WO2022179179A1 - 一种面向异构特性大型装备多智能体协同自主转运系统 - Google Patents
一种面向异构特性大型装备多智能体协同自主转运系统 Download PDFInfo
- Publication number
- WO2022179179A1 WO2022179179A1 PCT/CN2021/129478 CN2021129478W WO2022179179A1 WO 2022179179 A1 WO2022179179 A1 WO 2022179179A1 CN 2021129478 W CN2021129478 W CN 2021129478W WO 2022179179 A1 WO2022179179 A1 WO 2022179179A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- agent
- pose
- agents
- splicing
- navigation
- Prior art date
Links
- 238000012546 transfer Methods 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 claims abstract description 30
- 230000008569 process Effects 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 11
- 239000003795 chemical substances by application Substances 0.000 claims description 249
- 230000033001 locomotion Effects 0.000 claims description 90
- 238000005259 measurement Methods 0.000 claims description 20
- 230000008878 coupling Effects 0.000 claims description 14
- 238000010168 coupling process Methods 0.000 claims description 14
- 238000005859 coupling reaction Methods 0.000 claims description 14
- 230000000007 visual effect Effects 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000003550 marker Substances 0.000 claims description 6
- 230000007423 decrease Effects 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 abstract description 2
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 230000006978 adaptation Effects 0.000 abstract 1
- 238000003032 molecular docking Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000005755 formation reaction Methods 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G63/00—Transferring or trans-shipping at storage areas, railway yards or harbours or in opening mining cuts; Marshalling yard installations
- B65G63/002—Transferring or trans-shipping at storage areas, railway yards or harbours or in opening mining cuts; Marshalling yard installations for articles
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Definitions
- the invention relates to a multi-agent cooperative autonomous transport system for large-scale equipment with heterogeneous characteristics, which has the ability of single-agent independent operation and multi-agent cooperative operation, and has indoor and outdoor unmanned autonomous navigation and positioning capabilities.
- the technical problem solved by the present invention is: to overcome the deficiencies of the prior art, to provide a multi-agent cooperative autonomous transport system for large-scale equipment with heterogeneous characteristics, and to realize the efficient, flexible and generalized transport of high-end equipment, so as to adapt to the high-end equipment.
- the equipment exceeds the transfer requirements of the existing single transfer equipment size and carrying capacity, improves the flexibility and adaptability of the equipment, reduces the manual labor in the previous product transfer process, and shortens the product transfer time.
- a multi-agent cooperative autonomous transport system for large-scale equipment with heterogeneous characteristics comprising: a main controller, an instruction receiving and processing module, a navigation unit, a combined splicing path planning module, a wireless control module and multiple agents; instruction receiving and processing The module receives the control command input from the outside, the control command includes the movement command and the splicing mode of the transport system, and sends the movement command and the splicing mode of the transport system to the main controller; the main controller sends the movement command to the navigation unit, The navigation unit performs navigation calculation according to the measured value of the current position of the transfer system and the received motion commands, generates the overall yaw angle, speed and rotational angular speed of the transfer system, and feeds it back to the main controller;
- the main controller sends the splicing mode and the motion parameters fed back by the navigation unit to the combined splicing path planning module.
- the control module sends it to the corresponding agents; each agent performs trajectory planning according to the received theoretical motion parameters, combined with the current pose information of the agent and the pose information of other agents obtained through the wireless control module.
- the closed-loop compensation of generates actual motion parameters, and the current agent operates according to the actual motion parameters, so as to realize the cooperative autonomous transport of multiple agents.
- the external input control command received by the command receiving and processing module comes from a manual control command sent by the handset or an automatic control command from an external dispatching system.
- the navigation unit includes: a navigation control module, a navigation sensor receiving module, a visual navigation sensor, a laser navigation sensor and an iGPS navigation sensor;
- the visual navigation sensor collects the local relative position information between the marker and the visual navigation sensor;
- the laser navigation sensor and the iGPS navigation sensor collect the coordinate information and attitude angle of the sensor itself in the global coordinate system;
- the navigation sensor receiving module receives the outputs of the three sensors The result is provided to the navigation control module;
- the navigation control module When the required positioning accuracy is between 1mm and 5mm, the navigation control module performs coordinate transformation on the local relative position information to obtain the relative coordinates and attitude angle of the overall transfer center in the local coordinate system, and according to the obtained relative coordinates, Attitude angle and target path deviation, generate the overall yaw angle, speed and rotation angular speed of the transfer system, and feed it back to the main controller;
- the local coordinate system refers to the coordinate system established with the center of the marker as the origin;
- the navigation control module When the required positioning accuracy is greater than or equal to 5mm, the navigation control module performs coordinate transformation on the coordinate information and attitude angle of the sensor itself in the global coordinate system collected by the laser navigation sensor, and obtains the coordinates and attitude angle of the overall transfer center in the global coordinate system. ; And according to the obtained coordinates, attitude angle and target path deviation, generate the overall yaw angle, speed and rotational angular velocity of the transfer system, and feed it back to the main controller; the global coordinate system refers to: take the vertex of the agent transfer site as the origin establish a coordinate system;
- the navigation control module When the required positioning accuracy is less than or equal to 1mm, the navigation control module performs coordinate transformation on the coordinate information and attitude angle of the sensor itself in the global coordinate system collected by the iGPS navigation sensor, and obtains the coordinates and attitude angle of the overall transfer center in the global coordinate system. ; And according to the obtained coordinates, attitude angle and target path deviation, the overall yaw angle, speed and rotation angular speed of the transfer system are generated and fed back to the main controller.
- the movement command of the transfer system in the external input control command specifically includes: target coordinate X, target coordinate Y, target angle ⁇ , running command and running mode;
- Target coordinate X, target coordinate Y and target angle ⁇ are the positions in the global coordinate system.
- the running commands include stop, forward, backward, right lateral, left lateral, 90° counterclockwise and 90° clockwise; running modes include emergency Stop mode and normal mode; emergency stop mode refers to the immediate stop of the entire transfer system; normal mode refers to the mode in which the transfer system operates normally according to the instructions.
- the splicing mode of the transfer system in the external input control command specifically includes: L-shaped splicing, product type splicing, two-car splicing and four-car splicing;
- L-shaped splicing refers to the form in which three agents are arranged in an L shape
- product-type splicing refers to the form in which three agents are arranged according to their shape
- double-car splicing means that two agents are arranged in the form of horizontal or vertical side-by-side
- Four-car stitching refers to the form in which four agents are arranged in a rectangle.
- the combined splicing path planning module decomposes the overall motion parameters of the transport system into theoretical motion parameters of each agent according to the splicing mode, which specifically includes the following steps:
- the pose between the overall transfer center points O is: in is the distance between the i-th agent center O i relative to the overall transit center point O, is the angle between the i-th agent center O i relative to the overall transfer center point O;
- the X-direction, Y-direction and angle increment ( ⁇ s xi , ⁇ s yi , ⁇ zi ) of the i-th agent center in unit time ⁇ t are:
- agent is a transfer platform based on the Mecanum wheel, and each agent is provided with a motion trajectory planning and closed-loop control module, a pose measurement sensor interface module, and a pose measurement sensor combination;
- the pose measurement sensor interface module receives the current pose information of the agent collected by the combination of measurement sensors, and provides it to the motion trajectory planning and closed-loop control module.
- the current pose information of the agent and the pose information of other agents obtained through the wireless control module are used for closed-loop compensation of trajectory planning to generate actual motion data.
- the combination of pose measurement sensors includes a laser ranging sensor, a two-dimensional laser radar, and a two-dimensional photoelectric sensor PSD; the laser ranging sensor is used to measure the distance between two adjacent agents, and the two-dimensional laser radar is used to measure The relative attitude and angle between two adjacent agents, the two-dimensional photoelectric sensor PSD is used to measure the attitude deviation between two adjacent agents.
- the motion trajectory planning and closed-loop control module performs closed-loop compensation of trajectory planning to generate actual motion parameters, specifically:
- the most front-end agent in the multi-agent is used as the master agent, and the others are used as slave agents;
- the master agent moves according to the theoretical motion parameters of the agent, and other slave agents perform motion compensation according to the pose deviation from the master agent;
- Each slave agent selects a reference point on the rear face of the master agent; and determines the pose between each slave agent and the selected reference point of the slave agent as the initial pose;
- each slave agent acquires the pose between the slave agent and the reference point selected by the slave agent in real time as the real-time pose;
- S5 perform pose deviation calculation according to the initial pose and the real-time pose, and determine the pose deviation between each slave agent and the master agent;
- the rotational speed of each wheel of each slave agent is determined by using the compensation motion parameters described in S8.
