obstacles. At this time, the lidar calculates the distance to the obstacle by emitting a laser beam and measuring the reflection time. The camera detects obstacles by capturing environmental images and using image processing algorithms. In addition, the dual-arm robot also uses sonar sensors to detect obstacles at close range.

[0053] Once an obstacle is detected, the dual-arm robot will mark it on the environment map and determine the current position of the obstacle, including the horizontal orientation (such as X, Y coordinates) and height position (such as Z coordinate) of the obstacle, so that the dual-arm robot can accurately avoid it in subsequent planning; at this time, the dual-arm robot will calculate the position of the obstacle based on the data provided by the sensor; for lidar, the dual-arm robot uses triangulation or time difference measurement to calculate the distance and height of the obstacle; for cameras, the dual-arm robot uses image processing algorithms (such as edge detection, feature matching, etc.) to identify the outline and position of the obstacle.

[0054] Specifically, suppose a dual-arm robot is performing a handling task in a warehouse, and its actual moving path is from shelf A to shelf B; during the movement, the dual-arm robot needs to avoid obstacles on the path; the dual-arm robot uses its built-in navigation system and lidar to collect its moving path in real time; during the movement, the dual-arm robot will continuously update its position and environmental map; when the dual-arm robot approaches the aisle between shelf A and shelf B, its lidar detects that there is a large cargo box (obstacle 1) in the aisle blocking the way; at the same time, the camera also captures the image of the cargo box; the dual-arm robot marks the cargo box on the environmental map and calculates its current position; assuming that the horizontal position of the cargo box is (X1, Y1) and the height position is Z1, this information will be used for subsequent path planning and obstacle avoidance operations.

[0055] Furthermore, the dual-arm robot is equipped with a radar, and collects corresponding point cloud sets based on the detection of each obstacle by the radar, and determines the shape and type of the obstacle based on the identification of each point cloud set.

[0056] At this time, the radar (usually a lidar) equipped with the dual-arm robot will actively emit laser beams to the surrounding environment and receive the signals reflected by these laser beams from the surface of obstacles; the radar calculates the distance and direction of the obstacle by measuring the time difference between the emission and reception of the laser beam, as well as the deflection angle of the laser beam; at this time, the lidar continuously emits laser beams to the surrounding environment by rotating or scanning, forming a continuous scanning surface; when the laser beam encounters an obstacle, it will be reflected back and received by the radar; the processor inside the radar will calculate the three-dimensional coordinates of the obstacle based on the received signal.

[0057] When detecting obstacles, the radar collects a large amount of three-dimensional point data, which is called a point cloud set. Each point in the point cloud set represents a location on the surface of the obstacle, and the three-dimensional shape of the obstacle is constructed through the collection of these points. At this time, the lidar will continuously emit and receive laser beams, and each time it receives a reflected signal, it will calculate the coordinates of a three-dimensional point. As the radar scans continuously, a point cloud set containing a large number of points will be generated. This point cloud data is stored in the memory of the dual-arm robot for subsequent processing and analysis.

[0058] The dual-arm robot will process and analyze the collected point cloud set to identify the shape and type of obstacles, which usually involves filtering, segmentation, feature extraction and classification of point cloud data. The dual-arm robot uses point cloud processing algorithms (such as those in the PCL library) to process point cloud data. First, filtering is used to remove noise points and outliers. Then, a segmentation algorithm is used to segment the point cloud data into different parts, each corresponding to an obstacle. Next, the features of each part (such as shape, size, surface texture, etc.) are extracted. Finally, a classification algorithm (such as support vector machines, random forests, etc.) is used to classify the obstacles according to the features to determine their type and shape.

[0059] Therefore, the activity impact range corresponding to the obstacle is determined based on the shape and type of the obstacle, and multiple obstacle areas are determined according to the horizontal position, height position and activity impact range corresponding to the multiple obstacles. This is compatible with the overall consideration of the horizontal position, height position and activity impact range corresponding to the multiple obstacles, ensuring the accuracy of multiple obstacle areas.

[0060] At this time, the dual-arm robot will evaluate the activity impact range of each obstacle based on the shape and type of the obstacle previously detected and identified by radar; the activity impact range refers to the area around the obstacle that interferes with its movement or operation. The size and shape of this range depends on the specific characteristics of the obstacle, such as size, shape, stability, and whether it moves. At this time, there is a predefined obstacle impact range database inside the dual-arm robot. According to the type and shape of the obstacle, the dual-arm robot searches the database for the corresponding activity impact range template; if there is no directly matching template in the database, the dual-arm robot will use a rule-based or machine learning algorithm to dynamically calculate the activity impact range.

[0061] The dual-arm robot uses its navigation system or sensors to accurately determine the horizontal position (X, Y coordinates) and height position (Z coordinate) of each obstacle. This information is the basis for the subsequent division of obstacle areas. At this time, the dual-arm robot uses GPS, inertial navigation units, lidar or other types of sensors to obtain the location information of the obstacles. For determining the height position, sensors that can measure distance, such as lidar or ultrasonic sensors, are particularly useful.

[0062] The dual-arm robot will divide the environmental map into corresponding obstacle areas based on the activity impact range and location information of each obstacle. These areas will serve as a reference for the dual-arm robot in subsequent path planning and obstacle avoidance operations. At this time, the dual-arm robot will mark the activity impact range of each obstacle with polygons or other shapes in its internal environmental map. These marks are dynamically updated to reflect changes in the position or shape of the obstacle. When planning the path, the dual-arm robot will avoid these obstacle areas to ensure the safe and efficient completion of the task.

[0063] At this point, suppose the dual-arm robot is performing a handling task in a complex warehouse environment with multiple different types of obstacles, such as shelves, large cargo boxes, and mechanical equipment. The dual-arm robot first detects and identifies obstacles in the warehouse through radar. For shelves, the dual-arm robot knows that they are usually fixed, so the range of its activities is limited to the space occupied by the shelves themselves. For large cargo boxes, the dual-arm robot will evaluate their stability and the range of shaking caused by movement, thereby determining a slightly larger range of activities. For mechanical equipment, the dual-arm robot will consider its operating radius and the range of motion generated, thereby determining a wider range of activities.

[0064] The dual-arm robot uses its navigation system to accurately determine the horizontal position and height of each obstacle; for example, the coordinates of a shelf are (X1, Y1, Z1), the coordinates of a large cargo box are (X2, Y2, Z2), and the coordinates of a mechanical device are (X3, Y3, Z3); based on the obstacle's range of influence and location information, the dual-arm robot divides the corresponding obstacle areas in its internal environment map. These areas are marked with polygons or other shapes and are dynamically updated to reflect changes in the obstacle's position or shape; when planning subsequent paths, the dual-arm robot will avoid these obstacle areas to ensure safe execution of the task; through the above steps, the dual-arm robot can determine the range of influence of the obstacle based on its shape and type, and divide the obstacle areas based on the location information, which provides important support for the dual-arm robot's path planning and obstacle avoidance operations in complex environments.

