Aerial dual-track intelligent positioning method and system based on visual recognition

Abstract

The present invention provides an aerial double-track intelligent positioning method and system based on visual recognition, which acquire and analyze robot motion data in the ground space below the aerial double-track network, predict the motion change trends of all robots in the ground space, and generate a robot dynamic distribution map; based on the object transportation data inside the aerial double-track network and the robot dynamic distribution map, a target robot that can dock with the target object transported in the air is selected, and the motion state data of the target object in the air is obtained by visual recognition and tracking, and the target object is matched with the target robot to provide a guarantee for the process of docking the target object from the air to the ground; based on the motion state data, the target robot is controlled to maintain a relative relationship with the target object, and the target robot is visually positioned, and the target object is adjusted to land on the target robot's motion state, thereby improving the flexibility and reliability of aerial double-track transportation of objects.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to an aerial double-track intelligent positioning method and system based on visual recognition. Background Art

[0002] Intelligent industrial production has become the mainstream production model. In the air area and ground area of ​​the production workshop, an aerial double-track network and a robot group are arranged respectively. The aerial double-track network is used to transport parts and the robot group is used to assemble parts. This can realize three-dimensional operation in the production workshop, improve the space utilization rate of the production workshop, realize the dual-track synchronous operation of parts transportation and assembly, and effectively avoid the traditional production workshop that concentrates the transportation and assembly of parts in a limited range of ground areas. In order to ensure the effective and smooth connection of parts from the air area to the ground area for assembly, it is usually necessary to drop the parts from the air area to the ground area at a fixed ground position. Although this can ensure the normal and reliable circulation operation of parts, it reduces the flexibility of assembly and use of parts, cannot meet the needs of different assembly processes, and cannot dock parts from the air area to the ground area at any point in the ground area. Summary of the invention

[0003] The purpose of the present invention is to provide an aerial dual-track intelligent positioning method and system based on visual recognition, obtain and analyze robot motion data in the ground space below the aerial dual-track network, predict the motion change trends of all robots in the ground space, generate a robot dynamic distribution map, and perform a full-range dynamic prediction of the motion conditions of robots performing work tasks in the ground space, so as to provide a reference for the subsequent selection of robots for docking with objects transported in the air; based on the object transportation data and the robot dynamic distribution map inside the aerial dual-track network, select a target robot that can dock with the target object transported in the air, visually identify and track the motion state data of the target object in the air, match the target object with the target robot, and provide a guarantee for the process of docking the target object from the air to the ground; also based on the motion state data, control the target robot to maintain a relative relationship with the target object, visually locate the target robot, adjust the target object to land to the motion state of the target robot, improve the flexibility and reliability of aerial dual-track transportation of objects, and meet the needs of calling objects at different positions.

[0004] The present invention is achieved through the following technical solutions:

[0005] The aerial dual-track intelligent positioning method based on visual recognition includes:

[0006] Acquire robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space; and generate a dynamic distribution map of the robots in the ground space based on the motion change trends;

[0007] Acquire the object transportation data within the aerial double-track network, and select a target robot that can dock with the target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data; perform visual recognition and tracking on the target object to determine the motion state data of the target object in the aerial double-track network;

[0008] Based on the motion state data, the target robot is controlled to maintain a corresponding relative relationship with the target object; when the target object moves to the destination inside the aerial double-track network, the target robot is visually positioned to adjust the motion state of the target object to land on the target robot.

[0009] Optionally, obtaining robot motion data of the ground space below the aerial double-track network, analyzing the motion data, and predicting the motion change trends of all robots in the ground space; and generating a dynamic distribution map of the robots in the ground space based on the motion change trends, including:

[0010] Acquire the movement speed setting data and movement direction setting data of all robots in the ground space below the aerial double-track network; wherein the movement speed setting data refers to the movement speed change data set for the robot within a period of time in the future; the movement direction setting data refers to the movement direction change data set for the robot within a period of time in the future;

[0011] The motion speed setting data and the motion direction setting data are integrated and analyzed to predict the motion path change trend of all robots in the ground space within a future time interval; wherein the motion path change trend includes the position coordinate information of all characteristic time points of the robots in the ground space within a future time interval;

[0012] The motion path change trends of all robots are processed by spatial coordinate system transformation and time domain superposition integration to generate a dynamic distribution map of the robots in the ground space; wherein the dynamic distribution map of the robots includes a number of overall distribution sub-maps of the robots corresponding to all the characteristic time points within a future period of time.

[0013] Optionally, obtaining object transportation data within the aerial double-track network, selecting a target robot capable of docking with a target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data; and visually identifying and tracking the target object to determine motion state data of the target object in the aerial double-track network include:

[0014] Obtaining object transportation occupancy time data of each of all double-track tracks under the aerial double-track network, and determining the idle time data of each of all double-track tracks based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the transportation of the target object in the aerial double-track network;

[0015] Comparing the robot dynamic distribution map with the transport track path in the time domain, determining the time overlap between all the robots in the ground space and the transport track path; based on the time overlap, selecting one of the robots as a target robot that can dock with the target object;

[0016] Visual recognition and tracking are performed on the target object to determine the movement speed data and movement position data of the target object in the transportation track path of the aerial double-track network.

[0017] Optionally, based on the motion state data, controlling the target robot and the target object to maintain a corresponding relative relationship; when the target object moves to a destination inside the aerial double-track network, visually positioning the target robot, thereby adjusting the motion state of the target object landing on the target robot, includes:

[0018] Based on the movement speed data and movement position data of the target object in the transportation track of the aerial double-track network, the movement speed of the target robot is controlled so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transportation task;

[0019] When the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform object receiving operations; based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network is adjusted to land the target object.