- the present invention integrates multiple navigation technologies such as laser navigation technology, iGPS high-precision spatial positioning technology, and visual measurement, and realizes continuous navigation and high-precision spatial positioning capabilities for the global scene in the large space of the intelligent workshop;
- the present invention realizes the ability of the transport equipment to be compatible with the transport of large-scale heterogeneous products and conventional products through the coordinated transport of multi-agent heterogeneous marshalling, and solves the needs of product diversity and heterogeneity in the transport process of high-end equipment; and improves the flexibility of equipment. and adaptability, reduce the manual labor in the past product transfer and docking process, shorten the product transfer time, and realize the efficient application of intelligent equipment collaborative operation in the precise transfer docking and assembly and manufacturing links.
- the present invention adopts multi-agent multi-agent motion situation awareness and real-time online motion trajectory compensation.
- the agent performs real-time online analysis according to its own state and trajectory expectation, performs online trajectory compensation, and monitors the running fault state in real time for overall protection. measures to achieve autonomous health management and re-planning capabilities under fault conditions.
- the present invention realizes the innovative application of a flexible and efficient transfer docking process with diversified adaptive execution equipment and transfer modes, and provides better solutions for the shipping docking method of large-scale heavy-duty products.
- the cooperative operation mode based on multi-agent has the characteristics of high transfer accuracy, self-adaptive combination, and convenient and fast operation.
- the relative pose is periodically measured by the pose measurement system based on laser scanning radar contour recognition, and the deviation from the initial set pose is calculated; Fast wireless interaction of compensation parameters and state parameter data; through the mutual constraint analysis of the real-time pose deviation of each axis of the multi-agent, and establishing the corresponding pose compensation adjustment control strategy according to the analysis results, to realize the multi-agent cooperative transport online pose compensation adjustment.
- the present invention ensures the real-time relative pose control accuracy of the multi-agent cooperative transport process. Combined with the multi-agent autonomous heterogeneous formation collaborative control technology, it can adapt to the flexible transfer and docking of diverse and heterogeneous high-end equipment products, and realize the high-efficiency and generalized transfer and docking operation of multi-agent coordination. By adopting the inventive method, the problems of high-efficiency transfer, docking operation and high-precision positioning of large-scale heterogeneous high-end equipment in a narrow space are solved.
- Figure 1 is a schematic diagram of the main combination and splicing method of multi-agents of the present invention, wherein (a) is an L-shaped three-car splicing, (b) is a three-car splicing of a product type, (c) is a two-car splicing, and (d) is a four-car splicing splicing;
- Fig. 2 is the rotation center selection schematic diagram corresponding to four kinds of splicing modes of the present invention and Fig. 1;
- FIG. 3 is a schematic diagram of the system architecture of the present invention.
- Fig. 4 is the overall data flow diagram of the system of the present invention.
- FIG. 5 is a schematic diagram of the layout of the multi-agent splicing of the present invention.
- FIG. 6 is a schematic diagram of relative pose measurement during the multi-agent cooperative transport process of the present invention.
- a continuous navigation and high-precision spatial positioning method for the precise distribution link in the global transshipment scenario is constructed, which not only meets the cross-plant and cross-region joint transshipment of products, but also satisfies the high-precision positioning at key workstations.
- Multi-sensor information fusion technology Enhance system perception ability, enhance data reliability, improve accuracy, expand system time and space coverage, increase system real-time and information utilization, etc.
- the multi-agent cooperative autonomous transport system for large-scale equipment with heterogeneous characteristics involved in the present invention not only has the capability of single-vehicle operation and multi-body cooperative operation, but also has the capability of indoor and outdoor unmanned autonomous navigation and positioning.
- a multi-agent cooperative autonomous transport system for large-scale equipment with heterogeneous characteristics includes: a main controller, an instruction receiving and processing module, a navigation unit, a combined splicing path planning module, Wireless control module and multiple agents;
- the command receiving and processing module receives the control command input from the outside, the control command includes the movement command and the splicing mode of the transport system, and sends the movement command and the splicing mode of the transport system to the main controller; the main controller sends the movement command
- the navigation unit performs navigation calculation according to the measured value of the current position of the transfer system and the received motion command, generates the overall yaw angle, speed and rotational angular speed of the transfer system, and feeds it back to the main controller;
- the main controller sends the splicing mode and the motion parameters fed back by the navigation unit to the combined splicing path planning module.
- the control module sends it to the corresponding agents; each agent performs trajectory planning according to the received theoretical motion parameters, combined with the current pose information of the agent and the pose information of other agents obtained through the wireless control module.
- the closed-loop compensation of generates actual motion parameters, and the current agent operates according to the actual motion parameters, so as to realize the cooperative autonomous transport of multiple agents.
- the external input control command received by the command receiving and processing module comes from the manual control command sent by the hand-held device or the automatic control command from the external dispatching system, which specifically includes the following steps:
- the navigation unit includes: a navigation control module, a navigation sensor receiving module, a visual navigation sensor, a laser navigation sensor and an iGPS navigation sensor;
- the visual navigation sensor collects the local relative position information between the marker and the visual navigation sensor;
- the laser navigation sensor and the iGPS navigation sensor collect the coordinate information and attitude angle of the sensor itself in the global coordinate system;
- the navigation sensor receiving module receives the outputs of the three sensors The result is provided to the navigation control module;
- the navigation control module When the required positioning accuracy is between 1mm and 5mm, the navigation control module performs coordinate transformation on the local relative position information to obtain the relative coordinates and attitude angle of the overall transfer center in the local coordinate system, and according to the obtained relative coordinates, Attitude angle and target path deviation, generate the overall yaw angle, speed and rotation angular speed of the transfer system, and feed it back to the main controller;
- the local coordinate system refers to the coordinate system established with the center of the marker as the origin;
- the navigation control module When the required positioning accuracy is greater than or equal to 5mm, the navigation control module performs coordinate transformation on the coordinate information and attitude angle of the sensor itself in the global coordinate system collected by the laser navigation sensor, and obtains the coordinates and attitude angle of the overall transfer center in the global coordinate system. ; And according to the obtained coordinates, attitude angle and target path deviation, generate the overall yaw angle, speed and rotational angular velocity ( ⁇ , ⁇ , ⁇ z ) of the transfer system, and feed them back to the main controller; the global coordinate system refers to: The vertex of the agent's transfer site is the origin to establish a coordinate system;
- the navigation control module When the required positioning accuracy is less than or equal to 1mm, the navigation control module performs coordinate transformation on the coordinate information and attitude angle of the sensor itself in the global coordinate system collected by the iGPS navigation sensor, and obtains the coordinates and attitude angle of the overall transfer center in the global coordinate system. ; And according to the obtained coordinates, attitude angle and target path deviation, generate the overall yaw angle, speed and rotational angular velocity ( ⁇ , ⁇ , ⁇ z ) of the transfer system, and feed them back to the main controller.
- the movement command of the transfer system in the external input control command specifically including: target coordinate X, target coordinate Y, target angle ⁇ , running command and running mode;
- Target coordinate X, target coordinate Y and target angle ⁇ are the positions in the global coordinate system.
- the running commands include stop, forward, backward, right lateral, left lateral, 90° counterclockwise and 90° clockwise; running modes include emergency Stop mode and normal mode; emergency stop mode refers to the immediate stop of the entire transfer system; normal mode refers to the mode in which the transfer system operates normally according to the instructions.
- the splicing mode of the transfer system in the external input control command specifically includes: L-shaped splicing, pin-shaped splicing, two-car splicing and four-car splicing;
- L-shaped splicing refers to the form in which three agents are arranged in an L-shape
- Pin-shaped splicing refers to the form in which three agents are arranged in accordance with the shape
- Double-car splicing refers to the arrangement of two agents in the form of horizontal or vertical side by side.
- Four-car stitching refers to the form in which four agents are arranged in a rectangle.
