WO2021114888A1 - Dual-agv collaborative carrying control system and method - Google Patents

Dual-agv collaborative carrying control system and method Download PDF

Info

Publication number
WO2021114888A1
WO2021114888A1 PCT/CN2020/122756 CN2020122756W WO2021114888A1 WO 2021114888 A1 WO2021114888 A1 WO 2021114888A1 CN 2020122756 W CN2020122756 W CN 2020122756W WO 2021114888 A1 WO2021114888 A1 WO 2021114888A1
Authority
WO
WIPO (PCT)
Prior art keywords
agv
pilot
deviation
following
dual
Prior art date
Application number
PCT/CN2020/122756
Other languages
French (fr)
Chinese (zh)
Inventor
钱晓明
楼佩煌
宋阳
Original Assignee
南京航空航天大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南京航空航天大学 filed Critical 南京航空航天大学
Publication of WO2021114888A1 publication Critical patent/WO2021114888A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/4189—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by the transport system
    • G05B19/41895—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by the transport system using automatic guided vehicles [AGV]
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0287—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling
    • G05D1/0291—Fleet control
    • G05D1/0295—Fleet control by at least one leading vehicle of the fleet
    • 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]
    • 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
    • Y02P90/60—Electric or hybrid propulsion means for production processes

Definitions

  • the invention relates to a control system and a method for a dual AGV cooperative carrying system, in particular to a dual AGV cooperative carrying control method for the conveying of large objects, and belongs to the field of intelligent industrial robots.
  • the traditional long-distance AGV automated logistics method has gradually been unable to meet the tasks of manufacturing, logistics transfer and delivery.
  • most of the traditional transportation systems such as gantry cranes and tower cranes are still used for the transportation of large objects.
  • the C919 assembly workshop in Shanghai still uses the rail crane located in the upper part of the workshop to transport and dock the fuselage section and the partial wing during the docking work of the wing and fuselage.
  • the automated transportation of containers located in Shanghai Yangshan Port and Xiamen Yuanhai Automated Terminals uses the method of increasing the size of a single AGV to realize the automated logistics of large containers.
  • AGVs are used to form a multi-mobile robot system, so that it maintains a fixed formation while operating automatically, thereby providing a carrying capacity that a single AGV does not have.
  • the use of multiple AGVs for coordinated logistics handling has the following advantages: 1. It is convenient to manage AGVs; 2. It improves the stability and robustness of the logistics transportation system; 3. Expands the use of AGVs , To make its transportation methods more diversified; 4. Improve the carrying utilization rate of AGV; 5. The system has higher safety, robustness and flexibility.
  • Multiple AGVs carry out coordinated and synchronized operation and formation maintenance during the operation process to realize the coordinated handling of materials and other multiple AGVs, which will greatly improve the current application status of automated transportation of large-volume and heavy-weight materials, while further increasing the utilization rate of AGVs and optimizing logistics
  • the allocation of resources has a wide range of application scenarios.
  • the Chinese invention patent with publication number CN10418899A proposes a flexibly connected dual-mobile robot cooperative handling system and its composite navigation device.
  • the patent only relates to the mechanical structure of the dual-mobile robot, and does not involve the description of the control method. It is different from the multi-AGV cooperation method mentioned in an AGV-based automatic handling system and its multi-AGV cooperation method proposed by the Chinese invention patent with the publication number CN109062150A.
  • This cooperation method is formed on the basis of a single AGV operation system.
  • AGV operation task planning and scheduling management, the control subject at a single moment is still a single AGV, and the design purpose of the present invention is a dual AGV coordinated carrying control method.
  • the control subject at the same time is a dual AGV, so as to form a dual AGV for the same object
  • the collaboration is carried out simultaneously.
  • the designer of the present invention proposes a dual-AGV oriented based on the path tracking and combined with the pilot-following multi-agent basic architecture
  • the control method of coordinated transportation further expands the use value of AGV in the industry.
  • the present invention proposes a dual AGV cooperative carrying control method, which is used for dual AGV coordination The formation of the transportation system is maintained and the realization of automatic operation.
  • the present invention constructs a dual AGV cooperative carrier control method, adopts two omnidirectional mobile AGVs, combines the path tracking method and the pilot-follow method, and establishes a cooperative control model under the path tracking with a three-layer topology structure
  • the kinematics control model of the dual AGV cooperative delivery system, and the discrete control model based on time-domain rolling predictive control is used to optimize the kinematics model to realize the stable and reliable cooperative operation of the dual AGV system.
  • a dual AGV coordinated carrying control method the dual AGV is omnidirectional movement, the front and rear layout, the former is the pilot AGV, the latter is the follower AGV, the method includes the following steps:
  • Step 1 Collect the path distance deviation e x1 (t) of the pilot AGV at time t, the path angle deviation e ⁇ 1 (t), the formation angle deviation ⁇ 1 relative to the workpiece and the wheel speed, and the path distance deviation e of the following AGV x2 (t), path angle deviation e ⁇ 2 (t), formation distance deviation ⁇ L(t), formation angle deviation relative to the workpiece ⁇ 2 and wheel speed;
  • Step 2 According to the wheel speed of the pilot AGV and the following AGV, obtain the side shift velocity component v x1 , the forward velocity component v y1 and the angular velocity w 1 of the pilot AGV, the follow AGV side shift velocity component v x2 , the forward velocity component v y2 and Angular velocity w 2 ;
  • Step 3 Let among them, Input e x1 (t), e ⁇ 1 (t), e x2 (t), e ⁇ 2 (t), ⁇ L(t), v y1 , ⁇ 1 , ⁇ 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period expression;
  • Step 4 Estimate the input vector according to the speed state and deviation state of the pilot AGV and following the AGV at time t As well as the deviation change model, the optimization strategy of time-domain rolling predictive control is adopted to obtain the optimized input vector
  • Step 5 The input vector will be optimized It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system.
  • the kinematics control model is:
  • step 4 has the following form:
  • ⁇ L(t+1) ⁇ L(t)+T(v y1 cos ⁇ 1 -v x1 sin ⁇ 1 +v x2 sin ⁇ 2 +v y2 cos ⁇ 2 ).
  • e 1 (t) [e x1 (t) e ⁇ 1 (t)] T
  • u 1 (t) [v x1 (t) w ⁇ 1 (t)] T
  • the weight matrix Q and R are Positive semi-definite symmetric matrix; set the cost function weight matrix Q and R as follows:
  • step 4 the optimization strategy of time-domain rolling predictive control is adopted, and the optimized input vector obtained by solving the quadratic programming problem of the objective function H at time t
  • the specific steps include:
  • Step 4.1 Estimate the input vector Substituted into the deviation change model of claim 3 to predict the pilot deviation e x1 (t+T) and e ⁇ 1 (t+T) at t+T, and the following AGV deviation e x2 (t+T), e ⁇ 2 (t +T) and ⁇ L(t+T), and substituted into them to obtain the terminal penalty function G.
  • Step 4.2 Lead the AGV deviation state e x1 (t) and e ⁇ 1 (t) from time t and follow the AGV related deviation state e x2 (t), e ⁇ 2 (t) and ⁇ L(t), combined with the estimated input vector Obtain the cost function L.
  • Step 4.5 When the dual AGV system runs to time t+ ⁇ , update time t to t+ ⁇ , and repeat steps 1 to 5.
  • the dual AGV cooperative carrier control system includes a sensor communication layer, a data fusion processing layer and a motion control layer;
  • the motion control layer includes a pilot AGV controller and a follower AGV controller, a pilot AGV controller and a follower AGV control
  • the device is respectively connected to the wheel motor drivers of the pilot AGV and the following AGV;
  • the sensor communication layer is used to monitor the path deviation, formation deviation and wheel speed of the dual AGV, and transmit the monitored information to the data fusion processing layer for data fusion processing
  • the layer performs fusion and processing of the information to obtain the speed and angular velocity input of the pilot AGV and the following AGV, and send the speed and angular velocity input of the pilot AGV and the following AGV to the pilot AGV and the following AGV controller respectively;
  • the pilot AGV controller and the follower AGV controller respectively inversely solve the received speed and angular velocity input to obtain the rotation speed of each wheel, and send the rotation speed command to the motor driver.
  • the connection between the leading AGV and the following AGV is the carrying workpiece;
  • the connection between the leading AGV and the carrying workpiece is installed with an angle sensor to measure the formation angle deviation between the leading AGV and the workpiece, and it is installed at the connection between the leading AGV and the carrying workpiece
  • the pilot AGV and the following AGV are respectively installed with a vertical downward visual recognition module to measure the path angle deviation and path Distance deviation:
  • Encoders are installed at the wheels of the pilot AGV and the following AGV to collect wheel speed information.
  • the present invention provides a dual AGV coordinated carrying control method, which is applied to the coordinated operation of the path tracking of the dual AGV system in the front and rear layout. Based on the front and rear arrangement of dual AGVs, combined with the visual guidance path tracking method based on the pilot-follow mode, the dual AGV systems work together to ensure the stability of automated path tracking.
  • a method based on time-domain rolling predictive control is introduced to optimize the time discretization control sequence of dual AGVs, and a more optimized dual AGV coordinated path tracking operation control is obtained.
  • the method provided by the invention has the advantages of fast correction and convergence, stable path tracking, stable formation and the like. It is a coordinated transportation control method oriented to the front and rear layout structure of dual AGVs, which further expands the application scenarios and configuration optimization of AGVs, and opens up a new idea of multi-AGV coordinated transportation.
  • Fig. 1 is a schematic diagram of the dual AGV layout structure corresponding to a dual AGV cooperative carrying control method according to the present invention.
  • Fig. 2 is a three-layer topology structure cooperative control model according to the present invention.
  • Figure 3 shows the sensor communication layer star layout communication network according to the present invention.
  • Figure 4 is a flow chart of the overall control scheme of the present invention.
  • Fig. 5 is a schematic diagram of time dispersion based on rolling prediction in time domain according to the present invention.
  • a dual AGV coordinated carrier control method consisting of two omni-directional mobile AGV front and rear layouts, based on the reference to the existing multi-agent formation control method, using a heterogeneous pilot-following method, combined with visual guidance path tracking Methods: Construct a three-layer topology structure cooperative control model. Based on path deviation and formation deviation, a dual AGV kinematics control model under path tracking and pilot-following is established, and rolling predictive control based on time domain is adopted. The discrete control model is optimized for the kinematics control model to realize the stable and reliable cooperative operation of the dual AGV system.
  • the AGV used is an omnidirectional moving AGV, and the dual AGV is a front-to-back layout.