[0020] Optionally, determining a coverage space range in which the target robot can perform an object receiving operation includes:

[0021] Extracting the current speed of the target object;

[0022] extracting volume data and mass data of the target object;

[0023] Obtaining a coverage space range adjustment coefficient according to the current movement speed of the target object combined with the volume data and mass data of the target object;

[0024] The coverage space range adjustment coefficient is obtained by the following formula:

[0025]

[0026] Wherein, S represents the coverage space range adjustment coefficient; v represents the current movement speed of the target object; V and m represent the volume data and mass data of the target object; α and k represent the first adjustment coefficient and the second adjustment coefficient; and the value range of the first adjustment coefficient is 0.13-0.28; the value range of the second adjustment coefficient is 1.17-1.32;

[0027] Extracting the body coverage radius of the target robot;

[0028] The basic coverage space range is obtained according to the body coverage radius of the target robot combined with the coverage space range adjustment coefficient; wherein the area corresponding to the basic coverage space range is obtained by the following formula:

[0029]

[0030] Where A represents the area corresponding to the basic coverage space range; S represents the coverage space range adjustment coefficient; R represents the body coverage radius of the target robot; P v Indicates the minimum speed change ratio of the current target robot; n indicates the unit time that the target robot has been running, and the unit time is 1s; v i represents the running speed of the target robot corresponding to the i-th unit time; v i-1 represents the running speed of the target robot corresponding to the i-1th unit time;

[0031] The running speed and running speed acceleration response time of the current target robot are extracted to adjust the basic coverage space range to obtain the adjusted coverage space range; wherein the area corresponding to the adjusted coverage space range is obtained by the following formula:

[0032]

[0033] Among them, A t represents the area corresponding to the adjusted coverage space range; A represents the area corresponding to the basic coverage space range; v represents the current movement speed of the target object; v mrepresents the running speed of the target robot; r represents the number of accelerations during the running process of the target robot; t i represents the acceleration response time of the running speed corresponding to the i-th acceleration; t max represents the maximum value of the acceleration response time of the running speed corresponding to r accelerations; t m Indicates the running speed acceleration response time corresponding to the maximum acceleration of the target robot.

[0034] The aerial dual-track intelligent positioning system based on visual recognition includes:

[0035] A robot motion change prediction module is used to obtain the robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space;

[0036] A robot dynamic distribution determination module, used to generate a robot dynamic distribution map of the ground space based on the movement change trend;

[0037] A docking robot selection module is used to obtain the object transportation data within the aerial double-track network, and select a target robot that can dock with the target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data;

[0038] An object motion visual recognition module is used to perform visual recognition and tracking on the target object to determine the motion state data of the target object in the aerial dual-track network;

[0039] A robot motion control module, used for controlling the target robot to maintain a corresponding relative relationship with the target object based on the motion state data;

[0040] The object landing adjustment module is used to visually locate the target robot after the target object moves to the destination inside the aerial double-track network, so as to adjust the motion state of the target object landing to the target robot.

[0041] Optionally, the robot motion change prediction module is used to obtain robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space, including:

[0042] Acquire the movement speed setting data and movement direction setting data of all robots in the ground space below the aerial double-track network; wherein the movement speed setting data refers to the movement speed change data set for the robot within a period of time in the future; the movement direction setting data refers to the movement direction change data set for the robot within a period of time in the future;

[0043] The motion speed setting data and the motion direction setting data are integrated and analyzed to predict the motion path change trend of all robots in the ground space within a future time interval; wherein the motion path change trend includes the position coordinate information of all characteristic time points of the robots in the ground space within a future time interval;

[0044] The robot dynamic distribution determination module is used to generate a robot dynamic distribution map of the ground space based on the movement change trend, including:

[0045] The motion path change trends of all robots are processed by spatial coordinate system transformation and time domain superposition integration to generate a dynamic distribution map of the robots in the ground space; wherein the dynamic distribution map of the robots includes a number of overall distribution sub-maps of the robots corresponding to all the characteristic time points within a future period of time.

[0046] Optionally, the docking robot selection module is used to obtain object transportation data within the aerial double-track network, and select a target robot capable of docking with a target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data, including:

[0047] Obtaining object transportation occupancy time data of each of all double-track tracks under the aerial double-track network, and determining the idle time data of each of all double-track tracks based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the transportation of the target object in the aerial double-track network;

[0048] Comparing the robot dynamic distribution map with the transport track path in the time domain, determining the time overlap between all the robots in the ground space and the transport track path; based on the time overlap, selecting one of the robots as a target robot that can dock with the target object;

[0049] The object motion visual recognition module is used to perform visual recognition and tracking on the target object, and determine the motion state data of the target object in the aerial dual-track network, including:

[0050] Visual recognition and tracking are performed on the target object to determine the movement speed data and movement position data of the target object in the transportation track path of the aerial double-track network.

[0051] Optionally, the robot motion control module is used to control the target robot to maintain a corresponding relative relationship with the target object based on the motion state data, including:

[0052] Based on the movement speed data and movement position data of the target object in the transportation track of the aerial double-track network, the movement speed of the target robot is controlled so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transportation task;

[0053] The object landing adjustment module is used to perform visual positioning on the target robot after the target object moves to the destination inside the aerial double-track network, so as to adjust the motion state of the target object landing to the target robot, including:

[0054] When the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform object receiving operations; based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network is adjusted to land the target object.

[0055] Optionally, determining a coverage space range in which the target robot can perform an object receiving operation includes:

[0056] Extracting the current speed of the target object;

[0057] extracting volume data and mass data of the target object;

[0058] Obtaining a coverage space range adjustment coefficient according to the current movement speed of the target object combined with the volume data and mass data of the target object;

[0059] The coverage space range adjustment coefficient is obtained by the following formula:

[0060]

[0061] Wherein, S represents the coverage space range adjustment coefficient; v represents the current movement speed of the target object; V and m represent the volume data and mass data of the target object; α and k represent the first adjustment coefficient and the second adjustment coefficient; and the value range of the first adjustment coefficient is 0.13-0.28; the value range of the second adjustment coefficient is 1.17-1.32;

[0062] Extracting the body coverage radius of the target robot;

[0063] The basic coverage space range is obtained according to the body coverage radius of the target robot combined with the coverage space range adjustment coefficient; wherein the area corresponding to the basic coverage space range is obtained by the following formula:

[0064]

[0065] Where A represents the area corresponding to the basic coverage space range; S represents the coverage space range adjustment coefficient; R represents the body coverage radius of the target robot; P v Indicates the minimum speed change ratio of the current target robot; n indicates the unit time that the target robot has been running, and the unit time is 1s; v i represents the running speed of the target robot corresponding to the i-th unit time; v i-1 represents the running speed of the target robot corresponding to the i-1th unit time;

[0066] The running speed and running speed acceleration response time of the current target robot are extracted to adjust the basic coverage space range to obtain the adjusted coverage space range; wherein the area corresponding to the adjusted coverage space range is obtained by the following formula:

[0067]

[0068] Among them, A t represents the area corresponding to the adjusted coverage space range; A represents the area corresponding to the basic coverage space range; v represents the current movement speed of the target object; v m represents the running speed of the target robot; r represents the number of accelerations during the running process of the target robot; t i represents the acceleration response time of the running speed corresponding to the i-th acceleration; t max represents the maximum value of the acceleration response time of the running speed corresponding to r accelerations; t m Indicates the running speed acceleration response time corresponding to the maximum acceleration of the target robot.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The aerial dual-track intelligent positioning method and system based on visual recognition provided by the present application acquire and analyze the robot motion data of the ground space below the aerial dual-track network, predict the motion change trends of all robots in the ground space, generate a robot dynamic distribution map, and make a full-range dynamic prediction of the motion of robots performing work tasks in the ground space, so as to provide a reference for the subsequent selection of robots to dock with objects transported in the air; based on the object transportation data and the robot dynamic distribution map within the aerial dual-track network, select a target robot that can dock with the target object transported in the air, and visually identify and track the motion state data of the target object in the air, match the target object with the target robot, and provide a guarantee for the process of docking the target object from the air to the ground; also based on the motion state data, control the target robot to maintain a relative relationship with the target object, visually locate the target robot, adjust the target object to land to the motion state of the target robot, improve the flexibility and reliability of the aerial dual-track transportation of objects, and meet the needs of calling objects at different locations. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0072] Figure 1 A schematic diagram of the flow chart of the aerial dual-track intelligent positioning method based on visual recognition provided by the present invention.