[0065] In one embodiment of the present application, an activity influence range matching table is collected, and the activity influence range matching table is shown in Table 1:

[0066] Table 1 Activity influence range matching table

[0067] Obstacle type Scope of activity shelves A rectangular area the same size as the shelf Large cargo box A rectangular area that extends a certain distance outward based on the size of the cargo box Mechanical equipment A circular or elliptical area centered on the device, taking into account the operating radius

[0068] Suppose a dual-arm robot detects a large cargo box measuring 2m × 3m × 1.5m (length × width × height). Based on the activity impact range matching table, the dual-arm robot will find a rectangular area template that extends outward a certain distance based on the cargo box's dimensions. For example, if the area extends outward by 0.5m, the activity impact range will be a rectangular area of 2.5m × 3.5m. Based on the actual position of the cargo box (such as the (X, Y, Z) coordinates), the dual-arm robot will divide the corresponding obstacle area in the environment map.

[0069] refer to Figure 4 ,In step S13, during the movement of the dual-arm robot, a three-dimensional avoidance logic is determined according to the moving position of the dual-arm robot, the spatial positions of the two manipulator arms of the dual-arm robot, and multiple obstacle areas;

[0070] In the specific implementation process of the present invention, the specific steps are:

[0071] S131: monitoring the movement of the dual-arm robot in real time, collecting the movement position of the dual-arm robot, and determining multiple obstacle areas around the dual-arm robot based on the perimeter detection of the movement position of the dual-arm robot. At this time, the movable range of the dual-arm robot during movement is determined based on the detection of the dual-arm robot and the multiple obstacle areas around the dual-arm robot.

[0072] S132: In the dual-arm robot, construct a corresponding dual-arm robot coordinate system based on the dual-arm robot, and determine the spatial positions of the two robotic arms of the dual-arm robot according to the two robotic arms of the dual-arm robot and the dual-arm robot coordinate system;

[0073] S133: Determine the activity dimensions of the two robotic arms based on the traversal of the two robotic arms of the dual-arm robot, determine the activity range of the two robotic arms of the dual-arm robot based on the activity dimensions of the two robotic arms and the relative distance between the two robotic arms, and determine the three-dimensional avoidance logic based on the movable range of the dual-arm robot during movement and the activity range of the two robotic arms of the dual-arm robot.

[0074] In an embodiment of the present application, the movement of the dual-arm robot is monitored in real time, the moving position of the dual-arm robot is collected, and multiple obstacle areas around the dual-arm robot are determined based on the peripheral detection of the moving position of the dual-arm robot. At this time, the movable range of the dual-arm robot during the movement is determined based on the detection of the dual-arm robot and the multiple obstacle areas around the dual-arm robot, which is compatible with the overall consideration of the detection of the dual-arm robot and the multiple obstacle areas around the dual-arm robot, thereby ensuring the accuracy of the movable range of the dual-arm robot during the movement.

[0075] At this time, the dual-arm robot is equipped with a variety of sensors (such as encoders, gyroscopes, odometers, etc.), which can monitor the movement status of the dual-arm robot in real time, including position, speed, acceleration and other information; in addition, the dual-arm robot also uses external positioning systems (such as GPS, UWB positioning systems, etc.) to obtain more accurate global position information; at this time, sensor data is continuously collected and transmitted to the control system of the dual-arm robot; the control system processes this data through algorithms to update the position information of the dual-arm robot in real time; for the external positioning system, the dual-arm robot needs to receive the positioning signal through the communication module and convert it into the position coordinates of the dual-arm robot itself.

[0076] Based on real-time monitoring, the dual-arm robot control system will regularly or on-demand collect the current position information of the dual-arm robot. This information is usually expressed in the form of coordinates, including the X, Y, and Z coordinates (and rotation angles) of the dual-arm robot in a two-dimensional plane or three-dimensional space. At this time, the control system will calculate the current position coordinates of the dual-arm robot based on sensor data or signals from an external positioning system, and store them in memory for subsequent use.

[0077] When moving, the dual-arm robot will use radar, cameras, lidar and other sensors to scan and detect the surrounding environment. These sensors can identify obstacles around the dual-arm robot and mark the corresponding obstacle areas in the dual-arm robot's environmental map according to the shape, type and location information of the obstacles. At this time, the environmental data collected by the sensors are transmitted to the dual-arm robot's perception system; the perception system processes this data through algorithms, identifies obstacles, and draws obstacle areas in the environmental map. These areas are usually represented by polygons, circles or other shapes to reflect the actual space occupied by the obstacles.

[0078] After determining the current position of the dual-arm robot and the surrounding obstacle area, the dual-arm robot control system will calculate the dual-arm robot's movable range during movement based on this information; the movable range refers to the area where the dual-arm robot can move safely without colliding with any obstacles; at this time, the control system uses a path planning algorithm, combined with the dual-arm robot's current position, obstacle area and preset safety distance, to calculate one or more feasible paths from the current position to the target position. The area covered by these paths is the dual-arm robot's movable range during movement; the control system will also continuously update this range to reflect changes in the dual-arm robot's position and dynamic changes in the surrounding environment.

[0079] At this point, assume that a dual-arm robot is performing a handling task in an automated warehouse; the warehouse has multiple shelves, mechanical equipment, and transportation channels; the dual-arm robot uses built-in encoders, gyroscopes, and other sensors to monitor its movement status in real time, including information such as position, speed, and acceleration; at the same time, the dual-arm robot also uses the UWB positioning system to obtain its global position coordinates in the warehouse; when the dual-arm robot moves near a shelf, the control system collects its current position coordinates as (X, Y, Z), where X and Y represent the dual-arm robot's position in the two-dimensional plane, and Z represents the dual-arm robot's height.

[0080] The dual-arm robot uses lidar to scan the surrounding environment and identify nearby obstacles such as shelves and mechanical equipment. These obstacles are marked as polygonal areas in the dual-arm robot's environmental map to reflect their actual occupied space; based on the dual-arm robot's current position and obstacle area information, the control system calculates an optimal path from the current position to the shelf. This path avoids all obstacle areas and ensures that the dual-arm robot will not collide with any obstacles during movement; the control system also continuously updates this path and movable range to reflect changes in the dual-arm robot's position and dynamic changes in the surrounding environment; through the above steps, the dual-arm robot can monitor its movement status in real time and determine its movable range during movement based on the surrounding environment information, which provides important support for the dual-arm robot's autonomous navigation and obstacle avoidance in complex environments.

[0081] Furthermore, in the dual-arm robot, a corresponding dual-arm robot coordinate system is constructed based on the dual-arm robot, and the spatial positions of the two robotic arms of the dual-arm robot are determined according to the two robotic arms of the dual-arm robot and the dual-arm robot coordinate system, which is compatible with the overall consideration of the two robotic arms of the dual-arm robot and the dual-arm robot coordinate system, and ensures the accuracy of the spatial positions of the two robotic arms of the dual-arm robot.

[0082] At this time, in the dual-arm robot, it is first necessary to construct a coordinate system based on the dual-arm robot itself, that is, the dual-arm robot coordinate system. This coordinate system usually selects a fixed point of the dual-arm robot (such as the center of gravity, base center, etc.) as the origin, and defines three mutually perpendicular coordinate axes (X-axis, Y-axis, Z-axis) to describe the position and posture of the dual-arm robot in three-dimensional space; at this time, the construction of the dual-arm robot coordinate system is usually completed in the design stage of the dual-arm robot and programmed in the control system of the dual-arm robot; the selection and definition of the coordinate system need to take into account the structural characteristics, kinematic model and subsequent task requirements of the dual-arm robot.