  • the front AGV of the system is used as the pilot AGV (7), and the rear AGV is used as the following AGV ( 1), between the leading AGV and the following AGV is to carry the workpiece (3); between the leading AGV and the carrying workpiece adopts a rotating flexible connection with a rotating pair (8), and between the carrying workpiece and the following AGV adopts a rotating pair and a moving pair
  • the angle sensor is installed at the rotation flexible connection (8) of the leading AGV and the carrying workpiece to measure the formation angle deviation between the leading AGV and the workpiece, and the flexible connection between the following AGV and the carrying workpiece
  • Angle sensors and displacement sensors are installed to measure the formation angle deviation and formation distance deviation between the following AGV and the workpiece;
  • the pilot AGV body has a vertical downward visual recognition module (6) to identify the path (5), and follow the
  • the dual AGV cooperative carrying control method of the present invention includes the following steps:
  • Step 1 Collect the path distance deviation e x1 (t) of the pilot AGV at time t, the path angle deviation e ⁇ 1 (t), the formation angle deviation ⁇ 1 relative to the workpiece and the wheel speed, and the path distance deviation e of the following AGV x2 (t), path angle deviation e ⁇ 2 (t), formation distance deviation ⁇ L(t), formation angle deviation relative to the workpiece ⁇ 2 and wheel speed;
  • Step 2 According to the wheel speed of the pilot AGV and the following AGV, obtain the side shift velocity component v x1 , the forward velocity component v y1 and the angular velocity w 1 of the pilot AGV, the follow AGV side shift velocity component v x2 , the forward velocity component v y2 and Angular velocity w 2 ;
  • Step 3 Let among them, Input e x1 (t), e ⁇ 1 (t), e x2 (t), e ⁇ 2 (t), ⁇ L(t), v y1 , ⁇ 1 , ⁇ 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period expression;
  • Step 4 Estimate the input vector according to the speed state and deviation state of the pilot AGV and following the AGV at time t As well as the deviation change model, the optimization strategy of time-domain rolling predictive control is adopted to obtain the optimized input vector
  • Step 5 The input vector will be optimized It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system.
  • step 1 at time t, the path is identified by the color band placed on the ground, and the visual module on each AGV recognizes the color band of the road surface to obtain the path distance deviation of the pilot AGV relative to the path e x1 And the path angle deviation e ⁇ 1 , the path distance deviation e x2 and the path angle deviation e ⁇ 2 of the following AGV relative to the path; the formation angle deviation ⁇ 1 of the pilot AGV relative to the workpiece is obtained by the angle sensor installed at the connection between the workpiece and the pilot AGV , Obtain the formation angle deviation ⁇ 2 and formation distance deviation ⁇ L of the following AGV relative to the workpiece through the angle sensor and displacement sensor installed at the connection between the workpiece and the following AGV; through the speed code installed at the driving wheel of the pilot AGV and the following AGV The device obtains the rotation speed of each wheel of the dual AGV.
  • step 2 from the rotation speed of each wheel of the dual AGV, through the positive kinematics equation of the Mecanum wheel omnidirectional mobile robot, the pilot AGV speed v x1 , v y1 and angular velocity w 1 are obtained at the current moment, and follow the AGV speed v x2 , v y2 and angular velocity w 2 .
  • the solution model is as follows:
  • R is the radius of the Mecanum wheel
  • L is the distance from the wheel to the center of the AGV, which is 1/2 of the length of the vehicle
  • W is the distance from the center of the wheel to the center of the vehicle, which is 1/2 of the width of the vehicle .
  • step 3 the current deviation and the AGV velocity and angular velocity state are converted to the dual AGV control value in the fusion solution layer, so among them, Input e x1 (t), e ⁇ 1 (t), e x2 (t), e ⁇ 2 (t), ⁇ L(t), v y1 , ⁇ 1 , ⁇ 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period expression.
  • the kinematic control model adopted is as follows:
  • step 4 the dual AGV system corresponding to the present invention adopts the pilot AGV path tracking, follows the AGV for path tracking and is accompanied by the overall plan of formation deviation compensation in the dual AGV system, and has the following deviation change model:
  • Input vector Substitute the above-mentioned deviation change model to calculate the pilot AGV deviation e x1 (t+T) and e ⁇ 1 (t+T) at time t+T, and the following AGV deviation e x2 (t+T), e ⁇ 2 (t+T) And ⁇ L(t+T). Substitute into the terminal penalty function G; lead the AGV deviation state e x1 (t) and e ⁇ 1 (t) at time t and follow the AGV related deviation state e x2 (t), e ⁇ 2 (t) and ⁇ L(t), combined Estimate the input vector Obtain the cost function L.
  • step 4 the optimization strategy of time-domain rolling predictive control is adopted, and the optimized input vector obtained by solving the quadratic programming problem of the objective function H at time t
  • the specific steps include:
  • Step 4.1 Estimate the input vector Substituted into the deviation change model of claim 3 to predict the pilot deviation e x1 (t+T) and e ⁇ 1 (t+T) at t+T, and the following AGV deviation e x2 (t+T), e ⁇ 2 (t +T) and ⁇ L(t+T), and substituted into them to obtain the terminal penalty function G.
  • Step 4.2 Lead the AGV deviation state e x1 (t) and e ⁇ 1 (t) from time t and follow the AGV related deviation state e x2 (t), e ⁇ 2 (t) and ⁇ L(t), combined with the estimated input vector Obtain the cost function L.
  • Step 4.5 When the dual AGV system runs to time t+ ⁇ , update time t to t+ ⁇ , and repeat steps 1 to 5.
  • the purpose of rolling and optimizing the input vector is to make the penalty function G(e i (t+T)) and the cost
  • the objective function H of the sum of the function L(e i (t), u i (t)) is taken to the minimum value, and the input e i (t+T) of the penalty function G(e i (t+T)) is determined by the prediction stage Obtained, the penalty function G should be a continuously differentiable positive definite function.
  • the form of the terminal penalty function can be selected as follows:
  • e 1 (t) [e x1 (t) e ⁇ 1 (t)] T
  • u 1 (t) [v x1 (t) w ⁇ 1 (t)] T
  • the weight matrix Q and R are Symmetric positive semi-definite matrix.
  • the expression of the input vector is again obtained via the feedback control coefficients k 1 , k 2 , and the deviation vector e 1 (t) and speed v y1 (t) of the system at time t. Therefore, the final optimization goal becomes to find the optimal solution that can obtain the minimum value in the function H of the coefficients k 1 and k 2. Finally, substitute the solved coefficients k 1 and k 2 into the input vector And used for the control in the (t, t+ ⁇ ) time period.
  • the above content is the solution process of the control input of the pilot AGV and the follower AGV based on the rolling time domain predictive control in the dual AGV cooperative delivery system.
  • step 5 at the motion control layer, the input vector will be optimized It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system.
  • the pilot AGV and the controller following the AGV receive the motion control value from the data solution layer, and the omnidirectional AGV inverse kinematics equation is used to solve the rotation speed output of each wheel of the pilot AGV and the following AGV.
  • the driver drives the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated carrying operation of the dual AGV cooperative system.
  • the real input vector of the speed and angular velocity of the pilot AGV obtained by the aforementioned data solution layer Converted into the rotational speed input of each wheel, and sent to the driver through the controller to drive the motor of the pilot AGV; the real input vector of the following AGV speed and angular velocity obtained by the aforementioned data solution layer It is converted into the rotational speed input of each wheel and sent to the driver through the controller to drive the motor following the AGV to run.
  • the conversion model is as follows:
  • R is the radius of the Mecanum wheel
  • L is the distance from the wheel to the center of the AGV, which is 1/2 of the length of the vehicle
  • W is the distance from the center of the wheel to the center of the vehicle, which is 1/2 of the width of the vehicle
  • It is the speed component in the y direction (that is, the forward direction) of the pilot AGV, which is set to a fixed value when the system is running.
  • the present invention designs a three-layer topology structure model.
  • the three-layer topology collaborative control model of the present invention will be described.
  • sensors and communication modules constitute the sensor communication layer.
  • the vision module formed by the CCD camera and the DSP processor on each AGV is used to identify the path and obtain the path angle deviation and path distance deviation of the pilot AGV and the following AGV, respectively.
  • the encoder obtains the wheel speed of the pilot AGV and the following AGV.
  • Each AGV has a total of four encoders, which are respectively arranged at the output shaft of the motor that drives the four mecanum wheels.
  • the formation angle deviation between the leading AGV and the workpiece and the formation angle deviation between the following AGV and the workpiece are obtained by the angle sensor, and the formation distance deviation is obtained by the displacement sensor.
  • the information transmission network inside the dual AGV cooperative delivery system is constructed by wireless communication modules and serial communication methods.
  • the wireless communication module of Linghang AGV is used as the server, and the wireless communication module of the AGV, the upper computer and the integrated solution center is used as the client to access, forming a star-shaped communication network.
  • the data fusion solution center constitutes the data fusion processing layer, and the navigation AGV path deviation (distance, angle), formation angle deviation, and navigation AGV speed and angular velocity information are transmitted from the communication layer to the fusion solution center , Follow the AGV path deviation (distance, angle), formation angle deviation, formation distance deviation, and follow the AGV speed, angular velocity information is collected by the sensor and received into the fusion solution center, data fusion processing to obtain the leading AGV and the following AGV Speed and angular velocity input.
  • the motion control layer is composed of the pilot AGV and the controller, motor, driver and other actuators that follow the AGV.
  • the velocity and angular velocity input of each AGV obtained by the solution is obtained by the inverse kinematics of the mecanum wheel omnidirectional AGV to obtain the pilot AGV and follow
  • the rotation speed of each wheel of the AGV is sent through the driver to control the motor drive.
  • the interactive information in the dual AGV system mainly includes:
  • the information sent by the pilot AGV controller to the host computer system is the operating status information of the dual AGV system, including: the operating speed, obstacle avoidance information, and operating status of the dual AGV system.
  • the host computer system sends instruction information and task information to the pilot AGV controller, which is used to control the start and stop actions of the entire dual AGV system and task reception and execution;
  • the following AGV sends the current running speed and angular velocity to the fusion settlement center through the serial bus; the fusion settlement center sends the calculated following AGV speed and angular velocity control to the following AGV.
  • the path angle deviation, path distance deviation, formation angle deviation, formation distance deviation and other information of each AGV are used to fuse the movement speed and angular velocity of the pilot AGV and the following AGV at the current moment.
  • the Fusion Solution Center obtains the speed and angular velocity control of the dual AGV through the kinematics model; introduces the time-domain rolling optimization predictive control to optimize the time discrete control of the dual AGV kinematics, and obtains the discrete moment motion control input and sends it to the motion control
  • the pilot AGV controller and the follower AGV controller of the layer are used to fuse the movement speed and angular velocity of the pilot AGV and the following AGV at the current moment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

Provided are a dual-AGV collaborative carrying control system and method, which relate to the field of smart logistics carrying robots. The method is mainly for meeting the conveying requirements of large components, specifically of objects which have relatively large dimensions in the length direction, and is a combination of a path tracking method and a navigator-following method. By using a collaborative planning model that has a three-layer topological structure, the formation deviation under a dual-AGV navigator-following strategy is combined with the path deviation under path tracking to obtain a navigator-following and path tracking-based kinematic control model, and a time-domain rolling predictive control-based discrete control model is used to optimize the kinematic control model, thereby achieving the stable and reliable path tracking and collaborative carrying of the dual-AGV system.