[0073] Figure 2 A schematic diagram of the structure of the aerial dual-track intelligent positioning system based on visual recognition provided by the present invention. DETAILED DESCRIPTION

[0074] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0075] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0076] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0077] See also Figure 1 As shown, an aerial dual-track intelligent positioning method based on visual recognition is provided in one embodiment of the present application. The aerial dual-track intelligent positioning method based on visual recognition includes:

[0078] Acquire the robot motion data of the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space; based on the motion change trends, generate a dynamic distribution map of the robots in the ground space;

[0079] Obtaining object transportation data within the aerial double-track network, and selecting a target robot that can dock with a target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data; performing visual recognition and tracking on the target object to determine motion state data of the target object in the aerial double-track network;

[0080] Based on the motion state data, the target robot is controlled to maintain a corresponding relative relationship with the target object; when the target object moves to the destination inside the aerial double-track network, the target robot is visually positioned to adjust the motion state of the target object to land on the target robot.

[0081] The beneficial effects of the above embodiments are as follows: the aerial dual-track intelligent positioning method based on visual recognition acquires and analyzes the robot motion data of the ground space below the aerial dual-track network, predicts the motion change trends of all robots in the ground space, generates a robot dynamic distribution map, and performs a full-range dynamic prediction of the motion of robots performing work tasks in the ground space, providing a reference for the subsequent selection of robots to dock with objects transported in the air; based on the object transportation data and the robot dynamic distribution map within the aerial dual-track network, selects a target robot that can dock with the target object transported in the air, and visually identifies and tracks the motion state data of the target object in the air, matches the target object with the target robot, and provides a guarantee for the process of docking the target object from the air to the ground; based on the motion state data, controls the target robot to maintain a relative relationship with the target object, visually locates the target robot, adjusts the target object to land on the target robot, improves the flexibility and reliability of aerial dual-track transportation of objects, and meets the needs of calling objects at different locations.

[0082] In another embodiment, robot motion data of a ground space below an aerial double-track network is obtained, the motion data is analyzed, and the motion change trends of all robots in the ground space are predicted; based on the motion change trends, a dynamic distribution map of the robots in the ground space is generated, including:

[0083] Obtaining the movement speed setting data and movement direction setting data of each of the robots in the ground space below the aerial double-track network; wherein the movement speed setting data refers to the movement speed change data set for the robot within a period of time in the future; and the movement direction setting data refers to the movement direction change data set for the robot within a period of time in the future;

[0084] The motion speed setting data and the motion direction setting data are integrated and analyzed to predict the motion path change trend of all robots in the ground space within a future time interval; wherein the motion path change trend includes the position coordinate information of all characteristic time points of the robot in the ground space within the future time interval;

[0085] The motion path change trends of all robots are processed by spatial coordinate system transformation and time domain superposition integration to generate a dynamic distribution map of the robots in the ground space; wherein the dynamic distribution map of the robots includes a number of overall distribution sub-maps of the robots corresponding to all the characteristic time points within a certain period of time in the future.

[0086] The beneficial effect of the above embodiment is that an aerial double-track network and a robot group are arranged in the upper area and the ground area of ​​the production workshop, respectively, wherein the aerial double-track network includes a number of double-track tracks distributed continuously and / or crosswise in the upper area, and the corresponding double-track tracks can be used to transport objects such as parts to different locations in the upper area, and do not interfere with the robots moving in the ground area; the robot group includes a number of robots that can move freely in the ground area and perform tasks independently, each robot can independently perform corresponding operations, and can also dock objects transported in the aerial double-track network according to corresponding instructions (i.e., receive objects dropped from the aerial double-track network). All robots under the robot group are free to move in the ground area, so that each robot is located in a different position in the ground area at different time points. In order to accurately dispatch the appropriate robot to dock the object in a short time, it is necessary to first fully grasp the movement of all robots in the ground area, so that robots with a closer straight-line distance to the object or a faster movement speed in the ground area can be preferentially dispatched to dock the object. To this end, the motion speed setting data and motion direction setting data of all robots in the ground space below the aerial double-track network are obtained, and the motion speed change data and motion direction change data corresponding to the motion speed and motion direction set by each robot for the corresponding operation in the future period of time are comprehensively characterized. In addition, the motion speed setting data and motion direction setting data of each robot are integrated and analyzed to predict the motion path change trend of each robot in the ground space in the future period of time, and the position coordinates of all characteristic time points (such as several time points at fixed time intervals) of each robot in the future period of time are accurately calibrated in the ground space, so as to quantitatively characterize the motion position of each robot in the ground space at the time domain level. In addition, the motion path change trend of all robots is processed by a spatial coordinate system-transformation and time domain superposition integration, and a dynamic distribution map of the robots in the ground space is generated, so that the distribution position of all robots in the ground area at each characteristic time point in the future period of time can be globally characterized, providing a reliable basis for the subsequent selection of suitable robots to dock with objects transported by the aerial double-track network.

[0087] In another embodiment, object transportation data within the aerial double-track network is obtained, and based on the robot dynamic distribution map and the object transportation data, a target robot that can dock with a target object transported within the aerial double-track network is selected; and visual recognition and tracking of the target object is performed to determine the motion state data of the target object in the aerial double-track network, including:

[0088] Obtaining object transportation occupancy time data of each of all double-track tracks under the aerial double-track network, and determining the idle time data of each of all double-track tracks based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the transportation target object in the aerial double-track network;

[0089] Comparing the robot dynamic distribution map with the transport track path in the time domain, determining the time overlap between all the robots in the ground space and the transport track path; based on the time overlap, selecting one of the robots as a target robot that can dock with the target object;

[0090] The target object is visually identified and tracked to determine the movement speed data and movement position data of the target object in the transportation track path of the aerial double-track network.