[0083] After constructing the dual-arm robot coordinate system, it is necessary to determine the spatial positions of the two robotic arms of the dual-arm robot in this coordinate system. This is usually done by installing sensors (such as encoders, gyroscopes, etc.) on the robotic arms to measure the angles or displacements of each joint of the robotic arms in real time. The kinematic model of the dual-arm robot is then used to convert these measurements into three-dimensional coordinates of the end effector of the robotic arm in the dual-arm robot coordinate system. At this time, the spatial position determination of the robotic arm involves the forward kinematics calculation of the dual-arm robot. Forward kinematics refers to calculating the position of the end effector of the robotic arm in the dual-arm robot coordinate system based on the angles or displacements of each joint of the robotic arm. This is usually achieved through matrix transformation, which involves the calculation of rotation matrices and translation vectors.

[0084] Therefore, the activity dimensions of the two robotic arms of the dual-arm robot are determined based on the traversal of the two robotic arms, and the activity range of the two robotic arms of the dual-arm robot is determined according to the activity dimensions of the two robotic arms and the relative distance between the two robotic arms. The three-dimensional avoidance logic is determined according to the movable range of the dual-arm robot during the movement and the activity range of the two robotic arms of the dual-arm robot. It is compatible with the overall consideration of the movable range of the dual-arm robot during the movement and the activity range of the two robotic arms of the dual-arm robot, ensuring the accuracy of the three-dimensional avoidance logic. At the same time, it is compatible with the overall consideration of the moving position of the dual-arm robot, the spatial position of the two robotic arms of the dual-arm robot and multiple obstacle areas, ensuring the accuracy of the three-dimensional avoidance logic.

[0085] At this time, the two robotic arms of the dual-arm robot each have a certain range of motion and movement capabilities, which are achieved through the movement of the various joints of the robotic arms (such as rotational joints, translational joints, etc.); the activity dimension refers to the directions in which the robotic arm can move freely, which is usually related to the structural design and joint configuration of the robotic arm; for example, a robotic arm with three rotational joints performs rotational motion on three different planes; at this time, determining the activity dimension of the robotic arm usually involves analyzing the kinematic model of the robotic arm; by analyzing the joint configuration and connection method of the robotic arm, it is determined in which directions the robotic arm can move, as well as the range and limitations of these movements.

[0086] After determining the dimensions of the robot arm's motion, it is necessary to further determine the range of motion of the robot arm in these dimensions; the range of motion refers to the set of all positions that the robot arm's end effector can reach, which is usually affected by the robot arm's structure, joint limitations, and external environment (such as obstacles); at this time, determining the range of motion of the robot arm usually involves simulating and analyzing the robot arm's kinematic model; by simulating the motion state of the robot arm at different joint angles, all positions that the robot arm's end effector can reach are calculated, and a range of motion diagram of the robot arm is drawn.

[0087] After determining the movable range of the dual-arm robot and the range of movement of the two robotic arms, a set of three-dimensional avoidance logic needs to be developed to ensure that the dual-arm robot does not collide with obstacles or other robotic arms when performing tasks; the three-dimensional avoidance logic usually involves the planning, monitoring and adjustment of the motion trajectory of the dual-arm robot and the robotic arm; at this time, determining the three-dimensional avoidance logic usually involves the application of path planning algorithms, collision detection algorithms and real-time control strategies; the path planning algorithm is used to plan the motion trajectory of the dual-arm robot and the robotic arm to ensure that they can safely reach the target position; the collision detection algorithm is used to monitor the distance between the dual-arm robot and the robotic arm and obstacles or other robotic arms in real time, and make adjustments when necessary to avoid collisions; the real-time control strategy is used to quickly respond to and adjust the movement of the dual-arm robot and the robotic arm based on the collision detection results.

[0088] Specifically, for a dual-arm robot, arm 1 is responsible for grasping parts, and arm 2 is responsible for installing parts to the specified position; the movable range of the dual-arm robot has been determined through the previous steps; both arm 1 and arm 2 have three rotational joints, so they both perform rotational motions on three different planes, and these rotational motions constitute the active dimensions of the arms; by simulating and analyzing the kinematic models of arm 1 and arm 2, their ranges of motion at different joint angles are determined; for example, the end effector of arm 1 can perform rotational and translational motions within a certain radius centered on the base, while the range of motion of arm 2 is affected by its structure and joint limitations, and these ranges of motion form the reachable area of the arm in three-dimensional space.

[0089] When formulating the three-dimensional avoidance logic, the first thing to consider is the motion trajectory planning of the dual-arm robot and the robotic arm; for example, when robotic arm 1 needs to grab a part, it will move to the location of the part along a planned path; at the same time, robotic arm 2 will remain in a safe position within its range of activity to avoid collision with robotic arm 1; during the movement of the dual-arm robot, the collision detection algorithm will monitor the distance between the dual-arm robot and the robotic arm and the obstacle in real time, and make adjustments when necessary to avoid collision; for example, if robotic arm 1 detects that the distance to the obstacle is less than the preset safety distance during movement, it will immediately stop moving and make adjustments to ensure that the task can continue safely; through the above steps, the dual-arm robot can determine the activity dimensions and range of its two robotic arms, and formulate corresponding three-dimensional avoidance logic, which provides important support for the precise control and safe obstacle avoidance of the dual-arm robot when performing complex tasks.

[0090] refer to Figure 5In step S14, in the two manipulator arms of the dual-arm robot, the anti-interference logic between the two manipulator arms is determined according to the working paths of the two manipulator arms and the activity ranges of the two manipulator arms;

[0091] In the specific implementation process of the present invention, the specific steps are:

[0092] S141: determining the ranges of movement of the two robotic arms based on the detection of the two robotic arms of the dual-arm robot. At this time, further optimizing the ranges of movement of the two robotic arms to avoid overlapping areas between the ranges of movement of the two robotic arms;

[0093] S142: monitoring the two manipulator arms of the dual-arm robot in real time, collecting the work tasks of the dual-arm robot, and determining sub-work tasks corresponding to the two manipulator arms based on the analysis of the work tasks, where the sub-work tasks corresponding to the two manipulator arms are a first sub-work task and a second sub-work task, respectively;

[0094] S143: Determine the working paths of the two robotic arms according to the first sub-work task, the second sub-work task and the two robotic arms of the dual-arm robot, determine multiple anti-interference warning points according to the working paths of the two robotic arms and the activity range of the two robotic arms, and determine the anti-interference logic between the two robotic arms based on the multiple anti-interference warning points and the training of the two robotic arms. The anti-interference logic ensures that the two robotic arms do not contact each other.

[0095] In an embodiment of the present application, the activity ranges of the two robotic arms are determined based on the detection of the two robotic arms of the dual-arm robot. At this time, the activity ranges of the two robotic arms are further optimized to avoid overlapping areas between the activity ranges of the two robotic arms.