Description

一种双AGV协同运载控制系统及方法A Double AGV Cooperative Carrier Control System and Method 技术领域Technical field
本发明涉及一种面向双AGV协同运载系统的控制系统及方法，尤其涉及一种针对大型物件输送的双AGV协同运载控制方法，属于智能化工业机器人领域。The invention relates to a control system and a method for a dual AGV cooperative carrying system, in particular to a dual AGV cooperative carrying control method for the conveying of large objects, and belongs to the field of intelligent industrial robots.
背景技术Background technique
随着工业自动化水平的提高，传统采用的长距离AGV自动化物流方式已逐渐无法满足生产制造、物流转运派送等工作。在大型装备的总装，以及码头集装箱运载等场景中，目前仍大多采用龙门吊、塔式起重机等传统输送系统进行大型物件的运送，存在着占用空间大，能耗高，系统可拓展性有限等诸多局限性。如位于上海的C919总装车间，在进行机翼机身等的对接工作时，仍采用位于车间上部的轨道起重机进行机身段及部装机翼的输送和对接。而位于上海洋山港及厦门远海自动化码头的集装箱自动化运送，则采用了增大单台AGV体型的方法实现大型集装箱的自动化物流。With the improvement of the industrial automation level, the traditional long-distance AGV automated logistics method has gradually been unable to meet the tasks of manufacturing, logistics transfer and delivery. In the final assembly of large-scale equipment, as well as terminal container transportation, most of the traditional transportation systems such as gantry cranes and tower cranes are still used for the transportation of large objects. There are many reasons such as large space occupation, high energy consumption, and limited system scalability. limitation. For example, the C919 assembly workshop in Shanghai still uses the rail crane located in the upper part of the workshop to transport and dock the fuselage section and the partial wing during the docking work of the wing and fuselage. The automated transportation of containers located in Shanghai Yangshan Port and Xiamen Yuanhai Automated Terminals uses the method of increasing the size of a single AGV to realize the automated logistics of large containers.
采用多个AGV组成多移动机器人系统，使之在自动化运行的同时，保持固定队形，由此便具备了单个AGV所没有的运载能力。基于多智能体协作的模式，采用多AGV进行物流协同搬运的模式，有着以下优点：1.便于对AGV的管理；2.提高物流运送系统稳定性和鲁棒性；3.拓展AGV的利用领域，使其运载方式更加多样化；4.提高AGV的运载利用率；5.系统具备较高的安全性、鲁棒性及灵活性。多AGV在运行过程中进行协调同步运作和队形保持，实现物料等的多AGV协同搬运，将大大改善目前大体积、大重量物料自动化输送的应用现状，同时进一步提高AGV的利用率，优化物流资源的配置，具备广阔的应用场景。Multiple AGVs are used to form a multi-mobile robot system, so that it maintains a fixed formation while operating automatically, thereby providing a carrying capacity that a single AGV does not have. Based on the model of multi-agent collaboration, the use of multiple AGVs for coordinated logistics handling has the following advantages: 1. It is convenient to manage AGVs; 2. It improves the stability and robustness of the logistics transportation system; 3. Expands the use of AGVs , To make its transportation methods more diversified; 4. Improve the carrying utilization rate of AGV; 5. The system has higher safety, robustness and flexibility. Multiple AGVs carry out coordinated and synchronized operation and formation maintenance during the operation process to realize the coordinated handling of materials and other multiple AGVs, which will greatly improve the current application status of automated transportation of large-volume and heavy-weight materials, while further increasing the utilization rate of AGVs and optimizing logistics The allocation of resources has a wide range of application scenarios.
当下，对于多机器人队形控制方法，国内外学者提出了基于行为法、人工势场法、虚拟结构保持法、分布式控制法、循环法、领航-跟随法等诸多理论方法。其中，领航-跟随法因可靠性高，扩展性好，成为被广泛采用的一种多机器人队形控制及队形保持方法。基于领航-跟随的方法，国内外众多专家学者对于多机器人系统的导航构建及运行控制领域进行了全方位的拓展和研究。然而，应用于工业现场的多机器人技术需要针对实际设计。目前，设计可用于工业现场多AGV协同搬运系统的可实现方案仍然较少，对于多AGV协同搬运技术的应用仍不完善。公开号为CN10418899A的中国发明专利提出一种双移动机器人柔性连接的协同搬运系统及其复合导航装置，但在该专利中仅涉及了双 移动机器人的机械结构，并不涉及控制方法方面的描述。而区别于公开号为CN109062150A的中国发明专利提出的一种基于AGV的自动搬运系统及其多AGV协作方法中提及的多AGV协作方法，该种协作方法是在单AGV运行系统基础上形成的AGV运行任务规划、调度管理，单一时刻控制主体依旧为单台AGV，而本发明设计目的为双AGV协同运载控制方法，在同一时刻的控制主体为双AGV，以此来形成双AGV对同一物体的协作同步运载。At present, for multi-robot formation control methods, domestic and foreign scholars have proposed many theoretical methods based on behavior method, artificial potential field method, virtual structure maintenance method, distributed control method, loop method, pilot-follow method and so on. Among them, the pilot-following method has become a widely used multi-robot formation control and formation maintenance method due to its high reliability and good scalability. Based on the pilot-follow method, many experts and scholars at home and abroad have carried out all-round expansion and research in the field of navigation construction and operation control of multi-robot systems. However, the multi-robot technology applied to the industrial field needs to be designed for actual use. At present, there are still few achievable solutions designed for multi-AGV cooperative handling systems in industrial sites, and the application of multi-AGV cooperative handling technology is still incomplete. The Chinese invention patent with publication number CN10418899A proposes a flexibly connected dual-mobile robot cooperative handling system and its composite navigation device. However, the patent only relates to the mechanical structure of the dual-mobile robot, and does not involve the description of the control method. It is different from the multi-AGV cooperation method mentioned in an AGV-based automatic handling system and its multi-AGV cooperation method proposed by the Chinese invention patent with the publication number CN109062150A. This cooperation method is formed on the basis of a single AGV operation system. AGV operation task planning and scheduling management, the control subject at a single moment is still a single AGV, and the design purpose of the present invention is a dual AGV coordinated carrying control method. The control subject at the same time is a dual AGV, so as to form a dual AGV for the same object The collaboration is carried out simultaneously.
因此，综合上述对多移动机器人应用领域的发展现状及其广阔的应用价值和实用前景，本发明设计人以路径跟踪为基础，结合领航-跟随的多智能体基本架构，提出一种面向双AGV协同运载的控制方法，进一步拓展AGV在产业上的利用价值。Therefore, based on the above-mentioned development status of the application field of multi-mobile robots and its broad application value and practical prospects, the designer of the present invention proposes a dual-AGV oriented based on the path tracking and combined with the pilot-following multi-agent basic architecture The control method of coordinated transportation further expands the use value of AGV in the industry.
发明内容Summary of the invention
本发明为解决现有技术问题，针对大部件自动化运送的场景，面向双AGV柔性连接前后布局的方式，参考领航-跟随策略，提出了一种双AGV协同运载的控制方法，用于双AGV协同运载系统队形保持及自动化运行的实现。In order to solve the problems of the prior art, aiming at the scene of automatic transportation of large components, and oriented to the flexible connection of dual AGVs in the front and rear layout mode, referring to the pilot-follow strategy, the present invention proposes a dual AGV cooperative carrying control method, which is used for dual AGV coordination The formation of the transportation system is maintained and the realization of automatic operation.
本发明构建了一种双AGV协同运载控制方法，采用了两台全向移动AGV，结合路径跟踪方法及领航-跟随方法，以三层拓扑结构的协同控制模型，建立一种在路径跟踪下的双AGV协同运载系统的运动学控制模型，并采用了基于时域滚动预测控制的离散控制模型对该运动学模型进行运动控制优化，实现双AGV系统稳定可靠的协同运行。The present invention constructs a dual AGV cooperative carrier control method, adopts two omnidirectional mobile AGVs, combines the path tracking method and the pilot-follow method, and establishes a cooperative control model under the path tracking with a three-layer topology structure The kinematics control model of the dual AGV cooperative delivery system, and the discrete control model based on time-domain rolling predictive control is used to optimize the kinematics model to realize the stable and reliable cooperative operation of the dual AGV system.
在本发明中，一种双AGV协同运载控制方法，双AGV为全向移动，前后布局，前者为领航AGV，后者为跟随AGV，该方法包括如下步骤：In the present invention, a dual AGV coordinated carrying control method, the dual AGV is omnidirectional movement, the front and rear layout, the former is the pilot AGV, the latter is the follower AGV, the method includes the following steps:
步骤1：采集t时刻的领航AGV的路径距离偏差e x1(t)、路径角度偏差e θ1(t)、相对于工件的队形角度偏差α 1和车轮转速，以及跟随AGV的路径距离偏差e x2(t)、路径角度偏差e θ2(t)、队形距离偏差ΔL(t)、相对于工件的队形角度偏差α 2和车轮转速； Step 1: Collect the path distance deviation e x1 (t) of the pilot AGV at time t, the path angle deviation e θ1 (t), the formation angle deviation α 1 relative to the workpiece and the wheel speed, and the path distance deviation e of the following AGV x2 (t), path angle deviation e θ2 (t), formation distance deviation ΔL(t), formation angle deviation relative to the workpiece α 2 and wheel speed;
步骤2：根据领航AGV和跟随AGV的车轮转速，求得领航AGV的侧移速度分量v x1、前进速度分量v y1及角速度w 1，跟随AGV侧移速度分量v x2、前进速度分量v y2及角速度w 2； Step 2: According to the wheel speed of the pilot AGV and the following AGV, obtain the side shift velocity component v x1 , the forward velocity component v y1 and the angular velocity w 1 of the pilot AGV, the follow AGV side shift velocity component v x2 , the forward velocity component v y2 and Angular velocity w 2 ;
步骤3：令
Figure PCTCN2020122756-appb-000001
其中，
Figure PCTCN2020122756-appb-000002
将e x1(t)、e θ1(t)、e x2(t)、e θ2(t)、ΔL(t)、v y1、α 1、α 2输入运动学控制模型，得到时域t到t+T时段内的估计输入向量
Figure PCTCN2020122756-appb-000003
表达式； Step 3: Let
Figure PCTCN2020122756-appb-000001
among them,
Figure PCTCN2020122756-appb-000002
Input e x1 (t), e θ1 (t), e x2 (t), e θ2 (t), ΔL(t), v y1 , α 1 , α 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period
Figure PCTCN2020122756-appb-000003
expression;
步骤4：根据t时刻领航AGV及跟随AGV的速度状态和偏差状态、估计输入向量
Figure PCTCN2020122756-appb-000004
以及偏差变化模型，采用时域滚动预测控制的优化策略，获得优化输入向量
Figure PCTCN2020122756-appb-000005
Step 4: Estimate the input vector according to the speed state and deviation state of the pilot AGV and following the AGV at time t
Figure PCTCN2020122756-appb-000004
As well as the deviation change model, the optimization strategy of time-domain rolling predictive control is adopted to obtain the optimized input vector
Figure PCTCN2020122756-appb-000005
步骤5：将优化输入向量
Figure PCTCN2020122756-appb-000006
解算成领航AGV及跟随AGV各轮的转速输出量，并发往驱动器驱动领航AGV及跟随AGV的电机运转，从而进行双AGV协同系统的路径跟踪协同运载运行。 Step 5: The input vector will be optimized
Figure PCTCN2020122756-appb-000006
It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system.