- the combined splicing path planning module decomposes the overall motion parameters of the transport system into the theoretical motion parameters of each agent according to the splicing mode, which specifically includes the following steps:
- the pose between the overall transfer center points O is: in is the distance between the i-th agent center O i relative to the overall transit center point O, is the angle between the i-th agent center O i relative to the overall transfer center point O;
- the X- and Y-direction position increments and angle increments ( ⁇ s xi , ⁇ s yi , ⁇ zi ) of the center of the i-th agent in unit time ⁇ t are:
- the intelligent body of the present invention is a transfer platform based on the Mecanum wheel, and each intelligent body is provided with a motion trajectory planning and closed-loop control module, a position and attitude measurement sensor interface module, and a combination of position and attitude measurement sensors;
- the pose measurement sensor interface module adopts RS422, network port and RS232 to receive the current pose information of the agent collected by the combination of measurement sensors, and provide it to the motion trajectory planning and closed-loop control module.
- the theoretical motion parameters of the agent itself combined with the current pose information of the agent and the pose information of other agents obtained through the wireless control module, perform closed-loop compensation for trajectory planning to generate actual motion data.
- the combination of pose measurement sensors includes a laser ranging sensor, a two-dimensional laser radar, and a two-dimensional photoelectric sensor PSD; the laser ranging sensor is used to measure the distance between two adjacent agents, and the laser two-dimensional radar is used to measure the distance between two adjacent agents. It is used to measure the relative attitude and angle between two adjacent agents, and the two-dimensional photoelectric sensor PSD is used to measure the attitude deviation between two adjacent agents.
- the multi-agent high-speed synchronous wireless control module adopts a wireless communication link based on the WIFI network topology, with the access capability of up to 255 agents, and the response frequency can reach 100Hz.
- a high-speed 800MHz main frequency 32-bit processor is used to design the time synchronization algorithm, and the synchronization accuracy can reach the microsecond level.
- the TDMA time slot scheduling mechanism allocates channel resources for each node of the wireless network in advance, and each node intelligently sends data on the allocated time slot, which effectively guarantees the real-time and reliability of communication, especially when the time synchronization accuracy is required to be high. Further improve real-time and reliability.
- Each single agent communicates with each other through a WIFI wireless link.
- the motion trajectory planning and closed-loop control module performs closed-loop compensation of trajectory planning to generate actual motion parameters, specifically:
- the most front-end agent in the multi-agent is used as the master agent, and the others are used as slave agents;
- the master agent moves according to the theoretical motion parameters of the agent, and other slave agents perform motion compensation according to the pose deviation from the master agent;
- Each slave agent selects a reference point on the rear face of the master agent; and determines the pose between each slave agent and the selected reference point of the slave agent as the initial pose;
- each slave agent acquires the pose between the slave agent and the reference point selected by the slave agent in real time as the real-time pose;
- S5 perform pose deviation calculation according to the initial pose and the real-time pose, and determine the pose deviation between each slave agent and the master agent;
- the specific calculation of the motion trajectory planning and closed-loop control module is as follows:
- the shape splicing is based on the distribution of the three agents in the shape of a shape, and the geometric centroid O point is used as the overall transfer center to move, and the frontmost agent is used as the main agent.
- the two eight-agents are slave agents, and the initial relative poses between the three agents can be flexibly adapted according to the actual needs, that is, a and b in Figure 5 can be changed at will, where a is the front and rear of the main agent.
- the distance between the end face and the front face of the follower agent, b is the distance value between the geometric center points of the two follower agents.
- the coordinate system X 1 O 1 Y 1 is established from the center O 1 of the agent 1
- the coordinate system X 2 O 2 Y 2 is established from the center O 2 of the agent 2
- the main agent The center point O 0 of the rear face of the body establishes the coordinate system X 0 O 0 Y 0 , it can be known that:
- the initial pose from the center point O 1 of agent 1 in the coordinate system X 0 O 0 Y 0 of the main agent is:
- the initial pose from the center point O 2 of agent 2 in the coordinate system X 0 O 0 Y 0 of the main agent is:
- the contour data of end faces A and B are fitted to the distance and angle data of the two end face contour centers in the laser scanning radar, and the distance and angle data of the center point O 1 of the agent 1 and the center point O 2 of the agent 2 in the main agent are solved.
- the real-time measurement data of the center point of the front face of the slave agent 1 relative to the points A and B of the rear face of the main agent are ⁇ (d A1 ', ⁇ A1 '), (d B1 ', ⁇ B1 ') ⁇ , the real-time pose ( d x1 ', d y1 ' ,d z1 '):
- the two agents coordinately adjust the control strategy.
- the x, y, z three-axis pose deviation of the agent is recorded as ( ⁇ 1 , ⁇ 2 , ⁇ 3 , ⁇ 4 , ⁇ 5 , ⁇ 6 ), and is used as the input parameter of the adjustment control.
- the percentages ( ⁇ 1 , ⁇ 2 , ⁇ 3 , ⁇ 4 ) of the three-axis pose deviation data of the two agents are calculated , ⁇ 5 , ⁇ 6 ), and the maximum deviation percentage ⁇ max is obtained according to the percentage order.
- ⁇ (i) (e ⁇ (i) -e - ⁇ (i) )/(e ⁇ (i) +e - ⁇ (i) )
- the interpolation increment ⁇ i of each axis is adjusted to design the integral separation PID algorithm: when the interpolation increment ⁇ i is greater than the threshold value, the adjusted speed output should gradually increase, And when the error is small, the growth rate is small, and when the error is large, the growth rate is large; when the deviation value is less than or equal to the interpolation increment ⁇ i , the adjusted speed output should gradually decrease, namely:
- ⁇ i K pi ( ⁇ i- ⁇ i')+K ii * ⁇ i+K di *( ⁇ i-2* ⁇ i'+ ⁇ i"))
- the MECHATROLINK_II field motion bus is used to realize the topological connection of the multi-axis drive motors of the master and slave agents.
- the 20-axis linkage interpolation control is carried out according to the current speed V i of all motors of the master and slave agents and the target speed V i target , so as to realize the synchronous planning control of the master and slave agents.