[0091] The beneficial effects of the above embodiments are that the aerial double-track network includes a number of double-track tracks that are continuously and / or cross-distributed in the overhead area, and multiple objects are transported within the aerial double-track network in the same time period. In order to ensure that all objects can be effectively and continuously transported within the aerial double-track network, it is necessary to perform time domain analysis on the object transportation conditions of all double-track tracks, specifically obtain the object transportation occupancy time data of each double-track track under the aerial double-track network, and determine the idle time data of each double-track track based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the target object transported in the aerial double-track network. In this way, the transportation track path corresponding to the target object transported in the aerial double-track network can be planned based on the time interval when all double-track tracks are in an idle state, effectively avoiding transportation path conflicts between the target object and other objects during the transportation process within the aerial double-track network. The robot dynamic distribution map is also compared with the transport track path in the time domain to determine the time overlap between all the robots in the ground space and the transport track path. The time overlap refers to the time overlap between the time interval of each robot in the working state and the time interval corresponding to the transport track path transporting the target object. The robot with the smallest time overlap is used as the target robot that can dock with the target object. The robot with the smallest time overlap has a low probability of being in the working state during the time interval corresponding to the transport track path transporting the target object, which facilitates the subsequent scheduling of the robot to dock with the target object at any time. The target object is also visually identified and tracked to determine the movement speed data and movement position data of the target object in the transport track path of the aerial double-track network, so that the transportation status of the target object in the aerial double-track network can be characterized in multiple dimensions, providing a basis for the subsequent realization of the motion linkage control of the target robot and the target object.

[0092] In another embodiment, based on the motion state data, the target robot is controlled to maintain a corresponding relative relationship with the target object; when the target object moves to a destination inside the aerial double-track network, the target robot is visually positioned to adjust the motion state of the target object to land on the target robot, including:

[0093] Based on the movement speed data and movement position data of the target object in the transportation track of the aerial double-track network, the movement speed of the target robot is controlled so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transportation task;

[0094] When the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform object receiving operations; based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network is adjusted to perform the landing movement of the target object.

[0095] The beneficial effect of the above embodiment is to control the movement speed of the target robot with reference to the movement speed data and movement position data of the target object in the transport track of the aerial double-track network, so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transport task, such as maintaining the distance between the target robot and the target object in the corresponding distance range during the execution of the original transport task, ensuring that the target robot can change its own working state in a short time and dock with the target object. Also, when the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform the object receiving operation, and the coverage space range can be but is not limited to the operating space range formed by the mechanical arm of the target robot itself. And based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network to land the target object will be adjusted to ensure that the target object can be accurately aligned with the working space range formed by the robotic arm of the target robot itself during the process of landing from the aerial double-track network to the ground area, so as to avoid the target object from falling due to failure to be accurately and smoothly received by the target robot during the landing process.

[0096] In another embodiment, determining the coverage space range in which the target robot can perform an object receiving operation includes:

[0097] Extracting the current speed of the target object;

[0098] extracting volume data and mass data of the target object;

[0099] Obtaining a coverage space range adjustment coefficient according to the current movement speed of the target object combined with the volume data and mass data of the target object;

[0100] The coverage space range adjustment coefficient is obtained by the following formula:

[0101]

[0102] Wherein, S represents the coverage space range adjustment coefficient; v represents the current movement speed of the target object; V and m represent the volume data and mass data of the target object; α and k represent the first adjustment coefficient and the second adjustment coefficient; and the value range of the first adjustment coefficient is 0.13-0.28; the value range of the second adjustment coefficient is 1.17-1.32;

[0103] Extracting the body coverage radius of the target robot;

[0104] The basic coverage space range is obtained according to the body coverage radius of the target robot combined with the coverage space range adjustment coefficient; wherein the area corresponding to the basic coverage space range is obtained by the following formula:

[0105]

[0106] Where A represents the area corresponding to the basic coverage space range; S represents the coverage space range adjustment coefficient; R represents the body coverage radius of the target robot; P v Indicates the minimum speed change ratio of the current target robot; n indicates the unit time that the target robot has been running, and the unit time is 1s; v i represents the running speed of the target robot corresponding to the i-th unit time; v i-1 represents the running speed of the target robot corresponding to the i-1th unit time;

[0107] The running speed and running speed acceleration response time of the current target robot are extracted to adjust the basic coverage space range to obtain the adjusted coverage space range; wherein the area corresponding to the adjusted coverage space range is obtained by the following formula:

[0108]

[0109] Among them, A t represents the area corresponding to the adjusted coverage space range; A represents the area corresponding to the basic coverage space range; v represents the current movement speed of the target object; v m represents the running speed of the target robot; r represents the number of accelerations during the running process of the target robot; t irepresents the acceleration response time of the running speed corresponding to the i-th acceleration; t max represents the maximum value of the acceleration response time of the running speed corresponding to r accelerations; t m Indicates the running speed acceleration response time corresponding to the maximum acceleration of the target robot.

[0110] The beneficial effects of the above embodiments are that the scheme can dynamically calculate the coverage space range adjustment coefficient (S) by extracting the current movement speed, volume data and mass data of the target object. This coefficient can reflect the influence of the characteristics of the target object on the coverage space range of the target robot receiving operation, so that the robot can adjust its operation range according to the characteristics of different objects, thereby improving the adaptability and flexibility of the robot. By introducing the first adjustment coefficient (α) and the second adjustment coefficient (k), as well as a detailed calculation formula, the scheme can more accurately calculate the coverage space range adjustment coefficient, and then determine the basic coverage space range. This accuracy helps to ensure that the robot can accurately cover the target object when performing the object receiving task, thereby improving the success rate of task execution. The scheme takes into account factors such as the current movement speed of the target robot, the acceleration response time of the running speed, and the like, and adjusts the basic coverage space range in real time. This real-time performance enables the robot to quickly adjust its operation range according to the current motion state to adapt to the changing environmental conditions and task requirements. The scheme comprehensively considers the characteristics of the target object (such as volume, mass, movement speed) and the motion state of the target robot (such as running speed, acceleration response time, etc.), and determines the final coverage space range through the synergy of multiple parameters. This comprehensiveness makes the solution more comprehensive and reliable, capable of coping with complex and changing practical application scenarios.