进一步地，步骤3中所述，用于获得估计输入向量
Figure PCTCN2020122756-appb-000007
的运动学控制模型为： Further, as described in step 3, used to obtain the estimated input vector
Figure PCTCN2020122756-appb-000007
The kinematics control model is:
Figure PCTCN2020122756-appb-000008
Figure PCTCN2020122756-appb-000008
Figure PCTCN2020122756-appb-000009
Figure PCTCN2020122756-appb-000009
进一步地，步骤4中所述的偏差变化模型具有如下形式：Further, the deviation change model described in step 4 has the following form:
Figure PCTCN2020122756-appb-000010
Figure PCTCN2020122756-appb-000010
Figure PCTCN2020122756-appb-000011
Figure PCTCN2020122756-appb-000011
ΔL(t+1)＝ΔL(t)+T(v y1cosα 1-v x1sinα 1+v x2sinα 2+v y2cosα 2)。 ΔL(t+1)=ΔL(t)+T(v y1 cosα 1 -v x1 sinα 1 +v x2 sinα 2 +v y2 cosα 2 ).
进一步地，步骤6中描述的领航AGV的终端罚函数的形式如下：Further, the form of the terminal penalty function of the pilot AGV described in step 6 is as follows:
Figure PCTCN2020122756-appb-000012
Figure PCTCN2020122756-appb-000012
代价函数的形式如下：The form of the cost function is as follows:
L(e 1(t),u 1(t))＝e 1(t) TQe 1(t)+u 1(t) TRu 1(t) L(e 1 (t),u 1 (t))=e 1 (t) T Qe 1 (t)+u 1 (t) T Ru 1 (t)
其中，e 1(t)＝[e x1(t) e θ1(t)] T，u 1(t)＝[v x1(t) w θ1(t)] T，权值矩阵Q与R则为半正定的对称矩阵；设置代价函数权值矩阵Q和R如下： Among them, e 1 (t) = [e x1 (t) e θ1 (t)] T , u 1 (t) = [v x1 (t) w θ1 (t)] T , the weight matrix Q and R are Positive semi-definite symmetric matrix; set the cost function weight matrix Q and R as follows:
Figure PCTCN2020122756-appb-000013
且
Figure PCTCN2020122756-appb-000014
Figure PCTCN2020122756-appb-000013
And
Figure PCTCN2020122756-appb-000014
则，then,
L(e 1(t),u 1(t))＝q x1e x1(t) 2+q θ1e θ1(t) 2+r x1v x1(t) 2+r θ1ω 1(t) 2 L(e 1 (t),u 1 (t))=q x1 e x1 (t) 2 +q θ1 e θ1 (t) 2 +r x1 v x1 (t) 2 +r θ1 ω 1 (t) 2
跟随AGV模型的终端罚函数的形式如下：The form of the terminal penalty function following the AGV model is as follows:
Figure PCTCN2020122756-appb-000015
Figure PCTCN2020122756-appb-000015
代价函数的形式如下：The form of the cost function is as follows:
L(e 2(t),u 2(t),ΔL(t))＝q x2e x1(t) 2+q θ2e θ1(t) 2+q 3ΔL(t) 2+r x2v x2(t) 2+r θ2ω 2(t) 2+r y2v y2(t) 2 L(e 2 (t),u 2 (t),ΔL(t))=q x2 e x1 (t) 2 +q θ2 e θ1 (t) 2 +q 3 ΔL(t) 2 +r x2 v x2 (t) 2 +r θ2 ω 2 (t) 2 +r y2 v y2 (t) 2
进一步地，在步骤4中，采用时域滚动预测控制的优化策略，于t时刻将求解目标函数H的二次规划问题获得的优化输入向量
Figure PCTCN2020122756-appb-000016
作为当前时域(t，t+δ)内的控制量对系统进行控制，具体步骤包括： Further, in step 4, the optimization strategy of time-domain rolling predictive control is adopted, and the optimized input vector obtained by solving the quadratic programming problem of the objective function H at time t
Figure PCTCN2020122756-appb-000016
As the control variable in the current time domain (t, t+δ) to control the system, the specific steps include:
步骤4.1：估计输入向量
Figure PCTCN2020122756-appb-000017
代入权利要求3所述的偏差变化模型预测t+T时刻的领航偏差e x1(t+T)及e θ1(t+T)，和跟随AGV偏差e x2(t+T)、e θ2(t+T)及ΔL(t+T)，并代入获得终端罚函数G。 Step 4.1: Estimate the input vector
Figure PCTCN2020122756-appb-000017
Substituted into the deviation change model of claim 3 to predict the pilot deviation e x1 (t+T) and e θ1 (t+T) at t+T, and the following AGV deviation e x2 (t+T), e θ2 (t +T) and ΔL(t+T), and substituted into them to obtain the terminal penalty function G.
步骤4.2：由t时刻领航AGV偏差状态e x1(t)及e θ1(t)和跟随AGV相关偏差状态e x2(t)、e θ2(t)及ΔL(t)，结合估计输入向量
Figure PCTCN2020122756-appb-000018
获得代价函数L。 Step 4.2: Lead the AGV deviation state e x1 (t) and e θ1 (t) from time t and follow the AGV related deviation state e x2 (t), e θ2 (t) and ΔL(t), combined with the estimated input vector
Figure PCTCN2020122756-appb-000018
Obtain the cost function L.
步骤4.3：设置目标函数H＝G+L，求解H关于参数k i的二次规划问题，将求解获得的k i代入获得领航AGV与跟随AGV的优化输入向量
Figure PCTCN2020122756-appb-000019
步骤4.4：在控制时段(t,t+δ)内，令真实输入向量
Figure PCTCN2020122756-appb-000020
其中0<δ≤T； Step 4.3: Set the objective function H=G+L, solve the quadratic programming problem of H with respect to the parameter k i , and substitute the obtained k i into the optimized input vector of the pilot AGV and the following AGV
Figure PCTCN2020122756-appb-000019
Step 4.4: In the control period (t, t+δ), let the real input vector
Figure PCTCN2020122756-appb-000020
Where 0<δ≤T;
步骤4.5：当双AGV系统运行到t+δ时刻后，将时刻t更新为t+δ，重复步骤1到步骤5。Step 4.5: When the dual AGV system runs to time t+δ, update time t to t+δ, and repeat steps 1 to 5.
基于上述方法的双AGV协同运载控制系统，包括传感通讯层、数据融合处理层和运动控制层；所述运动控制层包括领航AGV控制器和跟随AGV控制器，领航AGV控制器和跟随AGV控制器分别连接领航AGV和跟随AGV的车轮电机驱动器；所述传感通讯层用于监测双AGV的路径偏差、队形偏差和车轮转速，并将所监测信息传送给数据融合处理层，数据融合处理层对信息进行融合解算处理，得到领航AGV及跟随AGV的速度及角速度输入量，并将领航AGV及跟随AGV的速度及角速度输入量分别发送给领航AGV控制器和跟随AGV控制器；所述领航AGV控制器和跟随AGV控制器分别对所接收的速度及角速度输入量进行反解，得到各车轮转速，并将转速命令发送至电机驱动器。The dual AGV cooperative carrier control system based on the above method includes a sensor communication layer, a data fusion processing layer and a motion control layer; the motion control layer includes a pilot AGV controller and a follower AGV controller, a pilot AGV controller and a follower AGV control The device is respectively connected to the wheel motor drivers of the pilot AGV and the following AGV; the sensor communication layer is used to monitor the path deviation, formation deviation and wheel speed of the dual AGV, and transmit the monitored information to the data fusion processing layer for data fusion processing The layer performs fusion and processing of the information to obtain the speed and angular velocity input of the pilot AGV and the following AGV, and send the speed and angular velocity input of the pilot AGV and the following AGV to the pilot AGV and the following AGV controller respectively; The pilot AGV controller and the follower AGV controller respectively inversely solve the received speed and angular velocity input to obtain the rotation speed of each wheel, and send the rotation speed command to the motor driver.
进一步地，领航AGV与跟随AGV之间为运载工件；领航AGV与运载工件的连接处安装有角度传感器用于测量领航AGV与工件之间的队形角度偏差，跟随AGV与运载工件的连接处安装有角度传感器与位移传感器测量跟随AGV与工件之间的队形角度偏差与队形距离偏差；领航AGV和跟随AGV车身中部分别安装有垂直向下的视觉识 别模块，用于测量路径角度偏差及路径距离偏差；领航AGV和跟随AGV车轮处分别安装有编码器采集车轮转速信息。Further, between the leading AGV and the following AGV is the carrying workpiece; the connection between the leading AGV and the carrying workpiece is installed with an angle sensor to measure the formation angle deviation between the leading AGV and the workpiece, and it is installed at the connection between the leading AGV and the carrying workpiece There are angle sensors and displacement sensors to measure the formation angle deviation and the formation distance deviation between the following AGV and the workpiece; the pilot AGV and the following AGV are respectively installed with a vertical downward visual recognition module to measure the path angle deviation and path Distance deviation: Encoders are installed at the wheels of the pilot AGV and the following AGV to collect wheel speed information.