[0096] At this point, a detailed kinematic analysis is performed on the two arms of the dual-arm robot. By considering the joint configuration, joint limitations (such as rotation angle, translation distance, etc.) and external environmental factors (such as obstacle position, workspace limitations, etc.) of the robotic arms, the set of all positions that each robotic arm can reach in three-dimensional space, that is, its range of motion, is determined. This usually requires the use of forward kinematics and inverse kinematics knowledge in dual-arm robotics. At this point, in actual operation, the robotic arm is modeled through simulation software, and its motion state at different joint angles is simulated to calculate the range of motion. In addition, sensors installed on the robotic arm (such as encoders, gyroscopes, etc.) are also used to collect joint angle data in real time, and the current position and range of motion of the robotic arm are calculated through algorithms.

[0097] After determining the range of motion of the two robotic arms, further optimization is required to avoid overlapping or conflicting areas between them. This is achieved by adjusting the joint limits of the robotic arms, changing the base position or orientation of the robotic arms, and replanning the workspace layout. The goal of the optimization is to ensure that the two robotic arms can work independently and efficiently within their respective ranges of motion without interfering with each other. At this time, the method of optimizing the range of motion varies depending on the specific situation. For example, if the ranges of motion of the two robotic arms overlap, consider adjusting the base position or orientation of one of the robotic arms to reduce the overlapping area. In addition, advanced algorithms (such as genetic algorithms, particle swarm algorithms, etc.) are also used to automatically optimize the range of motion of the robotic arms to find the optimal workspace layout.

[0098] Furthermore, the two robotic arms of the dual-arm robot are monitored in real time, the work tasks of the dual-arm robot are collected, and the sub-work tasks corresponding to the two robotic arms are determined based on the analysis of the work tasks. The sub-work tasks corresponding to the two robotic arms are the first sub-work task and the second sub-work task respectively.

[0099] At this time, the two robotic arms of the dual-arm robot are continuously monitored; the monitoring content includes the kinematic parameters such as the position, speed, acceleration, joint angle of the robotic arm, and the data collected by the installed sensors (such as force sensors, visual sensors, etc.); the purpose of real-time monitoring is to grasp the current status of the robotic arm in real time and provide accurate information for subsequent task allocation and path planning. At this time, in actual operation, sensors (such as encoders, gyroscopes, accelerometers, etc.) are installed on the robotic arm to collect kinematic parameters in real time; in addition, machine vision technology (such as cameras, lidar, etc.) is also used to monitor changes in the environment around the robotic arm. These sensors and data acquisition equipment are usually connected to the control system of the dual-arm robot to transmit and process data in real time.

[0100] The dual-arm robot receives work tasks from external systems or users. These tasks are predefined, periodic tasks (such as assembly tasks on a production line) or tasks dynamically generated according to real-time needs (such as picking and packaging specific products according to order requirements). Tasks are usually transmitted to the dual-arm robot's control system in the form of instructions or data packets. At this time, task acquisition is achieved in various ways, such as receiving task instructions from a remote server through wireless communication (such as Wi-Fi, Bluetooth, etc.), or transmitting data to a host connected to the control system through a wired connection (such as Ethernet). In addition, natural language processing (NLP) technology is used to parse user voice commands, or the user's task input is received through a graphical user interface (GUI).

[0101] After receiving the work task, the task needs to be parsed to determine the sub-tasks that each robot arm needs to complete. This step decomposes, sorts, and prioritizes the tasks. The purpose of the analysis is to decompose complex tasks into a series of simple actions or steps performed by the robot arm, and ensure that these actions or steps can be completed efficiently without interfering with each other. At this time, task parsing is achieved by writing dedicated parsing algorithms or using existing task planning software. These algorithms or software can usually analyze the task requirements, the capabilities of the robot arm, and the constraints of the working environment, and then generate one or more feasible task execution plans. When generating plans, the collaborative relationship between the robot arms as well as conflicts and interferences need to be considered.

[0102] Specifically, suppose there is a two-arm robot working in an automated warehouse. Its main task is to pick specific products from the shelves and place them into cartons. Before the task begins, the robot uses sensors installed on the robot arm (such as encoders, gyroscopes, etc.) to monitor parameters such as the position, speed, and joint angle of the robot arm in real time. At the same time, machine vision technology is used to monitor the position of the shelves and cartons and the status of the products. The robot receives picking tasks from the warehouse management system (WMS), including information about the products to be picked (such as product number, quantity, etc.) and the location information of the target cartons.

[0103] The picking task is parsed to determine the sub-tasks that each robot arm needs to complete; for example, robot arm 1 is assigned to pick up goods from the shelf and place them in the temporary storage area; while robot arm 2 is responsible for picking up the goods from the temporary storage area and placing them in the target packaging box; in order to avoid conflicts and interference between the robot arms, their motion paths and collaboration strategies need to be planned according to the layout of the shelves and packaging boxes and the motion capabilities of the robot arms; through the above steps, the two robot arms of the dual-arm robot are successfully monitored in real time, the work tasks are collected, and the tasks are parsed to determine the sub-tasks that each robot arm needs to complete, which provides an important guarantee for the efficient and collaborative execution of the dual-arm robot in subsequent tasks.

[0104] Therefore, the working paths of the two robotic arms are determined according to the first sub-work task, the second sub-work task and the two robotic arms of the dual-arm robot, and multiple anti-interference warning points are determined according to the working paths of the two robotic arms and the activity range of the two robotic arms. The anti-interference logic between the two robotic arms is determined based on the multiple anti-interference warning points and the training of the two robotic arms. The anti-interference logic ensures that the two robotic arms do not contact each other, is compatible with the overall consideration of multiple anti-interference warning points and the training of the two robotic arms, and ensures the accuracy of the anti-interference logic between the two robotic arms.

[0105] At this time, the work path that each of them needs to follow is determined based on the first sub-task and the second sub-task, as well as the current status and capabilities of the two manipulators of the dual-arm robot; the work path should be able to efficiently guide the manipulator from the starting position to the target position while complying with the kinematic constraints of the manipulator and the safety requirements of the working environment; at this time, in actual operation, the determination of the work path usually depends on path planning algorithms, which can consider multiple factors such as the joint limitations of the manipulator, obstacle locations, workspace layout, etc., and generate one or more feasible paths; the path planning algorithm is based on graph search (such as A* algorithm, Dijkstra algorithm, etc.), and is also based on sampling (such as fast random exploration tree RRT, probabilistic roadmap PRM, etc.).

[0106] After determining the working paths of the two robotic arms, a series of anti-interference warning points need to be set along these paths. These warning points are risk points where the robotic arms may collide with other objects (including another robotic arm) during movement. By setting sensors or monitoring mechanisms at these points, the relative position and distance between the robotic arms can be monitored in real time to trigger avoidance actions when necessary. At this time, the determination of anti-interference warning points usually relies on a detailed analysis of the range of motion and working paths of the robotic arms. Simulation software is used to simulate the movement of the robotic arms and calculate their collision risks at different positions. Based on the simulation results, warning points are set in high-risk areas and corresponding sensors or monitoring equipment are configured.

[0107] Anti-interference logic is a set of rules and strategies used to ensure that two robotic arms complete a task without contacting each other; it should be able to monitor the position and speed of the robotic arms in real time, predict potential collision risks, and trigger avoidance actions when necessary; anti-interference logic is based on a variety of algorithms and technical implementations, such as collision detection algorithms, motion planning algorithms, real-time control systems, etc. At this time, in actual operation, the implementation of anti-interference logic usually relies on advanced control systems and algorithms, which can monitor the status of the robotic arms and changes in the surrounding environment in real time, and make decisions based on preset rules and strategies; for example, when it is detected that the distance between the two robotic arms is less than a safety threshold, the control system automatically adjusts the motion path or speed of the robotic arms to avoid collision.