[0111] At the same time, the adjustment coefficient α plays a role of adjustment and balance. It helps to appropriately enhance or weaken the influence of speed, volume and mass on the calculation results when calculating the adjustment coefficient C. This is because the importance of these factors may vary in different application scenarios. By adjusting the value of α, the model can be flexibly adjusted according to specific application requirements and environmental conditions to make it more in line with the actual situation. This flexibility is very necessary to adapt to changing operating environments and task requirements. While ensuring the safety of robot operation, it is also necessary to consider the efficiency of operation. By setting α appropriately, efficiency can be maximized while ensuring safety. k is a safety factor used to consider the influence of inertia, which is mainly used to add additional safety margins in calculations. This is especially important for dealing with emergencies and unpredictable dynamic behaviors of objects. In a dynamic and potentially changing environment, the movement of the target object may change due to various external factors. k helps the robot better cope with such dynamic changes by adding additional coverage space range. By considering the inertial motion distance, k helps ensure the reliability of the robot when approaching or operating the target object, reducing accidents or errors caused by inertia. The introduction of the two adjustment coefficients significantly increases the adaptability and flexibility of the calculation model, enabling it to better meet the needs of specific tasks and environments. By precisely adjusting the values ​​of these coefficients, it is possible to effectively balance safety and efficiency, enhance the accuracy and reliability of robot operations, and thus maximize performance in practical applications.

[0112] In summary, this technical solution improves the efficiency and accuracy of the target robot in performing object receiving tasks through the advantages of dynamic adaptation, precise calculation, real-time adjustment, comprehensive consideration and scalability.

[0113] See also Figure 2 As shown, an aerial dual-track intelligent positioning system based on visual recognition is provided in one embodiment of the present application. The aerial dual-track intelligent positioning system based on visual recognition includes:

[0114] The robot motion change prediction module is used to obtain the robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space;

[0115] A robot dynamic distribution determination module, used to generate a robot dynamic distribution map of the ground space based on the movement change trend;

[0116] A docking robot selection module is used to obtain the object transportation data within the aerial double-track network, and select a target robot that can dock with the target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data;

[0117] An object motion visual recognition module is used to perform visual recognition and tracking on the target object, and determine the motion state data of the target object in the aerial dual-track network;

[0118] A robot motion control module, used for controlling the target robot to maintain a corresponding relative relationship with the target object based on the motion state data;

[0119] The object landing adjustment module is used to perform visual positioning on the target robot after the target object moves to the destination inside the aerial double-track network, so as to adjust the motion state of the target object landing on the target robot.

[0120] The beneficial effects of the above embodiments are as follows: the aerial dual-track intelligent positioning system based on visual recognition acquires and analyzes the robot motion data of the ground space below the aerial dual-track network, predicts the motion change trends of all robots in the ground space, generates a robot dynamic distribution map, and performs a full-range dynamic prediction of the motion conditions of robots performing work tasks in the ground space, providing a reference for the subsequent selection of robots to dock with objects transported in the air; based on the object transportation data and the robot dynamic distribution map within the aerial dual-track network, selects a target robot that can dock with the target object transported in the air, and visually identifies and tracks the motion state data of the target object in the air, matches the target object with the target robot, and provides a guarantee for the process of docking the target object from the air to the ground; based on the motion state data, controls the target robot to maintain a relative relationship with the target object, visually locates the target robot, adjusts the target object to land to the motion state of the target robot, improves the flexibility and reliability of aerial dual-track transportation of objects, and meets the needs of calling objects at different locations.

[0121] In another embodiment, the robot motion change prediction module is used to obtain robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space, including:

[0122] Obtaining the movement speed setting data and movement direction setting data of each of the robots in the ground space below the aerial double-track network; wherein the movement speed setting data refers to the movement speed change data set for the robot within a period of time in the future; and the movement direction setting data refers to the movement direction change data set for the robot within a period of time in the future;

[0123] The motion speed setting data and the motion direction setting data are integrated and analyzed to predict the motion path change trend of all robots in the ground space within a future time interval; wherein the motion path change trend includes the position coordinate information of all characteristic time points of the robot in the ground space within the future time interval;

[0124] The robot dynamic distribution determination module is used to generate a robot dynamic distribution map of the ground space based on the movement change trend, including:

[0125] The motion path change trends of all robots are processed by spatial coordinate system transformation and time domain superposition integration to generate a dynamic distribution map of the robots in the ground space; wherein the dynamic distribution map of the robots includes a number of overall distribution sub-maps of the robots corresponding to all the characteristic time points within a certain period of time in the future.

[0126] The beneficial effect of the above embodiment is that an aerial double-track network and a robot group are arranged in the upper area and the ground area of ​​the production workshop, respectively, wherein the aerial double-track network includes a number of double-track tracks distributed continuously and / or crosswise in the upper area, and the corresponding double-track tracks can be used to transport objects such as parts to different locations in the upper area, and do not interfere with the robots moving in the ground area; the robot group includes a number of robots that can move freely in the ground area and perform tasks independently, each robot can independently perform corresponding operations, and can also dock objects transported in the aerial double-track network according to corresponding instructions (i.e., receive objects dropped from the aerial double-track network). All robots under the robot group are free to move in the ground area, so that each robot is located in a different position in the ground area at different time points. In order to accurately dispatch the appropriate robot to dock the object in a short time, it is necessary to first fully grasp the movement of all robots in the ground area, so that robots with a closer straight-line distance to the object or a faster movement speed in the ground area can be preferentially dispatched to dock the object. To this end, the motion speed setting data and motion direction setting data of all robots in the ground space below the aerial double-track network are obtained, and the motion speed change data and motion direction change data corresponding to the motion speed and motion direction set by each robot for the corresponding operation in the future period of time are comprehensively characterized. In addition, the motion speed setting data and motion direction setting data of each robot are integrated and analyzed to predict the motion path change trend of each robot in the ground space in the future period of time, and the position coordinates of all characteristic time points (such as several time points at fixed time intervals) of each robot in the future period of time are accurately calibrated in the ground space, so as to quantitatively characterize the motion position of each robot in the ground space at the time domain level. In addition, the motion path change trend of all robots is processed by a spatial coordinate system-transformation and time domain superposition integration, and a dynamic distribution map of the robots in the ground space is generated, so that the distribution position of all robots in the ground area at each characteristic time point in the future period of time can be globally characterized, providing a reliable basis for the subsequent selection of suitable robots to dock with objects transported by the aerial double-track network.

[0127] In another embodiment, the docking robot selection module is used to obtain object transportation data within the aerial double-track network, and select a target robot capable of docking with a target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data, including:

[0128] Obtaining object transportation occupancy time data of each of all double-track tracks under the aerial double-track network, and determining the idle time data of each of all double-track tracks based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the transportation target object in the aerial double-track network;

[0129] Comparing the robot dynamic distribution map with the transport track path in the time domain, determining the time overlap between all the robots in the ground space and the transport track path; based on the time overlap, selecting one of the robots as a target robot that can dock with the target object;

[0130] The object motion visual recognition module is used to perform visual recognition and tracking on the target object, and determine the motion state data of the target object in the aerial dual-track network, including:

[0131] The target object is visually identified and tracked to determine the movement speed data and movement position data of the target object in the transportation track path of the aerial double-track network.