有益效果：本发明提供了一种双AGV协同运载控制方法，应用于前后布局的双AGV系统进行路径跟踪的协同运作场合。以双AGV前后安置为基本布局，在领航-跟随模式的基础上结合视觉导引路径跟踪的方式，双AGV系统协同运作的同时保证了自动化路径跟踪的稳定性。引入基于时域滚动预测控制的方法，对双AGV的时间离散化控制序列进行优化，得到了更优化的双AGV协同路径跟踪运行的控制。本发明提供的方法，具有纠偏收敛快，路径跟踪平稳，队形保持稳定等优点。是为一种面向双AGV前后布局结构的协同运载控制方法，进一步拓展了AGV的应用场合和配置优化，打开了多AGV协同运载方式的新思路。Beneficial effects: The present invention provides a dual AGV coordinated carrying control method, which is applied to the coordinated operation of the path tracking of the dual AGV system in the front and rear layout. Based on the front and rear arrangement of dual AGVs, combined with the visual guidance path tracking method based on the pilot-follow mode, the dual AGV systems work together to ensure the stability of automated path tracking. A method based on time-domain rolling predictive control is introduced to optimize the time discretization control sequence of dual AGVs, and a more optimized dual AGV coordinated path tracking operation control is obtained. The method provided by the invention has the advantages of fast correction and convergence, stable path tracking, stable formation and the like. It is a coordinated transportation control method oriented to the front and rear layout structure of dual AGVs, which further expands the application scenarios and configuration optimization of AGVs, and opens up a new idea of multi-AGV coordinated transportation.
附图说明Description of the drawings
图1为本发明所述的一种双AGV协同运载控制方法对应的双AGV布局结构简图。Fig. 1 is a schematic diagram of the dual AGV layout structure corresponding to a dual AGV cooperative carrying control method according to the present invention.
图2为本发明所述的三层拓扑结构协同控制模型。Fig. 2 is a three-layer topology structure cooperative control model according to the present invention.
图3为本发明所述的传感通讯层星形布局通讯网络。Figure 3 shows the sensor communication layer star layout communication network according to the present invention.
图4为本发明所述的总体控制方案流程图。Figure 4 is a flow chart of the overall control scheme of the present invention.
图5为本发明所述的基于时域滚动预测时间离散示意图。Fig. 5 is a schematic diagram of time dispersion based on rolling prediction in time domain according to the present invention.
具体实施方式Detailed ways
下面结合附图对本发明的具体实施方式作进一步的说明。The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings.
一种双AGV协同运载控制方法，由两台全向移动AGV前后布局组成，在参考现有的多智能体编队控制方法的基础上，采用异构的领航-跟随方法，结合视觉导引路径跟踪方法，构建三层拓扑结构的协同控制模型，在路径偏差和队形偏差的基础上建立一种在路径跟踪及领航-跟随下的双AGV运动学控制模型，并采用了基于时域滚动预测控制的离散控制模型对该运动学控制模型进行优化，实现双AGV系统稳定可靠的协同运行。A dual AGV coordinated carrier control method, consisting of two omni-directional mobile AGV front and rear layouts, based on the reference to the existing multi-agent formation control method, using a heterogeneous pilot-following method, combined with visual guidance path tracking Methods: Construct a three-layer topology structure cooperative control model. Based on path deviation and formation deviation, a dual AGV kinematics control model under path tracking and pilot-following is established, and rolling predictive control based on time domain is adopted. The discrete control model is optimized for the kinematics control model to realize the stable and reliable cooperative operation of the dual AGV system.
结合附图1，本发明所述的双AGV协同运载控制方法中，采用的AGV为全向移动AGV，双AGV为前后布局，以系统前AGV作为领航AGV(7)，后AGV作为跟随AGV(1)，领航AGV与跟随AGV之间为运载工件(3)；领航AGV与运载工件之间采用具备转动副的转动柔性连接(8)，运载工件与跟随AGV之间采用具备转动副和移动副的组合柔性连接(4)；在领航AGV与运载工件的转动柔性连接(8)处安装有角度传感器用于测量领航AGV与工件之间的队形角度偏差，在跟随AGV与运载工件组合柔性连 接处安装有角度传感器与位移传感器测量跟随AGV与工件之间的队形角度偏差与队形距离偏差；领航AGV车身中部有垂直向下的视觉识别模块(6)识别路径(5)，跟随AGV车身中部安装有垂直向下的视觉识别模块(4)识别路径；领航AGV车轮处安装有编码器(9)采集车轮转速信息，跟随AGV车轮处安装有编码器(10)采集车轮转速信息；领航AGV的主要功能为进行路径跟踪前进以及与上位机系统交换指令信息的功能，跟随AGV在路径跟踪前进的过程中根据传感器的数据及领航AGV反馈的数据进行实时调整控制。With reference to Figure 1, in the dual AGV cooperative carrier control method of the present invention, the AGV used is an omnidirectional moving AGV, and the dual AGV is a front-to-back layout. The front AGV of the system is used as the pilot AGV (7), and the rear AGV is used as the following AGV ( 1), between the leading AGV and the following AGV is to carry the workpiece (3); between the leading AGV and the carrying workpiece adopts a rotating flexible connection with a rotating pair (8), and between the carrying workpiece and the following AGV adopts a rotating pair and a moving pair The combined flexible connection (4); the angle sensor is installed at the rotation flexible connection (8) of the leading AGV and the carrying workpiece to measure the formation angle deviation between the leading AGV and the workpiece, and the flexible connection between the following AGV and the carrying workpiece Angle sensors and displacement sensors are installed to measure the formation angle deviation and formation distance deviation between the following AGV and the workpiece; the pilot AGV body has a vertical downward visual recognition module (6) to identify the path (5), and follow the AGV body A vertical downward visual recognition module (4) is installed in the middle to identify the path; an encoder (9) is installed at the wheels of the pilot AGV to collect wheel speed information, and an encoder (10) is installed at the wheels of the following AGV to collect wheel speed information; Its main function is to carry out the path tracking forward and the function of exchanging command information with the host computer system, and follow the AGV to carry out real-time adjustment and control according to the sensor data and the data feedback from the pilot AGV during the path tracking progress.
本发明所述的双AGV协同运载控制方法，包括下述步骤：The dual AGV cooperative carrying control method of the present invention includes the following steps:
步骤1：采集t时刻的领航AGV的路径距离偏差e x1(t)、路径角度偏差e θ1(t)、相对于工件的队形角度偏差α 1和车轮转速，以及跟随AGV的路径距离偏差e x2(t)、路径角度偏差e θ2(t)、队形距离偏差ΔL(t)、相对于工件的队形角度偏差α 2和车轮转速； Step 1: Collect the path distance deviation e x1 (t) of the pilot AGV at time t, the path angle deviation e θ1 (t), the formation angle deviation α 1 relative to the workpiece and the wheel speed, and the path distance deviation e of the following AGV x2 (t), path angle deviation e θ2 (t), formation distance deviation ΔL(t), formation angle deviation relative to the workpiece α 2 and wheel speed;
步骤2：根据领航AGV和跟随AGV的车轮转速，求得领航AGV的侧移速度分量v x1、前进速度分量v y1及角速度w 1，跟随AGV侧移速度分量v x2、前进速度分量v y2及角速度w 2； Step 2: According to the wheel speed of the pilot AGV and the following AGV, obtain the side shift velocity component v x1 , the forward velocity component v y1 and the angular velocity w 1 of the pilot AGV, the follow AGV side shift velocity component v x2 , the forward velocity component v y2 and Angular velocity w 2 ;
步骤3：令
Figure PCTCN2020122756-appb-000021
其中，
Figure PCTCN2020122756-appb-000022
将e x1(t)、e θ1(t)、e x2(t)、e θ2(t)、ΔL(t)、v y1、α 1、α 2输入运动学控制模型，得到时域t到t+T时段内的估计输入向量
Figure PCTCN2020122756-appb-000023
表达式； Step 3: Let
Figure PCTCN2020122756-appb-000021
among them,
Figure PCTCN2020122756-appb-000022
Input e x1 (t), e θ1 (t), e x2 (t), e θ2 (t), ΔL(t), v y1 , α 1 , α 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period
Figure PCTCN2020122756-appb-000023
expression;
步骤4：根据t时刻领航AGV及跟随AGV的速度状态和偏差状态、估计输入向量
Figure PCTCN2020122756-appb-000024
以及偏差变化模型，采用时域滚动预测控制的优化策略，获得优化输入向量
Figure PCTCN2020122756-appb-000025
Step 4: Estimate the input vector according to the speed state and deviation state of the pilot AGV and following the AGV at time t
Figure PCTCN2020122756-appb-000024
As well as the deviation change model, the optimization strategy of time-domain rolling predictive control is adopted to obtain the optimized input vector
Figure PCTCN2020122756-appb-000025
步骤5：将优化输入向量
Figure PCTCN2020122756-appb-000026
解算成领航AGV及跟随AGV各轮的转速输出量，并发往驱动器驱动领航AGV及跟随AGV的电机运转，从而进行双AGV协同系统的路径跟踪协同运载运行。 Step 5: The input vector will be optimized
Figure PCTCN2020122756-appb-000026
It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system.
结合附图各步骤展开详细叙述。Detailed description of each step in conjunction with the accompanying drawings.
结合附图4总体方案控制流程图，在步骤1中，t时刻，通过地面安置的色带标识路径，各AGV上的视觉模块识别路面的色带获得领航AGV相对于路径的路径距离偏差e x1和路径角度偏差e θ1，跟随AGV相对于路径的路径距离偏差e x2和路径角度偏差e θ2；通过安装于工件与领航AGV连接处的角度传感器获得领航AGV相对于工件的队形角度偏差α 1，通过安装于工件与跟随AGV连接处的角度传感器与位移传感器获得跟随AGV相对于工件的队形角度偏差α 2和队形距离偏差ΔL；通过安装于领航AGV及跟随AGV 驱动轮处的速度编码器获得双AGV各轮的转速。 With reference to the control flow chart of the overall scheme of Figure 4, in step 1, at time t, the path is identified by the color band placed on the ground, and the visual module on each AGV recognizes the color band of the road surface to obtain the path distance deviation of the pilot AGV relative to the path e x1 And the path angle deviation e θ1 , the path distance deviation e x2 and the path angle deviation e θ2 of the following AGV relative to the path; the formation angle deviation α 1 of the pilot AGV relative to the workpiece is obtained by the angle sensor installed at the connection between the workpiece and the pilot AGV , Obtain the formation angle deviation α 2 and formation distance deviation ΔL of the following AGV relative to the workpiece through the angle sensor and displacement sensor installed at the connection between the workpiece and the following AGV; through the speed code installed at the driving wheel of the pilot AGV and the following AGV The device obtains the rotation speed of each wheel of the dual AGV.