[0108] Specifically, for a dual-arm robot, arm 1 is responsible for grasping parts, and arm 2 is responsible for installing parts in the specified position; based on the current positions and task requirements of arm 1 and arm 2, a path planning algorithm is used to generate their respective work paths; for example, the path of arm 1 is from the parts storage area to the assembly work area, while the path of arm 2 is from the assembly work area to the conveyor belt of the next process.

[0109] By analyzing the working paths and ranges of motion of robot arms 1 and 2, potential collision risk points are identified. For example, near the assembly work area that both robot arms 1 and 2 need to pass through, multiple anti-interference warning points are set up. These warning points are configured to trigger an alarm or avoidance action when the distance between the robot arms is less than a certain threshold. Based on the setting of anti-interference warning points and the movement capabilities of the robot arms, a set of anti-interference logic is formulated. For example, when it is detected that the distance between robot arms 1 and 2 is less than a safety threshold, the control system automatically pauses the movement of one of the robot arms or adjusts its movement path to avoid collision. In addition, visual sensors and machine learning algorithms are used to monitor the relative position and posture between the robot arms in real time, and dynamic avoidance decisions are made based on this information. Through the above steps, the working paths, anti-interference warning points and anti-interference logic of the two robot arms of the dual-arm robot are successfully determined, which provides an important guarantee for the efficient and safe execution of the dual-arm robot in subsequent tasks.

[0110] In one embodiment of the present application, a dual-arm robot is performing an assembly task. Arm 1 is responsible for grasping parts, and Arm 2 is responsible for installing the parts in the designated location. There are some obstacles (such as workbenches, shelves, etc.) and potential conflict areas (such as the assembly area that both Arm 1 and Arm 2 pass through) in the workspace. The warning point matching table is shown in Table 2:

[0111] Table 2 Warning point matching table

[0112] robotic arm Early warning point Relative position (distance, angle) Timestamp 1 A (50cm, 45°) T1 2 A (45cm, 135°) T1+Δt1 1 B (80cm, 90°) T2 2 C (120cm, 0°) T2+Δt2

[0113] refer to Figure 6 In step S15, the multi-dimensional range of motion of the dual-arm robot is determined based on the moving direction of the dual-arm robot, the form of the dual-arm robot, and the range of motion of the two manipulator arms. The dual-arm robot is then triggered to perform multiple obstacle avoidance operations relative to various obstacle areas based on the multi-dimensional range of motion, anti-interference logic, and three-dimensional avoidance logic.

[0114] In the specific implementation process of the present invention, the specific steps are:

[0115] S151: Collect the moving direction of the dual-arm robot, and determine the first-dimensional range of movement according to the moving direction and the shape of the dual-arm robot;

[0116] S152: determining a second-dimensional range of motion of the dual-arm robot according to the moving direction of the dual-arm robot and the ranges of motion of the two manipulator arms, and determining a multi-dimensional range of motion of the dual-arm robot based on the first-dimensional range of motion, the second-dimensional range of motion, and the mapping relationship between the ranges of motion;

[0117] S153: In the multi-dimensional range of activity of the dual-arm robot, the anti-interference logic and the three-dimensional avoidance logic are combined, and a multiple obstacle avoidance mechanism is determined based on the combination of the anti-interference logic, the three-dimensional avoidance logic and the moving state of the dual-arm robot. Based on the multiple obstacle avoidance mechanism, the dual-arm robot is triggered to perform multiple obstacle avoidance relative to each obstacle area.

[0118] In an embodiment of the present application, the moving direction of the dual-arm robot is collected, and the first-dimensional activity range is determined based on the moving direction and the shape of the dual-arm robot. This is compatible with the overall consideration of the moving direction and the shape of the dual-arm robot, and ensures the accuracy of the first-dimensional activity range.

[0119] At this time, sensors or control systems are used to monitor and record the movement direction of the dual-arm robot; the movement direction is straight, curved, rotational, etc., depending on the dual-arm robot's movement capabilities and task requirements; the collected data include the dual-arm robot's speed, acceleration, steering angle, etc., which are crucial for the subsequent determination of the range of activity; at this time, in actual operation, gyroscopes, accelerometers, magnetometers and other sensors are installed on the dual-arm robot to collect movement direction data in real time. These sensors can sense the posture changes of the dual-arm robot and transmit the data to the control system for processing; in addition, machine vision technology is also used to monitor the movement trajectory of the dual-arm robot, thereby indirectly obtaining movement direction information.

[0120] This step involves analyzing the physical form of the dual-arm robot, including the length of the robotic arm, joint configuration, size of the end effector, etc., to determine the basic range of motion of the dual-arm robot when it is stationary; the basic range of motion refers to the spatial area that the robotic arm of the dual-arm robot can reach when it is not moving; at this time, determining the basic range of motion usually requires the use of 3D modeling software or robotic arm simulation tools. These tools can generate reachable space diagrams of the dual-arm robot in different postures based on the geometric dimensions and kinematic parameters of the dual-arm robot; by analyzing these graphics, the basic range of motion of the dual-arm robot when it is stationary is obtained.

[0121] After obtaining the dual-arm robot's moving direction and basic activity range, this step involves combining the two to determine the first-dimensional activity range of the dual-arm robot during movement; the first-dimensional activity range refers to the spatial area that the dual-arm robot can reach in a specific moving direction, which takes into account the influence of the dual-arm robot's mobility and physical form on the reachable space; at this time, determining the first-dimensional activity range usually requires the use of dynamic simulation or path planning algorithms, which can simulate the motion trajectory of the dual-arm robot in different moving directions, and combine the information of the basic activity range to generate a reachable space map of the dual-arm robot during movement; by analyzing these graphs, the first-dimensional activity range of the dual-arm robot in a specific moving direction is obtained.

[0122] Specifically, suppose there is a dual-arm robot working in an automated warehouse. Its main task is to pick goods from the shelves and place them on the conveyor belt. The dual-arm robot moves from the shelf area to the conveyor belt area in a straight line. The gyroscope and accelerometer installed on the dual-arm robot are used to collect data such as the speed, acceleration, and steering angle of the dual-arm robot in real time during the movement. The physical form of the dual-arm robot is modeled using 3D modeling software, including the length of the robotic arm, joint configuration, and size of the end effector. Through simulation analysis, the basic range of motion of the dual-arm robot in a stationary state is obtained, that is, the spatial area that the robotic arm can reach.

[0123] For the first-dimensional range of activity, a dynamic simulation algorithm or path planning software is used to simulate the motion trajectory of the dual-arm robot during linear movement. Combined with the information of the basic range of activity, a map of the reachable space of the dual-arm robot during movement is generated. By analyzing these graphs, the first-dimensional range of activity of the dual-arm robot in the linear movement direction is obtained, that is, the cargo area that the dual-arm robot can reach during movement. Through the above steps, the first-dimensional range of activity of the dual-arm robot in a specific movement direction is successfully determined, providing an important basis for subsequent task planning and obstacle avoidance strategies.