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
一种面向异构特性大型装备多智能体协同自主转运系统，包括主控制器、指令接收及处理模块、导航单元、组合拼接路径规划模块、无线控制模块以及多个智能体。本发明通过多智能体自主路径规划和协同作业实现高端装备高效转运精准配送，以适应高端装备转运精准配送过程中产品的多样性和异构性需求，提升装备柔性化和适应程度，实现智能装备协同作业在精准转运精准配送与装配制造环节高效应用。以多元化自适应执行装备和转运模式实现柔性化、高效化转运精准配送过程创新应用。
Description
本申请要求于2021年02月26日提交中国专利局、申请号为202110219888.8、申请名称为“一种面向异构特性大型装备多智能体协同自主转运系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本发明涉及一种面向异构特性大型装备多智能体协同自主转运系统，具备单智能体独立运行作业及多智能体协同作业能力，具备室内外无人自主导航及定位能力。
在高端装备转运过程中产品往往具有尺寸大、重量大、运输困难等特点，超出现有常规转运设备尺寸、承载、精度的能力范围。同时，现有常规转运装备转运能力，转运效率较低，兼容性差，其转运模式缺乏灵活性、柔性化和集成化的系统解决方案，无法满足高质量、高效率、柔性化的智能转运要求，成为制约高端装备广泛应用的瓶颈问题。
发明内容
本发明的技术解决问题是：克服现有技术的不足，提供了一种面向异构特性大型装备多智能体协同自主转运系统，实现了高端装备高效化、柔性化、通用化转运，以适应高端装备超出现有单一转运设备尺寸、承载能力范围的转运需求，提升装备柔性化和适应程度，减少已往产品转运过程中的人力劳动，缩短产品转运时间。
本发明的技术解决方案是：
一种面向异构特性大型装备多智能体协同自主转运系统，包括：主控制器、指令接收及处理模块、导航单元、组合拼接路径规划模块、无线控制模块以及多个智能体；指令接收及处理模块接收外部输入的控制指令，该控制指令包括转运系统的运动指令及拼接模式，并将转运系统的运动指令及拼接模式发送给主控制器；主控制器将所述运动指令发送给导航单元，导航单元根据转运系统当前位置的实测值以及接收到的运动指令进行导航解算，生成转运系统总体偏航角、速度以及旋转角速度，并反馈给主控制器；
主控制器将拼接模式以及导航单元反馈的运动参数发送给组合拼接路径规划模块，组 合拼接路径规划模块根据拼接模式，将转运系统总体的运动参数分解为各智能体的理论运动参数，并通过无线控制模块发送给对应的各个智能体；各个智能体根据接收到的自身的理论运动参数，结合本智能体当前的位姿信息以及通过无线控制模块获得的其他智能体的位姿信息，进行轨迹规划的闭环补偿，生成实际运动参数，当前智能体按照该实际运动参数进行运转，从而实现多个智能体的协同自主转运。
进一步的，所述指令接收及处理模块接收的外部输入控制指令，来自于手持器发送的手动控制指令或者来自于外部调度系统的自动控制指令。
进一步的，导航单元包括：导航控制模块、导航传感器接收模块、视觉导航传感器、激光导航传感器以及iGPS导航传感器；
视觉导航传感器采集标识物与视觉导航传感器之间的局部相对位置信息；激光导航传感器和iGPS导航传感器采集传感器自身在全局坐标系下的坐标信息和姿态角；导航传感器接收模块接收三种传感器的输出结果并提供给导航控制模块；
当需要的定位精度在1mm～5mm之间时，导航控制模块对所述局部相对位置信息进行坐标转换，得到整体转运中心在局部坐标系下的相对坐标和姿态角，并根据得到的相对坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器；所述局部坐标系是指以标识物的中心为原点建立的坐标系；
当需要的定位精度大于等于5mm时，导航控制模块对激光导航传感器采集的传感器自身在全局坐标系下的坐标信息和姿态角进行坐标转换，得到整体转运中心在全局坐标系下的坐标和姿态角；并根据得到的坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器；所述全局坐标系是指：以智能体转运场地的顶点为原点建立坐标系；
当需要的定位精度小于等于1mm时，导航控制模块对iGPS导航传感器采集的传感器自身在全局坐标系下的坐标信息和姿态角进行坐标转换，得到整体转运中心在全局坐标系下的坐标和姿态角；并根据得到的坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器。
进一步的，外部输入控制指令中的转运系统的运动指令，具体包括：目标坐标X、目标坐标Y、目标角度θ、运行指令以及运行模式；
目标坐标X、目标坐标Y和目标角度θ为全局坐标系下的位置，运行指令包括停车、前进、后退、右横行、左横行、逆时针旋转90°以及顺时针旋转90°；运行模式包括紧急停车模式和常规模式；紧急停车模式是指整个转运系统即刻停止运行；常规模式 是指转运系统按照指令正常运行的模式。
进一步的，外部输入控制指令中的转运系统的拼接模式，具体包括：L型拼接、品型拼接、双车拼接以及四车拼接；
L型拼接是指三个智能体按照L形排列的形式；品型拼接是指三个智能体按照品形排列的形式；双车拼接是指两个智能体按照横向或纵向并排的形式排列；四车拼接是指四个智能体按照矩形排列的形式。
进一步的，所述组合拼接路径规划模块根据拼接模式，将转运系统总体的运动参数分解为各智能体的理论运动参数，具体包括如下步骤：
(1)以整体转运中心O为理论形心，计算各个智能体中心点到整体转运中心O的相对位姿；
(2)当拼接整体以姿态(υ
x,υ
y,ω
z)进行运动时，在坐标系XOY中，计算单位时间Δt内各个智能体中心的位置增量；坐标系XOY是以整体转运中心O为圆心的坐标系；
(3)计算拼接整体转运过程中各智能体的实时理论运动数据。
进一步的，
首先将多智能体进行编号i＝1,2,……n，当已知拼接模式下第i个智能体中心O
i相对选取整体转运中心点O之间的位姿为
其中
为第i个智能体中心O
i相对整体转运中心点O之间的距离，
为第i个智能体中心O
i相对整体转运中心点O之间的角度；
单位时间Δt内第i个智能体中心的X向、Y向以及角度增量(Δs
xi,Δs
yi,Δθ
zi)分别为：
则可知拼接整体转运过程中第i个智能体的实时理论运动数据(υ
xi,υ
yi,υ
ωi)为：
进一步的，所述智能体为基于麦克纳姆轮的转运平台，每个智能体上均设置有运动轨迹规划及闭环控制模块、位姿测量传感器接口模块、以及位姿测量传感器组合；
位姿测量传感器接口模块接收测量传感器组合采集的智能体当前位姿信息，提供 给运动轨迹规划及闭环控制模块，运动轨迹规划及闭环控制模块根据外部输入的智能体自身的理论运动参数，结合本智能体当前的位姿信息以及通过无线控制模块获得的其他智能体的位姿信息，进行轨迹规划的闭环补偿，生成实际运动数据。
进一步的，位姿测量传感器组合包括激光测距传感器、激光二维雷达以及二维光电传感器PSD；激光测距传感器用于测量相邻两个智能体之间的距离，激光二维雷达用于测量相邻两个智能体之间的相对姿态和角度，二维光电传感器PSD用于测量相邻两个智能体之间的姿态偏差。
进一步的，所述运动轨迹规划及闭环控制模块进行轨迹规划的闭环补偿，生成实际运动参数，具体为：
S1、将多智能体中最前端的智能体作为主智能体，其它作为从智能体；
S2、主智能体按照该智能体理论运动参数进行运动，其它从智能体根据与主智能体的位姿偏差进行运动补偿；
S3、每个从智能体在主智能体的后端面选取一个参考点；并确定每个从智能体与该从智能体所选参考点之间的位姿作为初始位姿；
S4、在多智能体协同转运过程中，每个从智能体实时获取该从智能体与该从智能体所选参考点之间的位姿作为实时位姿；
S5、根据所述初始位姿和实时位姿进行位姿偏差计算，确定每个从智能体与主智能体的位姿偏差；
S6、对每个从智能体，设定位姿调整阈值，然后利用位姿偏差与阈值计算位姿偏差百分比；
S7、选取所有百分比中的最大值，然后进行归一化后，确定每个从智能体的调整幅值；
S8、对每个从智能体，利用调整幅值进行每个方向幅值的耦合重计算；利用耦合重计算结果建立各方向的控制律；然后设定插补间隔，利用耦合重计算结果、控制律确定各方向的插补增量；最后设定控制阈值，利用插补增量确定的生成从智能体补偿运动参数(Δυ
xi，Δυ
yi，Δυ
ωi)；
S9、根据从智能体补偿运动参数和从智能体理论运动参数累加生成从智能体实际运动参数(υ'
xi，υ'
yi，υ'
ωi)：
进一步的，利用S8中所述的补偿运动参数，确定每个从智能体的各轮转速。
进一步的，S8中，对每个从智能体，利用调整幅值进行每个方向幅值的耦合重计算时，根据高度方向的调整幅值，对其他方向的位姿调整偏差进行重计算，获得重计算后的各方向调整幅值；
利用重计算后的各方向调整幅值建立各方向的控制律；
利用重计算后的各方向调整幅值建立各方向的控制律时，首先对重计算后的各方向调整幅值进行归一化，然后建立各方向的指数趋近律。
进一步的，S8中，当插补间隔大于控制阈值时，运动控制量逐渐增大；当插补间隔不超过控制阈值时，运动控制量逐渐减小。
本发明与现有技术相比的有益效果是：