在步骤2中，由双AGV各轮的转速，通过麦克纳姆轮全向移动机器人的正运动学方程获得当前时刻领航AGV速度v x1、v y1及角速度w 1，跟随AGV速度v x2、v y2及角速度w 2。求解模型如下： In step 2, from the rotation speed of each wheel of the dual AGV, through the positive kinematics equation of the Mecanum wheel omnidirectional mobile robot, the pilot AGV speed v x1 , v y1 and angular velocity w 1 are obtained at the current moment, and follow the AGV speed v x2 , v y2 and angular velocity w 2 . The solution model is as follows:
Figure PCTCN2020122756-appb-000027
Figure PCTCN2020122756-appb-000027
R为麦克纳姆轮的半径；L为轮子到AGV中心车长方向的距离，即为车长的1/2；W为轮子车体中心车宽方向的距离，即为车宽的1/2。R is the radius of the Mecanum wheel; L is the distance from the wheel to the center of the AGV, which is 1/2 of the length of the vehicle; W is the distance from the center of the wheel to the center of the vehicle, which is 1/2 of the width of the vehicle .
在步骤3中，于融合解算层进行当前偏差及AGV速度角速度状态到双AGV控制量的转化，令
Figure PCTCN2020122756-appb-000028
其中，
Figure PCTCN2020122756-appb-000029
将e x1(t)、e θ1(t)、e x2(t)、e θ2(t)、ΔL(t)、v y1、α 1、α 2输入运动学控制模型，得到时域t到t+T时段内的估计输入向量
Figure PCTCN2020122756-appb-000030
表达式。采用的运动学控制模型如下： In step 3, the current deviation and the AGV velocity and angular velocity state are converted to the dual AGV control value in the fusion solution layer, so
Figure PCTCN2020122756-appb-000028
among them,
Figure PCTCN2020122756-appb-000029
Input e x1 (t), e θ1 (t), e x2 (t), e θ2 (t), ΔL(t), v y1 , α 1 , α 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period
Figure PCTCN2020122756-appb-000030
expression. The kinematic control model adopted is as follows:
Figure PCTCN2020122756-appb-000031
Figure PCTCN2020122756-appb-000031
Figure PCTCN2020122756-appb-000032
Figure PCTCN2020122756-appb-000032
在步骤4中，本发明对应的双AGV系统，采用领航AGV路径跟踪，跟随AGV进行路径跟踪并伴随双AGV系统内队形偏差补偿的总体方案，具有如下偏差变化模型：In step 4, the dual AGV system corresponding to the present invention adopts the pilot AGV path tracking, follows the AGV for path tracking and is accompanied by the overall plan of formation deviation compensation in the dual AGV system, and has the following deviation change model:
Figure PCTCN2020122756-appb-000033
Figure PCTCN2020122756-appb-000033
Figure PCTCN2020122756-appb-000034
Figure PCTCN2020122756-appb-000034
ΔL(t+1)＝ΔL(t)+T(v y1cosα 1-v x1sinα 1+v x2sinα 2+v y2cosα 2)       (6) ΔL(t+1)=ΔL(t)+T(v y1 cosα 1 -v x1 sinα 1 +v x2 sinα 2 +v y2 cosα 2 ) (6)
将输入向量
Figure PCTCN2020122756-appb-000035
代入上述偏差变化模型计算预测t+T时刻的领航AGV偏差e x1(t+T)及e θ1(t+T)，和跟随AGV偏差e x2(t+T)、e θ2(t+T)及ΔL(t+T)。并代入获得终端罚函数G；由t时刻领航AGV偏差状态e x1(t)及e θ1(t)和跟随AGV相关偏差状态e x2(t)、e θ2(t)及ΔL(t)，结合估计输入向量
Figure PCTCN2020122756-appb-000036
获得代价函数L。 Input vector
Figure PCTCN2020122756-appb-000035
Substitute the above-mentioned deviation change model to calculate the pilot AGV deviation e x1 (t+T) and e θ1 (t+T) at time t+T, and the following AGV deviation e x2 (t+T), e θ2 (t+T) And ΔL(t+T). Substitute into the terminal penalty function G; lead the AGV deviation state e x1 (t) and e θ1 (t) at time t and follow the AGV related deviation state e x2 (t), e θ2 (t) and ΔL(t), combined Estimate the input vector
Figure PCTCN2020122756-appb-000036
Obtain the cost function L.
在步骤4中，设置目标函数H为终端罚函数G及代价函数L的和，则有H＝G+L；求解H的二 次规划问题，获得优化参数k i，将求解获得的ki代入运动学控制模型获得领航AGV与跟随AGV的优化输入向量
Figure PCTCN2020122756-appb-000037
In step 4, set the objective function H as the sum of the terminal penalty function G and the cost function L, then H=G+L; solve the quadratic programming problem of H, obtain the optimized parameter k i , and substitute the obtained ki into the motion Learn the control model to obtain the optimized input vector of the pilot AGV and the following AGV
Figure PCTCN2020122756-appb-000037
在步骤4中，采用了时域滚动预测控制的优化策略，于t时刻将求解目标函数H的二次规划问题获得的优化输入向量
Figure PCTCN2020122756-appb-000038
作为当前时域(t，t+δ)内的控制量对系统进行控制，具体步骤包括： In step 4, the optimization strategy of time-domain rolling predictive control is adopted, and the optimized input vector obtained by solving the quadratic programming problem of the objective function H at time t
Figure PCTCN2020122756-appb-000038
As the control variable in the current time domain (t, t+δ) to control the system, the specific steps include:
步骤4.1：估计输入向量
Figure PCTCN2020122756-appb-000039
代入权利要求3所述的偏差变化模型预测t+T时刻的领航偏差e x1(t+T)及e θ1(t+T)，和跟随AGV偏差e x2(t+T)、e θ2(t+T)及ΔL(t+T)，并代入获得终端罚函数G。 Step 4.1: Estimate the input vector
Figure PCTCN2020122756-appb-000039
Substituted into the deviation change model of claim 3 to predict the pilot deviation e x1 (t+T) and e θ1 (t+T) at t+T, and the following AGV deviation e x2 (t+T), e θ2 (t +T) and ΔL(t+T), and substituted into them to obtain the terminal penalty function G.
步骤4.2：由t时刻领航AGV偏差状态e x1(t)及e θ1(t)和跟随AGV相关偏差状态e x2(t)、e θ2(t)及ΔL(t)，结合估计输入向量
Figure PCTCN2020122756-appb-000040
获得代价函数L。 Step 4.2: Lead the AGV deviation state e x1 (t) and e θ1 (t) from time t and follow the AGV related deviation state e x2 (t), e θ2 (t) and ΔL(t), combined with the estimated input vector
Figure PCTCN2020122756-appb-000040
Obtain the cost function L.
步骤4.3：设置目标函数H＝G+L，求解H关于参数k i的二次规划问题，将求解获得的k i代入获得领航AGV与跟随AGV的优化输入向量
Figure PCTCN2020122756-appb-000041
步骤4.4：在控制时段(t,t+δ)内，令真实输入向量
Figure PCTCN2020122756-appb-000042
其中0<δ≤T； Step 4.3: Set the objective function H=G+L, solve the quadratic programming problem of H with respect to the parameter k i , and substitute the obtained k i into the optimized input vector of the pilot AGV and the following AGV
Figure PCTCN2020122756-appb-000041
Step 4.4: In the control period (t, t+δ), let the real input vector
Figure PCTCN2020122756-appb-000042
Where 0<δ≤T;
步骤4.5：当双AGV系统运行到t+δ时刻后，将时刻t更新为t+δ，重复步骤1到步骤5。Step 4.5: When the dual AGV system runs to time t+δ, update time t to t+δ, and repeat steps 1 to 5.
下面结合附图4及附图5，对控制过程中的步骤4.1至4.3展开详细叙述，在预测控制中，滚动优化输入向量的目的在于使罚函数G(e i(t+T))与代价函数L(e i(t),u i(t))取和的目标函数H取到最小值，罚函数G(e i(t+T))的输入e i(t+T)由预测阶段得到，罚函数G应当为连续可微的正定函数。 The following is a detailed description of steps 4.1 to 4.3 in the control process with reference to Figures 4 and 5. In predictive control, the purpose of rolling and optimizing the input vector is to make the penalty function G(e i (t+T)) and the cost The objective function H of the sum of the function L(e i (t), u i (t)) is taken to the minimum value, and the input e i (t+T) of the penalty function G(e i (t+T)) is determined by the prediction stage Obtained, the penalty function G should be a continuously differentiable positive definite function.
参考上述方法，以领航AGV的运动学控制模型为例进行分析，可选择终端罚函数的形式如下：Refer to the above method and take the kinematics control model of the pilot AGV as an example for analysis. The form of the terminal penalty function can be selected as follows:
Figure PCTCN2020122756-appb-000043
Figure PCTCN2020122756-appb-000043
相应地，设置代价函数如下：Correspondingly, set the cost function as follows:
L(e 1(t),u 1(t))＝e 1(t) TQe 1(t)+u 1(t) TRu 1(t)         (8) L(e 1 (t),u 1 (t))=e 1 (t) T Qe 1 (t)+u 1 (t) T Ru 1 (t) (8)
其中，e 1(t)＝[e x1(t) e θ1(t)] T，u 1(t)＝[v x1(t) w θ1(t)] T，权值矩阵Q与R则为半正定的对称矩阵。设置代价函数权值矩阵Q和R如下： Among them, e 1 (t) = [e x1 (t) e θ1 (t)] T , u 1 (t) = [v x1 (t) w θ1 (t)] T , the weight matrix Q and R are Symmetric positive semi-definite matrix. Set the cost function weight matrix Q and R as follows:
Figure PCTCN2020122756-appb-000044
且
Figure PCTCN2020122756-appb-000045
Figure PCTCN2020122756-appb-000044
And
Figure PCTCN2020122756-appb-000045
则可得，Then you can get,
L(e 1(t),u 1(t))＝q x1e x1(t) 2+q θ1e θ1(t) 2+r x1v x1(t) 2+r θ1ω 1(t) 2       (9) L(e 1 (t),u 1 (t))=q x1 e x1 (t) 2 +q θ1 e θ1 (t) 2 +r x1 v x1 (t) 2 +r θ1 ω 1 (t) 2 (9)
问题转化为关于参数k i的目标函数H＝G(e 1(t+T))+L(e 1(t),u 1(t))的二次规划求解问题。式(2)中可得，偏差向量e 1(t+T)为与偏差向量e 1(t)与输入向量u 1(t)有关的值，t时刻偏差向量e 1(t)＝[e x1(t) e θ1(t)] T则由该时刻传感器测量数据解析得到，为已知值。在(5)中，输入向量的表达式又是经由反馈控制系数k 1，k 2，以及系统t时刻偏差向量e 1(t)与速度v y1(t)可得。因此，最终的优化目的变为寻求系数k 1，k 2的在函数H可获得最小值的最优解。最后，把解得的系数k 1，k 2代入得到输入向量
Figure PCTCN2020122756-appb-000046
并用于(t,t+δ)时间段内的控制。 The problem is transformed into a quadratic programming problem of the objective function H=G(e 1 (t+T))+L(e 1 (t), u 1 (t)) about the parameter k i. Equation (2) can be obtained, the deviation vector e 1 (t+T) is the value related to the deviation vector e 1 (t) and the input vector u 1 (t), the deviation vector e 1 (t) = [e x1 (t) e θ1 (t)] T is obtained by analyzing the sensor measurement data at that moment and is a known value. In (5), the expression of the input vector is again obtained via the feedback control coefficients k 1 , k 2 , and the deviation vector e 1 (t) and speed v y1 (t) of the system at time t. Therefore, the final optimization goal becomes to find the optimal solution that can obtain the minimum value in the function H of the coefficients k 1 and k 2. Finally, substitute the solved coefficients k 1 and k 2 into the input vector
Figure PCTCN2020122756-appb-000046
And used for the control in the (t, t+δ) time period.