[0132] The beneficial effects of the above embodiments are that the aerial double-track network includes a number of double-track tracks that are continuously and / or cross-distributed in the overhead area, and multiple objects are transported within the aerial double-track network in the same time period. In order to ensure that all objects can be effectively and continuously transported within the aerial double-track network, it is necessary to perform time domain analysis on the object transportation conditions of all double-track tracks, specifically obtain the object transportation occupancy time data of each double-track track under the aerial double-track network, and determine the idle time data of each double-track track based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the target object transported in the aerial double-track network. In this way, the transportation track path corresponding to the target object transported in the aerial double-track network can be planned based on the time interval when all double-track tracks are in an idle state, effectively avoiding transportation path conflicts between the target object and other objects during the transportation process within the aerial double-track network. The robot dynamic distribution map is also compared with the transport track path in the time domain to determine the time overlap between all the robots in the ground space and the transport track path. The time overlap refers to the time overlap between the time interval of each robot in the working state and the time interval corresponding to the transport track path transporting the target object. The robot with the smallest time overlap is used as the target robot that can dock with the target object. The robot with the smallest time overlap has a low probability of being in the working state during the time interval corresponding to the transport track path transporting the target object, which facilitates the subsequent scheduling of the robot to dock with the target object at any time. The target object is also visually identified and tracked to determine the movement speed data and movement position data of the target object in the transport track path of the aerial double-track network, so that the transportation status of the target object in the aerial double-track network can be characterized in multiple dimensions, providing a basis for the subsequent realization of the motion linkage control of the target robot and the target object.

[0133] In another embodiment, the robot motion control module is used to control the target robot to maintain a corresponding relative relationship with the target object based on the motion state data, including:

[0134] Based on the movement speed data and movement position data of the target object in the transportation track of the aerial double-track network, the movement speed of the target robot is controlled so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transportation task;

[0135] The object landing adjustment module is used to perform visual positioning on the target robot after the target object moves to the destination inside the aerial double-track network, so as to adjust the motion state of the target object landing on the target robot, including:

[0136] When the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform object receiving operations; based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network is adjusted to perform the landing movement of the target object.

[0137] The beneficial effect of the above embodiment is to control the movement speed of the target robot with reference to the movement speed data and movement position data of the target object in the transport track of the aerial double-track network, so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transport task, such as maintaining the distance between the target robot and the target object in the corresponding distance range during the execution of the original transport task, ensuring that the target robot can change its own working state in a short time and dock with the target object. Also, when the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform the object receiving operation, and the coverage space range can be but is not limited to the operating space range formed by the mechanical arm of the target robot itself. And based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network to land the target object will be adjusted to ensure that the target object can be accurately aligned with the working space range formed by the robotic arm of the target robot itself during the process of landing from the aerial double-track network to the ground area, so as to avoid the target object from falling due to failure to be accurately and smoothly received by the target robot during the landing process.

[0138] In another embodiment, determining the coverage space range in which the target robot can perform an object receiving operation includes:

[0139] Extracting the current speed of the target object;

[0140] extracting volume data and mass data of the target object;

[0141] Obtaining a coverage space range adjustment coefficient according to the current movement speed of the target object combined with the volume data and mass data of the target object;

[0142] The coverage space range adjustment coefficient is obtained by the following formula:

[0143]

[0144] Wherein, S represents the coverage space range adjustment coefficient; v represents the current movement speed of the target object; V and m represent the volume data and mass data of the target object; α and k represent the first adjustment coefficient and the second adjustment coefficient; and the value range of the first adjustment coefficient is 0.13-0.28; the value range of the second adjustment coefficient is 1.17-1.32;

[0145] Extracting the body coverage radius of the target robot;

[0146] The basic coverage space range is obtained according to the body coverage radius of the target robot combined with the coverage space range adjustment coefficient; wherein the area corresponding to the basic coverage space range is obtained by the following formula:

[0147]

[0148] Where A represents the area corresponding to the basic coverage space range; S represents the coverage space range adjustment coefficient; R represents the body coverage radius of the target robot; P v Indicates the minimum speed change ratio of the current target robot; n indicates the unit time that the target robot has been running, and the unit time is 1s; v i represents the running speed of the target robot corresponding to the i-th unit time; v i-1 represents the running speed of the target robot corresponding to the i-1th unit time;

[0149] The running speed and running speed acceleration response time of the current target robot are extracted to adjust the basic coverage space range to obtain the adjusted coverage space range; wherein the area corresponding to the adjusted coverage space range is obtained by the following formula:

[0150]

[0151] Among them, A t represents the area corresponding to the adjusted coverage space range; A represents the area corresponding to the basic coverage space range; v represents the current movement speed of the target object; v m represents the running speed of the target robot; r represents the number of accelerations during the running process of the target robot; t i represents the acceleration response time of the running speed corresponding to the i-th acceleration; t max represents the maximum value of the acceleration response time of the running speed corresponding to r accelerations; t m Indicates the running speed acceleration response time corresponding to the maximum acceleration of the target robot.

[0152] The beneficial effects of the above embodiments are that the scheme can dynamically calculate the coverage space range adjustment coefficient (S) by extracting the current movement speed, volume data and mass data of the target object. This coefficient can reflect the influence of the characteristics of the target object on the coverage space range of the target robot receiving operation, so that the robot can adjust its operation range according to the characteristics of different objects, thereby improving the adaptability and flexibility of the robot. By introducing the first adjustment coefficient (α) and the second adjustment coefficient (k), as well as a detailed calculation formula, the scheme can more accurately calculate the coverage space range adjustment coefficient, and then determine the basic coverage space range. This accuracy helps to ensure that the robot can accurately cover the target object when performing the object receiving task, thereby improving the success rate of task execution. The scheme takes into account factors such as the current movement speed of the target robot, the acceleration response time of the running speed, and the like, and adjusts the basic coverage space range in real time. This real-time performance enables the robot to quickly adjust its operation range according to the current motion state to adapt to the changing environmental conditions and task requirements. The scheme comprehensively considers the characteristics of the target object (such as volume, mass, movement speed) and the motion state of the target robot (such as running speed, acceleration response time, etc.), and determines the final coverage space range through the synergy of multiple parameters. This comprehensiveness makes the solution more comprehensive and reliable, capable of coping with complex and changing practical application scenarios.