(1)本发明对激光导航技术、iGPS高精度空间定位技术以及视觉测量等多导航技术进行融合，实现面向智能车间大空间范围内全局场景下的连续导航与高精度空间定位能力；
(2)本发明通过多智能体异型编组协同转运，实现转运设备同时兼容大型异构产品和常规产品转运的能力，解决高端装备转运过程中产品的多样性和异构性需求；提升装备柔性化和适应程度，减少已往产品转运对接过程中的人力劳动，缩短产品转运时间，实现智能装备协同作业在精准转运对接与装配制造环节高效应用。
(3)本发明采用多智能体间多智能体运动态势感知和实时在线运动轨迹补偿，智能体根据自身状态与轨迹预期进行实时在线分析，并进行在线轨迹补偿，实时监听运行故障状态进行整体保护措施，实现自主健康管理和在故障状态下的重规划能力。
(4)本发明以多元化自适应执行装备和转运模式实现柔性化、高效化转运对接过程创新应用，为大型重载产品的装运对接方式提供更好的解决措施。
(5)基于多智能体协同作业方式具有转运精度高、自适应组合、作业方便快捷等特点。
(6)多智能体协同作业时通过基于激光扫描雷达轮廓识别位姿测量系统周期性测量相对位姿，并与初始设定位姿进行偏差计算；多智能体之间通过5G网络实现运动指令、补偿参数和状态参数数据的快速无线交互；通过对多智能体的各轴实时位姿偏差 进行互约束性分析，并根据分析结果建立相应的位姿补偿调节控制策略，实现多智能体协同转运时的在线位姿补偿调节。
(7)本发明保证了多智能体协同转运过程的实时相对位姿控制精度。结合多智能体的自主异型编队协同控制技术，可以自适应多样性和异构型高端装备产品的柔性化转运对接，实现多智能体协同的高效率、通用化转运对接作业。采用该发明方法，解决了大型异构高端装备在狭小空间内的高效率转运、对接作业以及高精度定位问题。
图1为本发明多智能体主要组合拼接方式示意图，其中，(a)为L型三车拼接，(b)为品型三车拼接、(c)为双车拼接，(d)为四车拼接；
图2为本发明与图1四种拼接模式相对应的旋转中心选择示意图；
图3为本发明系统架构示意图；
图4为本发明系统整体数据流图；
图5为本发明品型拼接多智能体布局示意图；
图6为本发明多智能体协同转运过程中的相对位姿测量示意图。
下面结合附图对本发明的具体实施方式进行进一步的详细描述。
面向超大型载荷或者狭小异构空间内灵活转运需求，需要突破基于全向智能装备的多智能体协作和自主异型编队协同控制技术，具备基于多车异型编组及协同控制能力，实现高端装备高效化、柔性化、通用化转运，研制通用化转运平台，以适应高端装备产品转运多样性和异构性的要求，提升装备转运柔性化和适应程度，实现智能装备协同作业精准转运。
此外，构建转运全局场景下连续导航与面向精准配送环节的高精度空间定位方法，既满足产品的跨厂房、跨区域联合转运，又满足关键工位处的高精度定位，结合多传感器信息融合技术增强系统感知能力，增强数据的可信度，提高精度，扩展系统时间、空间覆盖率，增加系统的实时性和信息利用率等。保证高端装备在复杂环境中自动化转运精准配送过程的稳定可靠，提升现有设备装配精准配送过程执行效率与可靠性。
本发明涉及的面向异构特性大型装备多智能体协同自主转运系统既具备单车作业及多体协同作业能力，又具备室内外无人自主导航及定位能力。
如图3、图4所示，本发明提出的一种面向异构特性大型装备多智能体协同自主转运系统，包括：主控制器、指令接收及处理模块、导航单元、组合拼接路径规划模块、无线控制模块以及多个智能体；
指令接收及处理模块接收外部输入的控制指令，该控制指令包括转运系统的运动指令及拼接模式，并将转运系统的运动指令及拼接模式发送给主控制器；主控制器将所述运动指令发送给导航单元，导航单元根据转运系统当前位置的实测值以及接收到的运动指令进行导航解算，生成转运系统总体偏航角、速度以及旋转角速度，并反馈给主控制器；
主控制器将拼接模式以及导航单元反馈的运动参数发送给组合拼接路径规划模块，组合拼接路径规划模块根据拼接模式，将转运系统总体的运动参数分解为各智能体的理论运动参数，并通过无线控制模块发送给对应的各个智能体；各个智能体根据接收到的自身的理论运动参数，结合本智能体当前的位姿信息以及通过无线控制模块获得的其他智能体的位姿信息，进行轨迹规划的闭环补偿，生成实际运动参数，当前智能体按照该实际运动参数进行运转，从而实现多个智能体的协同自主转运。
指令接收及处理模块接收的外部输入控制指令，来自于手持器发送的手动控制指令或者来自于外部调度系统的自动控制指令，具体包括如下步骤：
(1)当为自动控制指令时，启动导航单元，解析自动导航指令，生成转运系统总体偏航角、速度以及旋转角速度(θ,υ,ω
z)，反馈给主控制器；
(2)当为手动控制指令时，解析手执器指令，获取车体运行的转运系统总体偏航角、速度以及旋转角速度(θ,υ,ω
z)，反馈给主控制器。
如图3所示，导航单元包括：导航控制模块、导航传感器接收模块、视觉导航传感器、激光导航传感器以及iGPS导航传感器；
视觉导航传感器采集标识物与视觉导航传感器之间的局部相对位置信息；激光导航传感器和iGPS导航传感器采集传感器自身在全局坐标系下的坐标信息和姿态角；导航传感器接收模块接收三种传感器的输出结果并提供给导航控制模块；
当需要的定位精度在1mm～5mm之间时，导航控制模块对所述局部相对位置信息进行坐标转换，得到整体转运中心在局部坐标系下的相对坐标和姿态角，并根据得到的相对坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器；所述局部坐标系是指以标识物的中心为原点建立的坐标系；
当需要的定位精度大于等于5mm时，导航控制模块对激光导航传感器采集的传感 器自身在全局坐标系下的坐标信息和姿态角进行坐标转换，得到整体转运中心在全局坐标系下的坐标和姿态角；并根据得到的坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度(θ,υ,ω
z)，反馈给主控制器；所述全局坐标系是指：以智能体转运场地的顶点为原点建立坐标系；
当需要的定位精度小于等于1mm时，导航控制模块对iGPS导航传感器采集的传感器自身在全局坐标系下的坐标信息和姿态角进行坐标转换，得到整体转运中心在全局坐标系下的坐标和姿态角；并根据得到的坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度(θ,υ,ω
z)，反馈给主控制器。
外部输入控制指令中的转运系统的运动指令，具体包括：目标坐标X、目标坐标Y、目标角度θ、运行指令以及运行模式；
目标坐标X、目标坐标Y和目标角度θ为全局坐标系下的位置，运行指令包括停车、前进、后退、右横行、左横行、逆时针旋转90°以及顺时针旋转90°；运行模式包括紧急停车模式和常规模式；紧急停车模式是指整个转运系统即刻停止运行；常规模式是指转运系统按照指令正常运行的模式。
如图1所示，外部输入控制指令中的转运系统的拼接模式，具体包括：L型拼接、品字型拼接、双车拼接以及四车拼接；
L型拼接是指三个智能体按照L形排列的形式；品字型拼接是指三个智能体按照品形排列的形式；双车拼接是指两个智能体按照横向或纵向并排的形式排列；四车拼接是指四个智能体按照矩形排列的形式。
组合拼接路径规划模块根据拼接模式，将转运系统总体的运动参数分解为各智能体的理论运动参数，具体包括如下步骤：
(1)以整体转运中心O为理论形心，计算各个智能体中心点到整体转运中心O的相对位姿；
(2)已知转运系统总体偏航角、速度以及旋转角速度为(θ,υ,ω
z)，可转换为拼接整体运行姿态(υ
x,υ
y,ω
z)，如下：
(3)当拼接整体以姿态(υ
x,υ
y,ω
z)进行运动时，在坐标系XOY中，计算单位时间 Δt内各个智能体中心的位置增量；组合拼接方式如图1所示，同时自定义组合拼接时的整体转运中心，整体转运中心一般定义为拼接整体几何中心、各单智能体中心或拼接连线的中心，如图2所示；坐标系XOY是以整体转运中心O为圆心的坐标系；
首先将多智能体进行编号i＝1,2,……n，当已知拼接模式下第i个智能体中心O
i相对选取整体转运中心点O之间的位姿为
其中
为第i个智能体中心O
i相对整体转运中心点O之间的距离，
为第i个智能体中心O
i相对整体转运中心点O之间的角度；
单位时间Δt内第i个智能体中心的X向、Y向位置增量以及角度增量(Δs
xi,Δs
yi,Δθ
zi)分别为：
(4)计算拼接整体转运过程中各智能体的实时理论运动数据。
可知拼接整体转运过程中第i个智能体的实时理论运动数据(υ
xi,υ
yi,υ
ωi)为：
本发明所述智能体为基于麦克纳姆轮的转运平台，每个智能体上均设置有运动轨迹规划及闭环控制模块、位姿测量传感器接口模块、以及位姿测量传感器组合；
位姿测量传感器接口模块采用RS422、网口和RS232用于接收测量传感器组合采集的智能体当前位姿信息，提供给运动轨迹规划及闭环控制模块，运动轨迹规划及闭环控制模块根据外部输入的智能体自身的理论运动参数，结合本智能体当前的位姿信息以及通过无线控制模块获得的其他智能体的位姿信息，进行轨迹规划的闭环补偿，生成实际运动数据。
如图3所示，位姿测量传感器组合包括激光测距传感器、激光二维雷达以及二维光电传感器PSD；激光测距传感器用于测量相邻两个智能体之间的距离，激光二维雷达用于测量相邻两个智能体之间的相对姿态和角度，二维光电传感器PSD用于测量相邻两个智能体之间的姿态偏差。