[0124] Furthermore, the second-dimensional activity range of the dual-arm robot is determined according to the moving direction of the dual-arm robot and the activity range of the two robotic arms, and the multi-dimensional activity range of the dual-arm robot is determined based on the first-dimensional activity range, the second-dimensional activity range and the activity range mapping relationship. This is compatible with the overall consideration of the first-dimensional activity range, the second-dimensional activity range and the activity range mapping relationship, ensuring the accuracy of the multi-dimensional activity range of the dual-arm robot.

[0125] At this time, the range of motion of the two arms of the dual-arm robot in a specific moving direction is analyzed; unlike the first dimension (usually referring to the horizontal or linear moving direction), the second dimension range of motion considers the accessible space of the dual-arm robot in the vertical direction, depth direction or other non-linear directions, which usually involves a comprehensive evaluation of the joint motion of the dual-arm robot's arms, the operating range of the end effector, and the existence of obstacles; at this time, determining the second dimension range of motion requires the use of three-dimensional space analysis tools or robotic arm simulation software, which can simulate the robotic arm motion of the dual-arm robot in different postures and positions, and calculate the spatial area that the robotic arm can reach in a specific direction; in addition, the structural limitations of the dual-arm robot itself (such as chassis height, robotic arm joint angle limitations, etc.) and obstacles in the environment need to be considered.

[0126] After determining the range of activity in the first and second dimensions, this step involves integrating the information of these two dimensions to form a more comprehensive description of the range of activity. This usually involves calculating the spatial intersection, union or other logical combination of the two-dimensional range of activity to derive the overall activity capability of the dual-arm robot in multi-dimensional space. At this time, integrating the first and second-dimensional range of activity requires the use of spatial geometry algorithms or three-dimensional modeling software. These tools can handle complex three-dimensional spatial relationships and calculate the intersection, union, etc. of the range of activity of the dual-arm robot in different dimensions. By integrating this information, the overall range of activity of the dual-arm robot in multi-dimensional space is obtained.

[0127] The activity range mapping relationship refers to the process of mapping the activity range of a dual-arm robot in different dimensions into a unified three-dimensional space coordinate system. This step involves converting the activity ranges of the first and second dimensions (and other dimensions) into representations in a unified coordinate system for spatial analysis and task planning. At this time, determining the multi-dimensional activity range usually requires establishing a three-dimensional space model and mapping the activity ranges of the dual-arm robot in different dimensions into this model. This requires the use of spatial coordinate conversion algorithms, three-dimensional visualization tools and other technical means. Through the mapping relationship, the activity ranges of the dual-arm robot in different dimensions are integrated into a unified three-dimensional space, thereby obtaining the overall activity range of the dual-arm robot in the multi-dimensional space.

[0128] Specifically, for a dual-arm robot, its main task is to pick goods from shelves at different heights and place them on the assembly line; determine the second-dimensional range of motion: the dual-arm robot needs to move in the vertical direction to reach shelves at different heights; by analyzing the dual-arm robot's arm joint angle limitations, the operating range of the end effector, and the height distribution of the shelves, determine the second-dimensional range of motion of the dual-arm robot in the vertical direction; for example, the dual-arm robot can reach goods on shelves from the ground to 3 meters high.

[0129] The horizontal range of motion of a dual-arm robot (the first dimension) is limited by the factory layout, the location of the assembly line, and the presence of other dual-arm robots. The vertical range of motion of a dual-arm robot (the second dimension) is also limited by the height of the shelves and the angle of the robot arm joints. By comprehensively evaluating the range of motion in these two dimensions, the overall mobility of the dual-arm robot in multi-dimensional space is derived. For example, a dual-arm robot can move 5 meters horizontally and reach goods on shelves from the ground to 3 meters high in the vertical direction.

[0130] In order to more intuitively represent the activity range of the dual-arm robot in multi-dimensional space, a three-dimensional space model is established, and the activity range of the dual-arm robot in different dimensions is mapped into this model; in this model, the reachable space of the dual-arm robot in the horizontal and vertical directions, as well as the overlapping parts between these spaces, are clearly seen; through the mapping relationship, the overall activity range of the dual-arm robot in multi-dimensional space is obtained, and task planning and obstacle avoidance strategies are formulated based on this; through the above steps, the activity range of the dual-arm robot in multi-dimensional space is successfully determined, providing an important basis for subsequent task execution and obstacle avoidance strategies.

[0131] Therefore, within the multi-dimensional activity range of the dual-arm robot, the anti-interference logic and the three-dimensional avoidance logic are combined, and a multiple obstacle avoidance mechanism is determined based on the combination of the anti-interference logic, the three-dimensional avoidance logic and the moving state of the dual-arm robot. Based on the multiple obstacle avoidance mechanism, the dual-arm robot triggers multiple obstacle avoidance relative to each obstacle area, which is compatible with the overall consideration of the multi-dimensional activity range, anti-interference logic and three-dimensional avoidance logic of the dual-arm robot, ensures the intelligence of the dual-arm robot's multiple obstacle avoidance relative to each obstacle area, and improves the obstacle avoidance accuracy of the dual-arm robot relative to each obstacle area.

[0132] At this time, the anti-interference logic (ensuring that the two arms of the dual-arm robot do not interfere with each other) and the three-dimensional avoidance logic (ensuring that the dual-arm robot does not collide with other obstacles in the environment) are combined; the anti-interference logic usually considers factors such as the motion trajectory, joint angle, and position of the end effector of the robot arm to ensure a safe distance between the robot arms; the three-dimensional avoidance logic considers the overall movement of the dual-arm robot in three-dimensional space, including translation, rotation, etc., to ensure that the dual-arm robot maintains a safe distance from fixed obstacles, moving obstacles, etc. in the environment; at this time, combining these two logics requires the use of technical means such as path planning algorithms, collision detection algorithms, and real-time monitoring systems; the path planning algorithm is used to generate a safe movement path for the dual-arm robot, the collision detection algorithm is used to monitor the distance between the dual-arm robot and obstacles in the environment in real time, and the real-time monitoring system is used to collect and process the motion status data of the dual-arm robot.

[0133] After combining the anti-interference logic and three-dimensional avoidance logic, it is necessary to determine the multiple obstacle avoidance mechanisms based on the real-time movement status of the dual-arm robot (such as position, speed, acceleration, etc.); the multiple obstacle avoidance mechanisms include different obstacle avoidance strategies, such as deceleration to avoid obstacles, detour to avoid obstacles, pause to avoid obstacles, etc. The selection of these strategies depends on factors such as the relative position, speed, acceleration between the dual-arm robot and the obstacle; at this time, determining the multiple obstacle avoidance mechanisms requires the use of decision-making algorithms such as decision trees, fuzzy logic, and neural networks. These algorithms can dynamically select appropriate obstacle avoidance strategies based on the real-time movement status of the dual-arm robot and obstacle information.

[0134] After determining the multiple obstacle avoidance mechanisms, it is necessary to trigger corresponding obstacle avoidance actions based on the real-time motion status of the dual-arm robot and obstacle information. This involves adjusting the speed of the dual-arm robot, changing its movement path, pausing its movement, and other actions to ensure that the dual-arm robot does not collide with any obstacles during the execution of the task; at this time, triggering the multiple obstacle avoidance mechanisms requires the use of hardware devices such as motion controllers and actuators; the motion controller is responsible for receiving instructions from the decision-making algorithm and adjusting the motion state of the dual-arm robot according to these instructions; the actuator is responsible for converting the instructions of the motion controller into actual movement of the dual-arm robot.