[0153] At the same time, the adjustment coefficient α plays a role of adjustment and balance. It helps to appropriately enhance or weaken the influence of speed, volume and mass on the calculation results when calculating the adjustment coefficient C. This is because the importance of these factors may vary in different application scenarios. By adjusting the value of α, the model can be flexibly adjusted according to specific application requirements and environmental conditions to make it more in line with the actual situation. This flexibility is very necessary to adapt to changing operating environments and task requirements. While ensuring the safety of robot operation, it is also necessary to consider the efficiency of operation. By setting α appropriately, efficiency can be maximized while ensuring safety. k is a safety factor used to consider the influence of inertia, which is mainly used to add additional safety margins in calculations. This is especially important for dealing with emergencies and unpredictable dynamic behaviors of objects. In a dynamic and potentially changing environment, the movement of the target object may change due to various external factors. k helps the robot better cope with such dynamic changes by adding additional coverage space range. By considering the inertial motion distance, k helps ensure the reliability of the robot when approaching or operating the target object, reducing accidents or errors caused by inertia. The introduction of the two adjustment coefficients significantly increases the adaptability and flexibility of the calculation model, enabling it to better meet the needs of specific tasks and environments. By precisely adjusting the values ​​of these coefficients, it is possible to effectively balance safety and efficiency, enhance the accuracy and reliability of robot operations, and thus maximize performance in practical applications.

[0154] In summary, this technical solution improves the efficiency and accuracy of the target robot in performing object receiving tasks through the advantages of dynamic adaptation, precise calculation, real-time adjustment, comprehensive consideration and scalability.

[0155] In general, the aerial dual-track intelligent positioning method and system based on visual recognition acquires and analyzes the robot motion data in the ground space below the aerial dual-track network, predicts the motion change trends of all robots in the ground space, generates a robot dynamic distribution map, and makes a full-range dynamic prediction of the motion of robots performing work tasks in the ground space, providing a reference for the subsequent selection of robots to dock with objects transported in the air; based on the object transportation data and the robot dynamic distribution map within the aerial dual-track network, selects a target robot that can dock with the target object transported in the air, and visually identifies and tracks the motion state data of the target object in the air, matches the target object with the target robot, and provides a guarantee for the process of docking the target object from the air to the ground; based on the motion state data, controls the target robot to maintain a relative relationship with the target object, visually locates the target robot, adjusts the target object to land on the target robot, improves the flexibility and reliability of aerial dual-track transportation of objects, and meets the needs of calling objects at different locations.

[0156] The above is only a specific implementation of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the protection scope of the present invention.

Claims

1. The aerial dual-track intelligent positioning method based on visual recognition is characterized by: include: Acquire robot motion data of the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space; generate a robot dynamic distribution map of the ground space based on the motion change trends; acquire object transportation data within the aerial double-track network, and select a target robot that can dock with a target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data; Perform visual recognition and tracking on the target object to determine the motion state data of the target object in the aerial dual-track network; Based on the motion state data, the target robot is controlled to maintain a corresponding relative relationship with the target object; when the target object moves to the destination inside the aerial double-track network, the target robot is visually positioned to adjust the motion state of the target object to land on the target robot.

2. The aerial dual-track intelligent positioning method based on visual recognition as claimed in claim 1, characterized in that: Acquire the robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space; Based on the movement change trend, a dynamic distribution map of the robots in the ground space is generated, including: Acquire the movement speed setting data and movement direction setting data of all robots in the ground space below the aerial double-track network; wherein the movement speed setting data refers to the movement speed change data set for the robot within a period of time in the future; the movement direction setting data refers to the movement direction change data set for the robot within a period of time in the future; The motion speed setting data and the motion direction setting data are integrated and analyzed to predict the motion path change trend of all robots in the ground space within a future time interval; wherein the motion path change trend includes the position coordinate information of all characteristic time points of the robots in the ground space within a future time interval; The motion path change trends of all robots are processed by spatial coordinate system transformation and time domain superposition integration to generate a dynamic distribution map of the robots in the ground space; wherein the dynamic distribution map of the robots includes a number of overall distribution sub-maps of the robots corresponding to all the characteristic time points within a future period of time.

3. The aerial dual-track intelligent positioning method based on visual recognition as claimed in claim 1, characterized in that: Acquire object transportation data within the aerial double-track network, and select a target robot capable of docking with a target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data; Performing visual recognition and tracking on the target object to determine the motion state data of the target object in the aerial dual-track network includes: Obtaining object transportation occupancy time data of each of all double-track tracks under the aerial double-track network, and determining the idle time data of each of all double-track tracks based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the transportation of the target object in the aerial double-track network; Comparing the robot dynamic distribution map with the transport track path in the time domain, determining the time overlap between all the robots in the ground space and the transport track path; based on the time overlap, selecting one of the robots as a target robot that can dock with the target object; Visual recognition and tracking are performed on the target object to determine the movement speed data and movement position data of the target object in the transportation track path of the aerial double-track network.

4. The aerial dual-track intelligent positioning method based on visual recognition as claimed in claim 1, characterized in that: Based on the motion state data, the target robot is controlled to maintain a corresponding relative relationship with the target object; when the target object moves to a destination inside the aerial double-track network, the target robot is visually positioned to adjust the motion state of the target object to land on the target robot, including: Based on the movement speed data and movement position data of the target object in the transportation track of the aerial double-track network, the movement speed of the target robot is controlled so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transportation task; When the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform object receiving operations; based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network is adjusted to land the target object.

5. The aerial dual-track intelligent positioning method based on visual recognition as claimed in claim 1, characterized in that: Determining the coverage space range in which the target robot can perform an object receiving operation includes: Extracting the current speed of the target object; extracting volume data and mass data of the target object; Obtaining a coverage space range adjustment coefficient according to the current movement speed of the target object combined with the volume data and mass data of the target object; The coverage space range adjustment coefficient is obtained by the following formula: Wherein, S represents the coverage space range adjustment coefficient; v represents the current movement speed of the target object; V and m represent the volume data and mass data of the target object; α and k represent the first adjustment coefficient and the second adjustment coefficient; and the value range of the first adjustment coefficient is 0.13-0.28; the value range of the second adjustment coefficient is 1.17-1.32; Extracting the body coverage radius of the target robot; The basic coverage space range is obtained according to the body coverage radius of the target robot combined with the coverage space range adjustment coefficient; wherein the area corresponding to the basic coverage space range is obtained by the following formula: Where A represents the area corresponding to the basic coverage space range; S represents the coverage space range adjustment coefficient; R represents the body coverage radius of the target robot; P v Indicates the minimum speed change ratio of the current target robot; n indicates the unit time that the target robot has been running, and the unit time is 1s; v i represents the running speed of the target robot corresponding to the i-th unit time; v i-1 represents the running speed of the target robot corresponding to the i-1th unit time; The running speed and running speed acceleration response time of the current target robot are extracted to adjust the basic coverage space range to obtain the adjusted coverage space range; wherein the area corresponding to the adjusted coverage space range is obtained by the following formula: Among them, A t represents the area corresponding to the adjusted coverage space range; A represents the area corresponding to the basic coverage space range; v represents the current movement speed of the target object; v m represents the running speed of the target robot; r represents the number of accelerations during the running process of the target robot; t i represents the acceleration response time of the running speed corresponding to the i-th acceleration; t max It represents the maximum value of the acceleration response time of the running speed corresponding to the r accelerations; t m Indicates the running speed acceleration response time corresponding to the maximum acceleration of the target robot.