在整个运动规划过程中，涉及到的运动数据、位姿信息传输都对实时性和同步性 有着高需求。多智能体高速同步无线控制模块采用基于WIFI网络拓扑方式架构无线通讯链路,具备多达255个智能体的接入能力，响应频率可达100Hz。同时采用高速的800MHz主频32位处理器进行时间同步算法设计，同步精度可达微秒级。TDMA时隙调度机制事先为无线网络各个节点分配通道资源，且各节点智能在分配的时隙上发送数据，有效保障了通信的实时性、可靠性，特别当时间同步精度要求较高时能更进一步提高实时和可靠性。各单智能体通过WIFI无线链路进行互相通讯。
本发明中，运动轨迹规划及闭环控制模块进行轨迹规划的闭环补偿，生成实际运动参数，具体为：
S1、将多智能体中最前端的智能体作为主智能体，其它作为从智能体；
S2、主智能体按照该智能体理论运动参数进行运动，其它从智能体根据与主智能体的位姿偏差进行运动补偿；
S3、每个从智能体在主智能体的后端面选取一个参考点；并确定每个从智能体与该从智能体所选参考点之间的位姿作为初始位姿；
S4、在多智能体协同转运过程中，每个从智能体实时获取该从智能体与该从智能体所选参考点之间的位姿作为实时位姿；
S5、根据所述初始位姿和实时位姿进行位姿偏差计算，确定每个从智能体与主智能体的位姿偏差；
S6、对每个从智能体，设定位姿调整阈值，然后利用位姿偏差与阈值计算位姿偏差百分比；
S7、选取所有百分比中的最大值，然后进行归一化后，确定每个从智能体的调整幅值；
S8、对每个从智能体，利用调整幅值进行每个方向幅值的耦合重计算；利用耦合重计算结果建立各方向的控制律；然后设定插补间隔，利用耦合重计算结果、控制律确定各方向的插补增量；最后设定控制阈值，利用插补增量确定的生成从智能体补偿运动参数(Δυ
xi，Δυ
yi，Δυ
ωi)；
S9、根据从智能体补偿运动参数和从智能体理论运动参数累加生成从智能体实际运动参数(υ'
xi，υ'
yi，υ'
ωi)：
利用S8中所述的补偿运动参数，确定每个从智能体的各轮转速。
S8中，对每个从智能体，利用调整幅值进行每个方向幅值的耦合重计算时，根据高度方向的调整幅值，对其他方向的位姿调整偏差进行重计算，获得重计算后的各方向调整幅值；
利用重计算后的各方向调整幅值建立各方向的控制律；
利用重计算后的各方向调整幅值建立各方向的控制律时，首先对重计算后的各方向调整幅值进行归一化，然后建立各方向的指数趋近律。
S8中，当插补间隔大于控制阈值时，运动控制量逐渐增大；当插补间隔不超过控制阈值时，运动控制量逐渐减小。
以品型拼接方式为例，运动轨迹规划及闭环控制模块具体计算如下：
(1)如图5所示，品型拼接是以三个智能体之间呈品字形分布并以几何形心O点为整体转运中心进行运动，以最前端的智能体作为主智能体，后两个八智能体为从智能体，三智能体之间的初始相对位姿可随实际需求进行柔性化自适应，即图5中的a、b可随意变化，其中a为前主智能体后端面与后从智能体前端面的距离值，b为后两个从智能体几何中心点之间的距离值。
(2)如图6所示，以从智能体1的中心O
1建立坐标系X
1O
1Y
1，以从智能体2的中心O
2建立坐标系X
2O
2Y
2，以主智能体后端面中心点O
0建立坐标系X
0O
0Y
0，则可知：
从智能体1中心点O
1在主智能体的坐标系X
0O
0Y
0的初始位姿为：
从智能体2中心点O
2在主智能体的坐标系X
0O
0Y
0的初始位姿为：
(3)如图5所示，按照转运对象的实际质量特性选择合适的a、b参数，即形成了多智能体的初始分布设计。根据多智能体协同自主转运系统的物理模型计算从智能体1中心点O
1在主智能体的坐标系X
0O
0Y
0的初始位姿以及从智能体2中心点O
2在主智能体的坐标系X
0O
0Y
0的理论初始位姿；同时在多智能体协同作业过程中通过TCP/IP通信协议以100Hz的通信周期读取激光扫描雷达实时测量从智能体相对主智能体后端 面A、B的轮廓数据并拟合出两个端面轮廓中心在激光扫描雷达的距离和角度数据，并求解从智能体1中心点O
1、从智能体2中心点O
2在主智能体的坐标系X
0O
0Y
0的实时位姿。
(4)多智能体协同转运系统在运动过程中，从智能体1的前端面中心点相对于主智能体后端面A、B点的实时测量数据为{(d
A1',θ
A1'),(d
B1',θ
B1')}，则可得从智能体1几何中心点O
1在主智能体的坐标系X
0O
0Y
0的实时位姿(d
x1',d
y1',d
z1')：
从智能体2的前端面中心点相对于主智能体后端面A、B点的实时测量数据为{(d
A2',θ
A2'),(d
B2',θ
B2')}，则可得从智能体2几何中心点O
1在主智能体的坐标系X
0O
0Y
0的实时位姿(d
x2',d
y2',d
z2')：
(5)计算从智能体相对主智能体后端面的实时位姿(d
xi',d
yi',d
zi')与初始设定位姿(d
xi,d
yi,d
zi)的偏差为(Δε
xi,Δε
yi,Δε
zi):
其中，实时根据后两从智能体中心点在主智能体的坐标系X
0O
0Y
0的实时位姿与初 始位姿之差联合建立后双智能体协同调整控制策略：首先以同一时刻两智能体的x，y，z三轴位姿偏差记为(Δε
1,Δε
2,Δε
3,Δε
4,Δε
5,Δε
6)，并作为调整控制的输入参数。
按照设定姿态调整阈值(ξ
1,ξ
2,ξ
3,ξ
4,ξ
5,ξ
6)计算两智能体的三轴位姿偏差数据的百分比(ρ
1,ρ
2,ρ
3,ρ
4,ρ
5,ρ
6)，按照百分比大小排序求得最大偏差百分比ρ
max。按照如下步骤进行：
a)以最大偏差百分比ρ
max的轴姿态偏差进行两智能体的各轴调整幅值Δμ
i计算，可知有：
b)根据两智能体的各轴调整幅值进行耦合重计算：
c)根据两智能体耦合重计算后各轴的调整幅值建立各自的控制律：
δ(i)＝(Δμ'
i/ξ
i)/max(Δμ'
i/ξ
i)
λ(i)＝(e
δ(i)-e
-δ(i))/(e
δ(i)+e
-δ(i))
d)以Τ
τ为插补间隔，计算各轴的插补增量：
Δσ
i＝Δμ'
iΤ
τλ(i)
e)设当前的各轴给定控制参数为
以当前控制参数的1/10为阈值为依据对各轴姿态调整插补增量Δσ
i设计积分分离PID算法：当插补增量Δσ
i大于阈值时，调整的速度输出量应逐渐增大，且误差小时增长率小，误差大时增长率大；当偏差值小于或等于插补增量Δσ
i时，调整的速度输出量应逐渐减小，即有：
Δυ
i＝K
pi(Δσi-Δσi')+K
ii*Δσi+K
di*(Δσi-2*Δσi'+Δσi”))
故知姿态调整后的运动控制量为:
故可以求出各个轮的转速：
其中，采用MECHATROLINK_II现场运动总线实现主、从智能体多轴驱动电机的拓扑连接。根据主、从智能体所有电机的当前速度V
i实际和目标速度V
i目标进行二十轴的联动插补控制，实现主、从智能体的同步规划控制。
本发明未详细说明部分属本领域技术人员公知常识。
Claims (13)
- 一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于包括：主控制器、指令接收及处理模块、导航单元、组合拼接路径规划模块、无线控制模块以及多个智能体；指令接收及处理模块接收外部输入的控制指令，该控制指令包括转运系统的运动指令及拼接模式，并将转运系统的运动指令及拼接模式发送给主控制器；主控制器将所述运动指令发送给导航单元，导航单元根据转运系统当前位置的实测值以及接收到的运动指令进行导航解算，生成转运系统总体偏航角、速度以及旋转角速度，并反馈给主控制器；主控制器将拼接模式以及导航单元反馈的运动参数发送给组合拼接路径规划模块，组合拼接路径规划模块根据拼接模式，将转运系统总体的运动参数分解为各智能体的理论运动参数，并通过无线控制模块发送给对应的各个智能体；各个智能体根据接收到的自身的理论运动参数，结合本智能体当前的位姿信息以及通过无线控制模块获得的其他智能体的位姿信息，进行轨迹规划的闭环补偿，生成实际运动参数，当前智能体按照该实际运动参数进行运转，从而实现多个智能体的协同自主转运。
- 根据权利要求1所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：所述指令接收及处理模块接收的外部输入控制指令，来自于手持器发送的手动控制指令或者来自于外部调度系统的自动控制指令。