[0135] Specifically, suppose there is a dual-arm robot working in a busy automated warehouse. Its main task is to pick goods from the shelves and place them on the conveyor belt. There are multiple shelves, conveyor belts, other dual-arm robots, and obstacles such as people in the warehouse. The dual-arm robot needs to consider both anti-interference between the two robotic arms and three-dimensional avoidance of other obstacles in the environment. For example, when the dual-arm robot moves from one shelf to another, it needs to ensure that the two robotic arms do not collide with each other, and the dual-arm robot itself does not collide with the shelves, conveyor belts, or other dual-arm robots. This requires the use of a path planning algorithm to generate a safe moving path for the dual-arm robot, and a collision detection algorithm to monitor the distance between the dual-arm robot and obstacles in real time.

[0136] Based on the dual-arm robot's real-time movement status and obstacle information, multiple obstacle avoidance mechanisms are determined; for example, when the dual-arm robot approaches a shelf, it needs to slow down to avoid colliding with the shelf; when the dual-arm robot detects that another dual-arm robot is on its path, it needs to detour to avoid colliding with it; when the dual-arm robot detects an emergency ahead (such as a person suddenly breaking into the work area), it needs to immediately suspend movement to ensure safety.

[0137] Once the multiple obstacle avoidance mechanisms are determined, the corresponding obstacle avoidance actions need to be triggered based on the dual-arm robot's real-time motion status and obstacle information; for example, when the dual-arm robot needs to slow down, the motion controller adjusts the speed of the dual-arm robot; when the dual-arm robot needs to detour, the path planning algorithm generates a new movement path; when the dual-arm robot needs to pause movement, the actuator immediately stops all movements of the dual-arm robot; through the above steps, the anti-interference logic and three-dimensional avoidance logic are successfully combined, and the multiple obstacle avoidance mechanisms are determined based on the real-time movement status of the dual-arm robot, thereby ensuring the safety of the dual-arm robot during the execution of the task.

[0138] In one embodiment of the present application, an obstacle avoidance strategy matching table is constructed. The obstacle avoidance strategy matching table lists the obstacle avoidance strategies that the dual-arm robot should adopt in different situations. The columns of the matching table include anti-interference logic, three-dimensional avoidance logic, and the movement status of the dual-arm robot (such as speed, direction, etc.), while the rows correspond to different obstacle avoidance strategies (such as deceleration, detour, pause, etc.). The obstacle avoidance strategy matching table is shown in Table 3:

[0139] Table 3. Obstacle avoidance strategy matching table

[0140] No anti-interference requirements, no three-dimensional avoidance requirements, low-speed movement Normal driving There is a need for anti-interference, no need for three-dimensional avoidance, and medium-speed movement Slow down and avoid obstacles No anti-interference requirements, three-dimensional avoidance requirements, high-speed movement Obstacle avoidance There are anti-interference requirements, three-dimensional avoidance requirements, and any speed Pause obstacle avoidance

[0141] Assume that a dual-arm robot detects both anti-interference and three-dimensional avoidance requirements while performing a task and is moving at a medium speed. According to the matching table, the dual-arm robot will adopt a "pause obstacle avoidance" strategy to ensure safety.

[0142] See also Figure 7 , Figure 7 : is a schematic diagram of the structural composition of the obstacle avoidance system of the dual-arm robot in an embodiment of the present invention; the obstacle avoidance system of the dual-arm robot includes:

[0143] A moving path module 21 is used to determine the actual moving path of the dual-arm robot according to the current position and target position of the dual-arm robot;

[0144] an obstacle area module 22 for determining current positions of a plurality of obstacles based on obstacle detection in an actual moving path of the dual-arm robot, and determining a plurality of obstacle areas according to the current positions of the plurality of obstacles and the shapes of the plurality of obstacles;

[0145] A three-dimensional avoidance module 23 is used to determine a three-dimensional avoidance logic according to the moving position of the dual-arm robot, the spatial positions of the two manipulator arms of the dual-arm robot, and multiple obstacle areas during the movement of the dual-arm robot;

[0146] The anti-interference logic module 24 is used to determine the anti-interference logic between the two manipulator arms of the dual-arm robot according to the working paths and the movable ranges of the two manipulator arms;

[0147] The multiple obstacle avoidance module 25 is used to determine the multi-dimensional activity range of the dual-arm robot based on the moving direction of the dual-arm robot, the shape of the dual-arm robot and the activity range of the two robotic arms, and trigger the multiple obstacle avoidance of the dual-arm robot relative to each obstacle area based on the multi-dimensional activity range of the dual-arm robot, anti-interference logic and three-dimensional avoidance logic.

[0148] The technical features of the above embodiments are arbitrarily combined. In order to make the description more concise, not all combinations of the technical features in the above embodiments are described. However, as long as there is no technical contradiction in the combination of these technical features, they should be considered as the main scope recorded in this specification.

Claims

1. A dual-arm robot obstacle avoidance method, characterized in that: include: Determining an actual moving path of the dual-arm robot according to the current position and target position of the dual-arm robot, including: collecting the current position and target position of the dual-arm robot, determining a moving space of the dual-arm robot according to the current position and target position of the dual-arm robot, and determining a plurality of alternative moving paths based on the moving space and the previous moving paths of the dual-arm robot; determining three better moving paths based on preliminary screening of the detection of the plurality of alternative moving paths, the matching coefficients of the three better moving paths and the dual-arm robot satisfying a preset matching coefficient threshold; determining a moving distance of each better moving path based on the detection of the three better moving paths, and determining the actual moving path of the dual-arm robot according to the moving distance of each better moving path, the shape of the dual-arm robot, and the power of the dual-arm robot; Determining current positions of a plurality of obstacles based on obstacle detection in an actual moving path of the dual-arm robot, and determining a plurality of obstacle areas according to the current positions of the plurality of obstacles and the shapes of the plurality of obstacles; During the movement of the dual-arm robot, a three-dimensional avoidance logic is determined based on the movement position of the dual-arm robot, the spatial positions of the two arms of the dual-arm robot, and multiple obstacle areas; In the two manipulator arms of the dual-arm robot, the anti-interference logic between the two manipulator arms is determined according to the working paths and the activity ranges of the two manipulator arms; The multi-dimensional activity range of the dual-arm robot is determined based on the movement direction of the dual-arm robot, the shape of the dual-arm robot and the activity range of the two robotic arms, and the multiple obstacle avoidance of the dual-arm robot relative to each obstacle area is triggered based on the multi-dimensional activity range of the dual-arm robot, anti-interference logic and three-dimensional avoidance logic.