而对于跟随AGV模型，类似地设置终端罚函数：For the following AGV model, the terminal penalty function is set similarly:
Figure PCTCN2020122756-appb-000047
Figure PCTCN2020122756-appb-000047
设置代价函数：Set the cost function:
Figure PCTCN2020122756-appb-000048
Figure PCTCN2020122756-appb-000048
同样地求解优化解k 3，k 4，k 5，并代入(6)得到Follower的控制输入向量
Figure PCTCN2020122756-appb-000049
并作用于时间段(t,t+δ)内。等到达时刻t+δ后重新上述步骤，并更新反馈控制系数k 3，k 4，k 5。 Similarly, solve the optimized solutions k 3 , k 4 , k 5 , and substitute (6) to obtain the control input vector of Follower
Figure PCTCN2020122756-appb-000049
And it acts on the time period (t, t+δ). After reaching the time t+δ, repeat the above steps, and update the feedback control coefficients k 3 , k 4 , and k 5 .
至此，上述内容即为双AGV协同运载系统内领航AGV及跟随AGV基于滚动时域预测控制的控制输入量求解过程。So far, the above content is the solution process of the control input of the pilot AGV and the follower AGV based on the rolling time domain predictive control in the dual AGV cooperative delivery system.
结合附图4，在步骤5中，于运动控制层处，将优化输入向量
Figure PCTCN2020122756-appb-000050
解算成领航AGV及跟随AGV各轮的转速输出量，并发往驱动器驱动领航AGV及跟随AGV的电机运转，从而进行双AGV协同系统的路径跟踪协同运载运行。在运动控制层中，领航AGV及跟随AGV的控制器接收来自数据解算层的运动控制量，通过全向AGV逆运动学方程解算成领航AGV及跟随AGV各轮的转速输出量，并发往驱动器驱动领航AGV及跟随AGV的电机运转，从而进行双AGV协同系统的路径跟踪协同运载运行。通过基于麦克纳姆轮的全向移动AGV运动学模型，将前述数据解算层获得的领航AGV的速度、角速度真实输入向量
Figure PCTCN2020122756-appb-000051
转化为各轮的转速输入量，并通过控制器发往驱动器从而驱动领航AGV的电机运行；将前述数据解算层获得的跟随AGV速 度、角速度真实输入向量
Figure PCTCN2020122756-appb-000052
转化为各轮的转速输入量，并通过控制器发往驱动器从而驱动跟随AGV的电机运行。转化模型如下： With reference to Figure 4, in step 5, at the motion control layer, the input vector will be optimized
Figure PCTCN2020122756-appb-000050
It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system. In the motion control layer, the pilot AGV and the controller following the AGV receive the motion control value from the data solution layer, and the omnidirectional AGV inverse kinematics equation is used to solve the rotation speed output of each wheel of the pilot AGV and the following AGV. The driver drives the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated carrying operation of the dual AGV cooperative system. Through the omnidirectional moving AGV kinematics model based on the Mecanum wheel, the real input vector of the speed and angular velocity of the pilot AGV obtained by the aforementioned data solution layer
Figure PCTCN2020122756-appb-000051
Converted into the rotational speed input of each wheel, and sent to the driver through the controller to drive the motor of the pilot AGV; the real input vector of the following AGV speed and angular velocity obtained by the aforementioned data solution layer
Figure PCTCN2020122756-appb-000052
It is converted into the rotational speed input of each wheel and sent to the driver through the controller to drive the motor following the AGV to run. The conversion model is as follows:
Figure PCTCN2020122756-appb-000053
Figure PCTCN2020122756-appb-000053
R为麦克纳姆轮的半径；L为轮子到AGV中心车长方向的距离，即为车长的1/2；W为轮子车体中心车宽方向的距离，即为车宽的1/2；
Figure PCTCN2020122756-appb-000054
为领航AGV的y方向(即前进方向)的速度分量，系统运行时设定为固定值。
Figure PCTCN2020122756-appb-000055
为AGV(i＝1为领航AGV，i＝2为跟随AGV)的各轮转速控制量。 R is the radius of the Mecanum wheel; L is the distance from the wheel to the center of the AGV, which is 1/2 of the length of the vehicle; W is the distance from the center of the wheel to the center of the vehicle, which is 1/2 of the width of the vehicle ；
Figure PCTCN2020122756-appb-000054
It is the speed component in the y direction (that is, the forward direction) of the pilot AGV, which is set to a fixed value when the system is running.
Figure PCTCN2020122756-appb-000055
It is the AGV (i=1 is the pilot AGV, i=2 is the follower AGV) of each wheel speed control amount.
为适用于双AGV协同运行构型的控制方法，本发明设计了三层拓扑结构模型。结合附图2，对本发明中的三层拓扑结构协同控制模型进行叙述。In order to be suitable for the control method of the dual AGV cooperative operation configuration, the present invention designs a three-layer topology structure model. With reference to Figure 2, the three-layer topology collaborative control model of the present invention will be described.
在该模型中，以各传感器及通讯模块构成传感通讯层。其中，在各AGV上由CCD摄像头及DSP处理器形成的视觉模块用于识别路径并分别获得领航AGV及跟随AGV的路径角度偏差及路径距离偏差。由编码器获得领航AGV及跟随AGV各轮转速，每个AGV共四个编码器，分别安置在驱动四个麦克纳姆轮的电机输出轴处。由角度传感器获得领航AGV与工件之间的队形角度偏差以及跟随AGV与工件之间的队形角度偏差，由位移传感器获得队形距离偏差。由无线通讯模块及串口通讯方式构建双AGV协同运载系统内部的信息传输网络。其中，以领航AGV的无线通讯模块作为服务器端，跟随AGV、上位机及融合解算中心的无线通讯模块作为客户端接入，形成星形布局的通讯网络。三层拓扑结构模型中，以数据融合解算中心构成数据融合处理层，领航AGV的路径偏差(距离、角度)、队形角度偏差及领航AGV速度、角速度信息由通讯层传送至融合解算中心，跟随AGV的路径偏差(距离、角度)、队形角度偏差、队形距离偏差及跟随AGV各速度、角速度信息由传感器采集并接收进融合解算中心，数据融合处理得到领航AGV及跟随AGV的速度及角速度输入量。以领航AGV及跟随AGV的控制器、电机、驱动器等执行机构构成运动控制层，通过解算得到的各AGV速度角速度输入量根据麦克纳姆轮全向AGV的运动学反解得到领航AGV及跟随AGV的各轮转速，发送速度指令经由驱动器控制电机驱动。以上述传感通信层、数据解算层及运动控制层为基本框架构成三层拓扑结构的协同控制模型。In this model, sensors and communication modules constitute the sensor communication layer. Among them, the vision module formed by the CCD camera and the DSP processor on each AGV is used to identify the path and obtain the path angle deviation and path distance deviation of the pilot AGV and the following AGV, respectively. The encoder obtains the wheel speed of the pilot AGV and the following AGV. Each AGV has a total of four encoders, which are respectively arranged at the output shaft of the motor that drives the four mecanum wheels. The formation angle deviation between the leading AGV and the workpiece and the formation angle deviation between the following AGV and the workpiece are obtained by the angle sensor, and the formation distance deviation is obtained by the displacement sensor. The information transmission network inside the dual AGV cooperative delivery system is constructed by wireless communication modules and serial communication methods. Among them, the wireless communication module of Linghang AGV is used as the server, and the wireless communication module of the AGV, the upper computer and the integrated solution center is used as the client to access, forming a star-shaped communication network. In the three-layer topology model, the data fusion solution center constitutes the data fusion processing layer, and the navigation AGV path deviation (distance, angle), formation angle deviation, and navigation AGV speed and angular velocity information are transmitted from the communication layer to the fusion solution center , Follow the AGV path deviation (distance, angle), formation angle deviation, formation distance deviation, and follow the AGV speed, angular velocity information is collected by the sensor and received into the fusion solution center, data fusion processing to obtain the leading AGV and the following AGV Speed and angular velocity input. The motion control layer is composed of the pilot AGV and the controller, motor, driver and other actuators that follow the AGV. The velocity and angular velocity input of each AGV obtained by the solution is obtained by the inverse kinematics of the mecanum wheel omnidirectional AGV to obtain the pilot AGV and follow The rotation speed of each wheel of the AGV is sent through the driver to control the motor drive. Taking the above-mentioned sensor communication layer, data solution layer and motion control layer as the basic framework, a cooperative control model with a three-layer topology structure is constructed.
在传感通信层中，通过系统内部构建的星形布局通信网络，如附图3所示，网络内部存在信息交互，其中实线箭头表示通过无线传输的数据流，虚线箭头表示通过串行总线传输的数据流。双AGV系统内的交互信息主要有：In the sensor communication layer, through the star-shaped communication network built within the system, as shown in Figure 3, there is information interaction inside the network, where the solid arrow represents the data stream transmitted through wireless, and the dashed arrow represents the serial bus The transmitted data stream. The interactive information in the dual AGV system mainly includes:
(1)领航AGV与跟随AGV之间的AGV避障及启停信息交互，直接通过连接于各AGV控制器上的无线模块进行通讯，达到双AGV系统同步启停的功能；(1) The AGV obstacle avoidance and start-stop information interaction between the pilot AGV and the following AGV is directly communicated through the wireless module connected to each AGV controller to achieve the function of synchronous start and stop of the dual AGV system;
(2)领航AGV与融合解算中心之间的领航AGV速度信息、偏差信息及控制量信息交互。领航AGV当时刻运行速度、路径偏差(角度、距离)及队形角度偏差，经由领航AGV控制器与融合解算中心之间的无线通讯发往融合解算中心；融合解算中心经过数学模型解算得到领航AGV的速度及角速度控制量，并经由无线通讯发往领航AGV控制；(2) The speed information, deviation information and control amount information interaction between the pilot AGV and the fusion solution center. The running speed, path deviation (angle, distance) and formation angle deviation of the leading AGV at the moment are sent to the fusion solving center through the wireless communication between the leading AGV controller and the fusion solution center; the fusion solution center is solved by the mathematical model Calculate the speed and angular velocity control of the pilot AGV, and send it to the pilot AGV via wireless communication for control;
(3)领航AGV与上位机系统之间的状态及指令信息交互。领航AGV控制器发往上位机系统的信息为双AGV系统运行的状态信息，包括：双AGV系统的运行速度、避障信息、运行状态。上位机系统往领航AGV控制器发送指令信息和任务信息，用于控制整个双AGV系统的启停动作及任务接收执行；(3) The status and command information interaction between the pilot AGV and the host computer system. The information sent by the pilot AGV controller to the host computer system is the operating status information of the dual AGV system, including: the operating speed, obstacle avoidance information, and operating status of the dual AGV system. The host computer system sends instruction information and task information to the pilot AGV controller, which is used to control the start and stop actions of the entire dual AGV system and task reception and execution;
(4)跟随AGV与融合解算中心之间存在信息交互。跟随AGV将当前运行的速度角速度通过串行总线发往融合结算中心；融合结算中心将解算后的跟随AGV速度、角速度控制量发往跟随AGV。(4) There is information interaction between the following AGV and the fusion solution center. The following AGV sends the current running speed and angular velocity to the fusion settlement center through the serial bus; the fusion settlement center sends the calculated following AGV speed and angular velocity control to the following AGV.