- 根据权利要求1所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：导航单元包括：导航控制模块、导航传感器接收模块、视觉导航传感器、激光导航传感器以及iGPS导航传感器；视觉导航传感器采集标识物与视觉导航传感器之间的局部相对位置信息；激光导航传感器和iGPS导航传感器采集传感器自身在全局坐标系下的坐标信息和姿态角；导航传感器接收模块接收三种传感器的输出结果并提供给导航控制模块；当需要的定位精度在1mm～5mm之间时，导航控制模块对所述局部相对位置信息进行坐标转换，得到整体转运中心在局部坐标系下的相对坐标和姿态角，并根据得到的相对坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器；所述局部坐标系是指以标识物的中心为原点建立的坐标系；当需要的定位精度大于等于5mm时，导航控制模块对激光导航传感器采集的传感器自身在全局坐标系下的坐标信息和姿态角进行坐标转换，得到整体转运中心在全局坐标系下的坐标和姿态角；并根据得到的坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器；所述全局坐标系是指：以智能体转运场地的顶点为原点建立坐标系；当需要的定位精度小于等于1mm时，导航控制模块对iGPS导航传感器采集的传感器自身在全局坐标系下的坐标信息和姿态角进行坐标转换，得到整体转运中心在全局坐标系下的坐标和姿态角；并根据得到的坐标、姿态角以及目标路径偏差，生成转运系统总体偏航角、速度以及旋转角速度，反馈给主控制器。
- 根据权利要求3所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：外部输入控制指令中的转运系统的运动指令，具体包括：目标坐标X、目标坐标Y、目标角度θ、运行指令以及运行模式；目标坐标X、目标坐标Y和目标角度θ为全局坐标系下的位置，运行指令包括停车、前进、后退、右横行、左横行、逆时针旋转90°以及顺时针旋转90°；运行模式包括紧急停车模式和常规模式；紧急停车模式是指整个转运系统即刻停止运行；常规模式是指转运系统按照指令正常运行的模式。
- 根据权利要求3所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：外部输入控制指令中的转运系统的拼接模式，具体包括：L型拼接、品型拼接、双车拼接以及四车拼接；L型拼接是指三个智能体按照L形排列的形式；品型拼接是指三个智能体按照品型排列的形式；双车拼接是指两个智能体按照横向或纵向并排的形式排列；四车拼接是指四个智能体按照矩形排列的形式。
- 根据权利要求1所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：所述组合拼接路径规划模块根据拼接模式，将转运系统总体的运动参数分解为各智能体的理论运动参数，具体包括如下步骤：(1)以整体转运中心O为理论形心，计算各个智能体中心点到整体转运中心O的相对位姿；(2)当拼接整体以姿态(υ x,υ y,ω z)进行运动时，在坐标系XOY中，计算单位时间Δt内各个智能体中心的位置增量；坐标系XOY是以整体转运中心O为圆心的坐标系；(3)计算拼接整体转运过程中各智能体的实时理论运动数据。
- 根据权利要求6所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：首先将多智能体进行编号i＝1,2,……n，当已知拼接模式下第i个智能体中心O i相对选取整体转运中心点O之间的位姿为其中
为第i个智能体中心O i相对整体转运中心点O之间的距离，
为第i个智能体中心O i相对整体转运中心点O之间的角度；
单位时间Δt内第i个智能体中心的X向、Y向位置增量以及角度增量(Δs xi,Δs yi,Δθ zi)分别为：
则可知拼接整体转运过程中第i个智能体的实时理论运动数据(υ xi,υ yi,υ ωi)为：
- 根据权利要求1所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：所述智能体为基于麦克纳姆轮的转运平台，每个智能体上均设置有运动轨迹规划及闭环控制模块、位姿测量传感器接口模块、以及位姿测量传感器组合；位姿测量传感器接口模块接收测量传感器组合采集的智能体当前位姿信息，提供给运动轨迹规划及闭环控制模块，运动轨迹规划及闭环控制模块根据外部输入的智能体自身的理论运动参数，结合本智能体当前的位姿信息以及通过无线控制模块获得的其他智能体的位姿信息，进行轨迹规划的闭环补偿，生成实际运动数据。
- 根据权利要求8所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：位姿测量传感器组合包括激光测距传感器、激光二维雷达以及二维光电传感器PSD；激光测距传感器用于测量相邻两个智能体之间的距离，激光二维雷达用于测量相邻两个智能体之间的相对姿态和角度，二维光电传感器PSD用于测量相邻两个智能体之间的姿态偏差。
- 根据权利要求8所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：所述运动轨迹规划及闭环控制模块进行轨迹规划的闭环补偿，生成实际运动参数，具体为：S1、将多智能体中最前端的智能体作为主智能体，其它作为从智能体；S2、主智能体按照该智能体理论运动参数进行运动，其它从智能体根据与主智能体的位姿偏差进行运动补偿；S3、每个从智能体在主智能体的后端面选取一个参考点；并确定每个从智能体与该从智能体所选参考点之间的位姿作为初始位姿；S4、在多智能体协同转运过程中，每个从智能体实时获取该从智能体与该从智能体所选参考点之间的位姿作为实时位姿；S5、根据所述初始位姿和实时位姿进行位姿偏差计算，确定每个从智能体与主智能体的位姿偏差；S6、对每个从智能体，设定位姿调整阈值，然后利用位姿偏差与阈值计算位姿偏差百分比；S7、选取所有百分比中的最大值，然后进行归一化后，确定每个从智能体的调整幅值；S8、对每个从智能体，利用调整幅值进行每个方向幅值的耦合重计算；利用耦合重计算结果建立各方向的控制律；然后设定插补间隔，利用耦合重计算结果、控制律确定各方向的插补增量；最后设定控制阈值，利用插补增量确定的生成从智能体补偿运动参数(Δυ xi，Δυ yi，Δυ ωi)；S9、根据从智能体补偿运动参数和从智能体理论运动参数累加生成从智能体实际运动参数(υ' xi，υ' yi，υ' ωi)：
- 根据权利要求10所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：利用S8中所述的补偿运动参数，确定每个从智能体的各轮转速。
- 根据权利要求10所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：S8中，对每个从智能体，利用调整幅值进行每个方向幅值的耦合重计算时，根据高度方向的调整幅值，对其他方向的位姿调整偏差进行重计算，获得重计算后的各方向调整幅值；利用重计算后的各方向调整幅值建立各方向的控制律；利用重计算后的各方向调整幅值建立各方向的控制律时，首先对重计算后的各方向调整幅值进行归一化，然后建立各方向的指数趋近律。
- 根据权利要求10所述的一种面向异构特性大型装备多智能体协同自主转运系统，其特征在于：S8中，当插补间隔大于控制阈值时，运动控制量逐渐增大；当插补间隔不超过控制阈值时，运动控制量逐渐减小。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110219888.8 | 2021-02-26 | ||
| CN202110219888.8A CN113003229B (zh) | 2021-02-26 | 2021-02-26 | 一种面向异构特性大型装备多智能体协同自主转运系统 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022179179A1 true WO2022179179A1 (zh) | 2022-09-01 |
Family
ID=76386575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2021/129478 WO2022179179A1 (zh) | 2021-02-26 | 2021-11-09 | 一种面向异构特性大型装备多智能体协同自主转运系统 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113003229B (zh) |
| WO (1) | WO2022179179A1 (zh) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115893032A (zh) * | 2022-11-23 | 2023-04-04 | 哈尔滨工业大学芜湖机器人产业技术研究院 | 用于厢式集装箱的自动装车系统 |
| CN118170034A (zh) * | 2024-05-16 | 2024-06-11 | 山东科技大学 | 一种具有通信丢包的异构多智能体系统均方一致控制方法 |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113031595B (zh) * | 2021-02-26 | 2023-02-03 | 北京卫星制造厂有限公司 | 一种基于多智能体协同转运实时在线位姿补偿控制方法 |
| CN113003229B (zh) * | 2021-02-26 | 2023-03-24 | 北京卫星制造厂有限公司 | 一种面向异构特性大型装备多智能体协同自主转运系统 |