6. The aerial dual-track intelligent positioning system based on visual recognition is characterized by: include: A robot motion change prediction module is used to obtain the robot motion data in the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space; A robot dynamic distribution determination module, used to generate a robot dynamic distribution map of the ground space based on the movement change trend; A docking robot selection module is used to obtain the object transportation data within the aerial double-track network, and select a target robot that can dock with the target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data; An object motion visual recognition module is used to perform visual recognition and tracking on the target object to determine the motion state data of the target object in the aerial dual-track network; A robot motion control module, used for controlling the target robot to maintain a corresponding relative relationship with the target object based on the motion state data; The object landing adjustment module is used to visually locate the target robot after the target object moves to the destination inside the aerial double-track network, so as to adjust the motion state of the target object landing to the target robot.

7. The aerial dual-track intelligent positioning system based on visual recognition as claimed in claim 6, characterized in that: The robot motion change prediction module is used to obtain the robot motion data of the ground space below the aerial double-track network, analyze the motion data, and predict the motion change trends of all robots in the ground space, including: Acquire the movement speed setting data and movement direction setting data of all robots in the ground space below the aerial double-track network; wherein the movement speed setting data refers to the movement speed change data set for the robot within a period of time in the future; the movement direction setting data refers to the movement direction change data set for the robot within a period of time in the future; The motion speed setting data and the motion direction setting data are integrated and analyzed to predict the motion path change trend of all robots in the ground space within a future time interval; wherein the motion path change trend includes the position coordinate information of all characteristic time points of the robots in the ground space within a future time interval; The robot dynamic distribution determination module is used to generate a robot dynamic distribution map of the ground space based on the movement change trend, including: The motion path change trends of all robots are processed by spatial coordinate system transformation and time domain superposition integration to generate a dynamic distribution map of the robots in the ground space; wherein the dynamic distribution map of the robots includes a number of overall distribution sub-maps of the robots corresponding to all the characteristic time points within a future period of time.

8. The aerial dual-track intelligent positioning system based on visual recognition as claimed in claim 6, characterized in that: The docking robot selection module is used to obtain the object transportation data within the aerial double-track network, and select a target robot that can dock with the target object transported within the aerial double-track network based on the robot dynamic distribution map and the object transportation data, including: Obtaining object transportation occupancy time data of each of all double-track tracks under the aerial double-track network, and determining the idle time data of each of all double-track tracks based on the object transportation occupancy time data, so as to determine the transportation track path corresponding to the transportation of the target object in the aerial double-track network; Comparing the robot dynamic distribution map with the transport track path in the time domain, determining the time overlap between all the robots in the ground space and the transport track path; based on the time overlap, selecting one of the robots as a target robot that can dock with the target object; The object motion visual recognition module is used to perform visual recognition and tracking on the target object, and determine the motion state data of the target object in the aerial dual-track network, including: Visual recognition and tracking are performed on the target object to determine the movement speed data and movement position data of the target object in the transportation track path of the aerial double-track network.

9. The aerial dual-track intelligent positioning system based on visual recognition as claimed in claim 6, characterized in that: The robot motion control module is used to control the target robot to maintain a corresponding relative relationship with the target object based on the motion state data, including: Based on the movement speed data and movement position data of the target object in the transportation track of the aerial double-track network, the movement speed of the target robot is controlled so that the target robot maintains a corresponding relative position relationship with the target object during the execution of the original transportation task; The object landing adjustment module is used to perform visual positioning on the target robot after the target object moves to the destination inside the aerial double-track network, so as to adjust the motion state of the target object landing to the target robot, including: When the target object moves to the destination inside the aerial double-track network, the target robot is identified and located to determine the coverage space range in which the target robot can perform object receiving operations; based on the coverage space range, the movement direction of the robotic arm that grabs the target object inside the aerial double-track network is adjusted to land the target object.

10. The aerial dual-track intelligent positioning system based on visual recognition as claimed in claim 6, characterized in that: Determining the coverage space range in which the target robot can perform an object receiving operation includes: Extracting the current speed of the target object; extracting volume data and mass data of the target object; Obtaining a coverage space range adjustment coefficient according to the current movement speed of the target object combined with the volume data and mass data of the target object; The coverage space range adjustment coefficient is obtained by the following formula: Wherein, S represents the coverage space range adjustment coefficient; v represents the current movement speed of the target object; V and m represent the volume data and mass data of the target object; α and k represent the first adjustment coefficient and the second adjustment coefficient; and the value range of the first adjustment coefficient is 0.13-0.28; the value range of the second adjustment coefficient is 1.17-1.32; Extracting the body coverage radius of the target robot; The basic coverage space range is obtained according to the body coverage radius of the target robot combined with the coverage space range adjustment coefficient; wherein the area corresponding to the basic coverage space range is obtained by the following formula: Where A represents the area corresponding to the basic coverage space range; S represents the coverage space range adjustment coefficient; R represents the body coverage radius of the target robot; P v Indicates the minimum speed change ratio of the current target robot; n indicates the unit time that the target robot has been running, and the unit time is 1s; v i represents the running speed of the target robot corresponding to the i-th unit time; v i-1 represents the running speed of the target robot corresponding to the i-1th unit time; The running speed and running speed acceleration response time of the current target robot are extracted to adjust the basic coverage space range to obtain the adjusted coverage space range; wherein the area corresponding to the adjusted coverage space range is obtained by the following formula: Among them, A t represents the area corresponding to the adjusted coverage space range; A represents the area corresponding to the basic coverage space range; v represents the current movement speed of the target object; v m represents the running speed of the target robot; r represents the number of accelerations during the running process of the target robot; t i represents the acceleration response time of the running speed corresponding to the i-th acceleration; t max It represents the maximum value of the acceleration response time of the running speed corresponding to the r accelerations; t m Indicates the running speed acceleration response time corresponding to the maximum acceleration of the target robot.

Images