2. The obstacle avoidance method of the dual-arm robot according to claim 1, characterized in that: The obstacle detection based on the actual moving path of the dual-arm robot determines the current positions of the multiple obstacles, and determines the multiple obstacle areas according to the current positions of the multiple obstacles and the shapes of the multiple obstacles, including: collecting an actual movement path of the dual-arm robot, marking a plurality of obstacles in the actual movement path based on the detection of the actual movement path of the dual-arm robot, and determining current positions of the plurality of obstacles; the current positions of the plurality of obstacles include horizontal positions and height positions of the plurality of obstacles; The dual-arm robot is equipped with a radar, and collects corresponding point cloud sets based on the radar's detection of each obstacle, and determines the shape and type of the obstacle based on the recognition of each point cloud set; The activity influence range corresponding to the obstacle is determined based on the shape and type of the obstacle, and multiple obstacle areas are determined according to the horizontal positions and height positions of the multiple obstacles and the activity influence ranges corresponding to the obstacles.

3. The obstacle avoidance method of the dual-arm robot according to claim 1, characterized in that: During the movement of the dual-arm robot, the three-dimensional avoidance logic is determined according to the movement position of the dual-arm robot, the spatial positions of the two manipulator arms of the dual-arm robot, and multiple obstacle areas, including: Monitor the movement of the dual-arm robot in real time, collect the moving position of the dual-arm robot, and determine multiple obstacle areas around the dual-arm robot based on the peripheral detection of the moving position of the dual-arm robot. At this time, the movable range of the dual-arm robot during the movement is determined based on the detection of the dual-arm robot and multiple obstacle areas around the dual-arm robot.

4. The obstacle avoidance method for a dual-arm robot according to claim 3, characterized in that: During the movement of the dual-arm robot, determining the three-dimensional avoidance logic according to the movement position of the dual-arm robot, the spatial positions of the two mechanical arms of the dual-arm robot, and the plurality of obstacle areas also includes: In a dual-arm robot, a corresponding dual-arm robot coordinate system is constructed based on the dual-arm robot, and the spatial positions of the two robotic arms of the dual-arm robot are determined according to the two robotic arms of the dual-arm robot and the dual-arm robot coordinate system; The activity dimensions of the two robotic arms of the dual-arm robot are determined based on the traversal of the two robotic arms, and the activity ranges of the two robotic arms of the dual-arm robot are determined based on the activity dimensions of the two robotic arms and the relative distance between the two robotic arms. The three-dimensional avoidance logic is determined based on the movable range of the dual-arm robot during movement and the activity ranges of the two robotic arms of the dual-arm robot.

5. The obstacle avoidance method for a dual-arm robot according to claim 1, characterized in that: The method of determining the anti-interference logic between the two robotic arms of the dual-arm robot according to the working paths and the movable ranges of the two robotic arms includes: The moving ranges of the two robotic arms are determined based on the detection of the two robotic arms of the dual-arm robot. At this time, the moving ranges of the two robotic arms are further optimized to avoid overlapping areas between the moving ranges of the two robotic arms.

6. The obstacle avoidance method for a dual-arm robot according to claim 5, characterized in that: The method further comprises: determining the anti-interference logic between the two robotic arms of the dual-arm robot according to the working paths and the movable ranges of the two robotic arms; Real-time monitoring of the two robotic arms of the dual-arm robot, collecting the work tasks of the dual-arm robot, and determining the sub-work tasks corresponding to the two robotic arms based on the analysis of the work tasks, where the sub-work tasks corresponding to the two robotic arms are the first sub-work task and the second sub-work task respectively; The working paths of the two robotic arms are determined according to the first sub-work task, the second sub-work task and the two robotic arms of the dual-arm robot, and multiple anti-interference warning points are determined according to the working paths of the two robotic arms and the activity range of the two robotic arms. The anti-interference logic between the two robotic arms is determined based on the multiple anti-interference warning points and the training of the two robotic arms, and the anti-interference logic ensures that the two robotic arms do not contact each other.

7. The obstacle avoidance method for a dual-arm robot according to claim 1, characterized in that: The method of determining the multi-dimensional range of motion of the dual-arm robot based on the moving direction of the dual-arm robot, the form of the dual-arm robot, and the range of motion of the two manipulator arms, and triggering the dual-arm robot to avoid multiple obstacles relative to various obstacle areas based on the multi-dimensional range of motion of the dual-arm robot, anti-interference logic, and three-dimensional avoidance logic, includes: Collect the moving direction of the dual-arm robot, and determine the first-dimensional range of motion based on the moving direction and shape of the dual-arm robot; The second dimensional range of motion of the dual-arm robot is determined according to the moving direction of the dual-arm robot and the range of motion of the two robotic arms, and the multi-dimensional range of motion of the dual-arm robot is determined based on the first dimensional range of motion, the second dimensional range of motion and the mapping relationship between the ranges of motion.

8. The obstacle avoidance method for a dual-arm robot according to claim 7, characterized in that: The method further includes determining the multi-dimensional range of motion of the dual-arm robot based on the moving direction of the dual-arm robot, the form of the dual-arm robot, and the range of motion of the two manipulator arms, and triggering multiple obstacle avoidance of the dual-arm robot relative to various obstacle areas based on the multi-dimensional range of motion of the dual-arm robot, anti-interference logic, and three-dimensional avoidance logic. In the multi-dimensional activity range of the dual-arm robot, the anti-interference logic and the three-dimensional avoidance logic are combined, and a multiple obstacle avoidance mechanism is determined based on the combination of the anti-interference logic, the three-dimensional avoidance logic and the movement state of the dual-arm robot. Based on the multiple obstacle avoidance mechanism, the dual-arm robot is triggered to avoid multiple obstacles relative to each obstacle area.

9. An obstacle avoidance system for a dual-arm robot, characterized in that: The dual-arm robot obstacle avoidance system is applied to the dual-arm robot obstacle avoidance method according to any one of claims 1 to 8, and the dual-arm robot obstacle avoidance system includes: A movement path module is used to determine the actual movement path of the dual-arm robot based on the current position and target position of the dual-arm robot, including: collecting the current position and target position of the dual-arm robot, determining the movement space of the dual-arm robot based on the current position and target position of the dual-arm robot, and determining multiple alternative movement paths based on the movement space and the previous movement paths of the dual-arm robot; determining three better movement paths based on preliminary screening of the detection of multiple alternative movement paths, and the matching coefficients of the three better movement paths and the dual-arm robot meet a preset matching coefficient threshold; determining the moving distance of each better movement path based on the detection of the three better movement paths, and determining the actual movement path of the dual-arm robot based on the moving distance of each better movement path, the shape of the dual-arm robot and the power of the dual-arm robot; An obstacle area module, configured to determine current positions of a plurality of obstacles based on obstacle detection in an actual moving path of the dual-arm robot, and to determine a plurality of obstacle areas according to the current positions of the plurality of obstacles and the shapes of the plurality of obstacles; A three-dimensional avoidance module is used to determine the three-dimensional avoidance logic according to the moving position of the dual-arm robot, the spatial position of the two arms of the dual-arm robot, and multiple obstacle areas during the movement of the dual-arm robot; An anti-interference logic module is used to determine the anti-interference logic between the two robotic arms of the dual-arm robot based on the working paths and activity ranges of the two robotic arms; The multi-obstacle avoidance module is used to determine the multi-dimensional activity range of the dual-arm robot based on the dual-arm robot's movement direction, the dual-arm robot's shape and the activity range of the two robotic arms, and trigger the dual-arm robot's multi-obstacle avoidance relative to each obstacle area based on the dual-arm robot's multi-dimensional activity range, anti-interference logic and three-dimensional avoidance logic.

Images