| CN113433822B (zh) * | 2021-08-26 | 2021-11-09 | 深圳市创能亿科科技开发有限公司 | 智能实验室的管理系统 |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6049745A (en) * | 1997-02-10 | 2000-04-11 | Fmc Corporation | Navigation system for automatic guided vehicle |
| CN110837257A (zh) * | 2019-11-19 | 2020-02-25 | 北京卫星制造厂有限公司 | 一种基于iGPS与视觉的AGV复合定位导航系统 |
| CN111367285A (zh) * | 2020-03-18 | 2020-07-03 | 华东理工大学 | 一种轮式移动小车协同编队与路径规划方法 |
| CN111399509A (zh) * | 2020-03-24 | 2020-07-10 | 华中科技大学 | 一种多移动机器人协同转运控制方法及系统 |
| CN111766879A (zh) * | 2020-06-24 | 2020-10-13 | 天津大学 | 一种基于自主协同导航的智能车编队系统 |
| CN111813122A (zh) * | 2020-07-14 | 2020-10-23 | 南京航空航天大学苏州研究院 | 一种基于全向移动agv的多车协同搬运快速变换队列方法 |
| CN112034864A (zh) * | 2020-11-04 | 2020-12-04 | 天津航天机电设备研究所 | 智能移动机器人多车协同控制浮动系统及其控制方法 |
| CN113003229A (zh) * | 2021-02-26 | 2021-06-22 | 北京卫星制造厂有限公司 | 一种面向异构特性大型装备多智能体协同自主转运系统 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI590019B (zh) * | 2016-03-04 | 2017-07-01 | 美和學校財團法人美和科技大學 | 雙載具協同運輸系統及其控制方法 |
| KR101984922B1 (ko) * | 2017-09-13 | 2019-05-31 | 엘지전자 주식회사 | 차량의 군집 주행 방법 및 차량 |
| CN108664024B (zh) * | 2018-05-08 | 2020-09-25 | 清华大学 | 无人车网络编队的运动规划和协作定位方法及装置 |
| CN109144076B (zh) * | 2018-10-31 | 2020-05-22 | 吉林大学 | 一种多车辆横纵向耦合协同控制系统及控制方法 |
| CN111123985B (zh) * | 2020-01-14 | 2023-08-11 | 北京卫星制造厂有限公司 | 一种基于激光扫描的多车自动调姿跟随系统 |
-
2021
- 2021-02-26 CN CN202110219888.8A patent/CN113003229B/zh active Active
- 2021-11-09 WO PCT/CN2021/129478 patent/WO2022179179A1/zh active Application Filing
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6049745A (en) * | 1997-02-10 | 2000-04-11 | Fmc Corporation | Navigation system for automatic guided vehicle |
| CN110837257A (zh) * | 2019-11-19 | 2020-02-25 | 北京卫星制造厂有限公司 | 一种基于iGPS与视觉的AGV复合定位导航系统 |
| CN111367285A (zh) * | 2020-03-18 | 2020-07-03 | 华东理工大学 | 一种轮式移动小车协同编队与路径规划方法 |
| CN111399509A (zh) * | 2020-03-24 | 2020-07-10 | 华中科技大学 | 一种多移动机器人协同转运控制方法及系统 |
| CN111766879A (zh) * | 2020-06-24 | 2020-10-13 | 天津大学 | 一种基于自主协同导航的智能车编队系统 |
| CN111813122A (zh) * | 2020-07-14 | 2020-10-23 | 南京航空航天大学苏州研究院 | 一种基于全向移动agv的多车协同搬运快速变换队列方法 |
| CN112034864A (zh) * | 2020-11-04 | 2020-12-04 | 天津航天机电设备研究所 | 智能移动机器人多车协同控制浮动系统及其控制方法 |
| CN113003229A (zh) * | 2021-02-26 | 2021-06-22 | 北京卫星制造厂有限公司 | 一种面向异构特性大型装备多智能体协同自主转运系统 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115893032A (zh) * | 2022-11-23 | 2023-04-04 | 哈尔滨工业大学芜湖机器人产业技术研究院 | 用于厢式集装箱的自动装车系统 |
| CN118170034A (zh) * | 2024-05-16 | 2024-06-11 | 山东科技大学 | 一种具有通信丢包的异构多智能体系统均方一致控制方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113003229A (zh) | 2021-06-22 |
| CN113003229B (zh) | 2023-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022179179A1 (zh) | 一种面向异构特性大型装备多智能体协同自主转运系统 | |
| WO2021114888A1 (zh) | 一种双agv协同运载控制系统及方法 | |
| CN110244772A (zh) | 移动机器人的领航跟随系统和领航跟随控制方法 | |
| CN111399509B (zh) | 一种多移动机器人协同转运控制方法及系统 | |
| CN108286949B (zh) | 一种可移动式三维检测机器人系统 | |
| CN110341711A (zh) | 一种基于码头环境的行驶轨迹生成系统及方法 | |
| CN110703751B (zh) | 一种多移动机器人系统半自主编队及避障控制方法 | |
| CN109445438B (zh) | 基于地图分享的巡航装置的巡航控制方法及系统 | |
| WO2022252221A1 (zh) | 一种移动机器人队列系统及路径规划、跟随方法 | |
| CN105751200B (zh) | 全液压自主移动机械臂的动作方法 | |
| CN111367285B (zh) | 一种轮式移动小车协同编队与路径规划方法 | |
| CN105527960A (zh) | 一种基于领航跟随的移动机器人编队控制方法 | |
| CN111813121A (zh) | 基于距离-角度优先级的多移动机器人编队避障方法 | |
| CN111634636B (zh) | 一种斗轮机的全自动取料控制系统 | |
| CN114684202B (zh) | 一种用于自动驾驶车辆的智能系统及其集成控制方法 | |
| CN107102641A (zh) | 一种基于激光引导双轮差速agv的原地偏航旋转方法 | |
| CN110450991B (zh) | 微纳卫星集群捕获空间非合作目标的方法 | |
| CN111921733B (zh) | 一种喷涂作业方法和系统 | |
| CN108638052A (zh) | 一种闭链式多臂机器人柔顺控制方法 | |
| CN109814550A (zh) | 一种用于封闭园区的无人运输车 | |
| CN214846390U (zh) | 基于自动导引小车的动态环境避障系统 | |
| WO2022179180A1 (zh) | 一种基于多智能体协同转运实时在线位姿补偿控制方法 | |
| CN117546681A (zh) | 引导机器人采摘和运送水果的协同作业方法、系统及平台 | |
| CN108890640A (zh) | 一种基于同步定位与地图构建技术的机器人设备校准方法 | |
| CN115284293A (zh) | 适应复杂任务的空间站机械臂多模式路径规划系统和方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21927603 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 21927603 Country of ref document: EP Kind code of ref document: A1 |