在附图2的数据融合处理层中，通过各AGV的路径角度偏差、路径距离偏差、队形角度偏差、队形距离偏差等信息，融合当前时刻领航AGV及跟随AGV的运动速度及角速度，在融合解算中心通过运动学模型获得双AGV速度及角速度控制量；引入基于时域滚动优化预测控制对双AGV运动学的时间离散控制进行优化，获得离散时刻的运动控制输入量并发往运动控制层的领航AGV控制器及跟随AGV控制器。In the data fusion processing layer of Figure 2, the path angle deviation, path distance deviation, formation angle deviation, formation distance deviation and other information of each AGV are used to fuse the movement speed and angular velocity of the pilot AGV and the following AGV at the current moment. The Fusion Solution Center obtains the speed and angular velocity control of the dual AGV through the kinematics model; introduces the time-domain rolling optimization predictive control to optimize the time discrete control of the dual AGV kinematics, and obtains the discrete moment motion control input and sends it to the motion control The pilot AGV controller and the follower AGV controller of the layer.
以上所述仅是本发明的优选实施方式，应当指出：对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。The above are only preferred embodiments of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications are also It should be regarded as the protection scope of the present invention.

Claims (7)

  1. 一种双AGV协同运载控制方法，其特征在于，所述双AGV为全向移动，前后布局，前者为领航AGV，后者为跟随AGV，所述方法包括如下步骤：A dual AGV coordinated carrier control method, characterized in that the dual AGV is omni-directional movement, front and rear layout, the former is a pilot AGV, and the latter is a follower AGV, the method includes the following steps:
    步骤1：采集t时刻的领航AGV的路径距离偏差e x1(t)、路径角度偏差e θ1(t)、相对于工件的队形角度偏差α 1和车轮转速，以及跟随AGV的路径距离偏差e x2(t)、路径角度偏差e θ2(t)、队形距离偏差ΔL(t)、相对于工件的队形角度偏差α 2和车轮转速； Step 1: Collect the path distance deviation e x1 (t) of the pilot AGV at time t, the path angle deviation e θ1 (t), the formation angle deviation α 1 relative to the workpiece and the wheel speed, and the path distance deviation e of the following AGV x2 (t), path angle deviation e θ2 (t), formation distance deviation ΔL(t), formation angle deviation relative to the workpiece α 2 and wheel speed;
    步骤2：根据领航AGV和跟随AGV的车轮转速，求得领航AGV的侧移速度分量v x1、前进速度分量v y1及角速度w 1，跟随AGV侧移速度分量v x2、前进速度分量v y2及角速度w 2； Step 2: According to the wheel speed of the pilot AGV and the following AGV, obtain the side shift velocity component v x1 , the forward velocity component v y1 and the angular velocity w 1 of the pilot AGV, the follow AGV side shift velocity component v x2 , the forward velocity component v y2 and Angular velocity w 2 ;
    步骤3：令
    Figure PCTCN2020122756-appb-100001
    其中，
    Figure PCTCN2020122756-appb-100002
    将e x1(t)、e θ1(t)、e x2(t)、e θ2(t)、ΔL(t)、v y1、α 1、α 2输入运动学控制模型，得到时域t到t+T时段内的估计输入向量
    Figure PCTCN2020122756-appb-100003
    表达式；     Step 3: Let
    Figure PCTCN2020122756-appb-100001
    among them,
    Figure PCTCN2020122756-appb-100002
    Input e x1 (t), e θ1 (t), e x2 (t), e θ2 (t), ΔL(t), v y1 , α 1 , α 2 into the kinematics control model to obtain the time domain t to t + Estimated input vector during T period
    Figure PCTCN2020122756-appb-100003
    expression;
    步骤4：根据t时刻领航AGV及跟随AGV的速度状态和偏差状态、估计输入向量
    Figure PCTCN2020122756-appb-100004
    以及偏差变化模型，采用时域滚动预测控制的优化策略，获得优化输入向量
    Figure PCTCN2020122756-appb-100005
    Step 4: Estimate the input vector according to the speed state and deviation state of the pilot AGV and following the AGV at time t
    Figure PCTCN2020122756-appb-100004
    As well as the deviation change model, the optimization strategy of time-domain rolling predictive control is adopted to obtain the optimized input vector
    Figure PCTCN2020122756-appb-100005
    步骤5：将优化输入向量
    Figure PCTCN2020122756-appb-100006
    解算成领航AGV及跟随AGV各轮的转速输出量，并发往驱动器驱动领航AGV及跟随AGV的电机运转，从而进行双AGV协同系统的路径跟踪协同运载运行。     Step 5: The input vector will be optimized
    Figure PCTCN2020122756-appb-100006
    It is calculated into the speed output of each wheel of the pilot AGV and the following AGV, and sent to the driver to drive the pilot AGV and the motor following the AGV to operate, so as to carry out the path tracking and coordinated transportation operation of the dual AGV cooperative system.
  2. 根据权利要求1所述的一种双AGV协同运载控制方法，其特征在于，所述运动学控制模型为：The dual AGV cooperative carrier control method according to claim 1, wherein the kinematics control model is:
    Figure PCTCN2020122756-appb-100007
    Figure PCTCN2020122756-appb-100007
    Figure PCTCN2020122756-appb-100008
    Figure PCTCN2020122756-appb-100008
  3. 根据权利要求1所述的一种双AGV协同运载控制方法，其特征在于，所述偏差变化模型为：The dual AGV cooperative carrier control method according to claim 1, wherein the deviation change model is:
    Figure PCTCN2020122756-appb-100009
    Figure PCTCN2020122756-appb-100009
    Figure PCTCN2020122756-appb-100010
    Figure PCTCN2020122756-appb-100010
    ΔL(t+1)＝ΔL(t)+T(v y1cosα 1-v x1sinα 1+v x2sinα 2+v y2cosα 2)。 ΔL(t+1)=ΔL(t)+T(v y1 cosα 1 -v x1 sinα 1 +v x2 sinα 2 +v y2 cosα 2 ).
  4. 根据权利要求1所述的一种双AGV协同运载控制方法，其特征在于，领航AGV的终端罚函数的形式如下：A dual AGV cooperative carrier control method according to claim 1, wherein the terminal penalty function of the pilot AGV has the following form:
    Figure PCTCN2020122756-appb-100011
    Figure PCTCN2020122756-appb-100011
    代价函数的形式如下：The form of the cost function is as follows:
    L(e 1(t),u 1(t))＝e 1(t) TQe 1(t)+u 1(t) TRu 1(t) L(e 1 (t),u 1 (t))=e 1 (t) T Qe 1 (t)+u 1 (t) T Ru 1 (t)
    其中，e 1(t)＝[e x1(t) e θ1(t)] T，u 1(t)＝[v x1(t) w θ1(t)] T，权值矩阵Q与R则为半正定的对称矩阵；设置代价函数权值矩阵Q和R如下： Among them, e 1 (t) = [e x1 (t) e θ1 (t)] T , u 1 (t) = [v x1 (t) w θ1 (t)] T , the weight matrix Q and R are Positive semi-definite symmetric matrix; set the cost function weight matrix Q and R as follows:
    Figure PCTCN2020122756-appb-100012
    且
    Figure PCTCN2020122756-appb-100013
    Figure PCTCN2020122756-appb-100012
    And
    Figure PCTCN2020122756-appb-100013
    则，then,
    L(e 1(t),u 1(t))＝q x1e x1(t) 2+q θ1e θ1(t) 2+r x1v x1(t) 2+r θ1ω 1(t) 2 L(e 1 (t),u 1 (t))=q x1 e x1 (t) 2 +q θ1 e θ1 (t) 2 +r x1 v x1 (t) 2 +r θ1 ω 1 (t) 2
    跟随AGV模型的终端罚函数的形式如下：The form of the terminal penalty function following the AGV model is as follows:
    Figure PCTCN2020122756-appb-100014
    Figure PCTCN2020122756-appb-100014
    代价函数的形式如下：The form of the cost function is as follows:
    L(e 2(t),u 2(t),ΔL(t))＝q x2e x1(t) 2+q θ2e θ1(t) 2+q 3ΔL(t) 2 L(e 2 (t),u 2 (t),ΔL(t))=q x2 e x1 (t) 2 +q θ2 e θ1 (t) 2 +q 3 ΔL(t) 2
    +r x2v x2(t) 2+r θ2ω 2(t) 2+r y2v y2(t) 2。 +r x2 v x2 (t) 2 +r θ2 ω 2 (t) 2 +r y2 v y2 (t) 2 .
  5. 根据权利要求1所述的一种双AGV协同运载控制方法，其特征在于，步骤4中，采用时域滚动预测控制的优化策略，于t时刻将求解目标函数H的二次规划问题获得的优化输入向量
    Figure PCTCN2020122756-appb-100015
    作为当前时域(t，t+δ)内的控制量对系统进行控制，具体步骤包括：     A dual AGV cooperative carrier control method according to claim 1, characterized in that, in step 4, an optimization strategy of time-domain rolling predictive control is adopted, and the optimization obtained by solving the quadratic programming problem of the objective function H at time t Input vector
    Figure PCTCN2020122756-appb-100015
    As the control variable in the current time domain (t, t+δ) to control the system, the specific steps include:
    步骤4.1：估计输入向量
    Figure PCTCN2020122756-appb-100016
    代入权利要求3所述的偏差变化模型预测t+T时刻的领航偏差e x1(t+T)及e θ1(t+T)，和跟随AGV偏差e x2(t+T)、e θ2(t+T)及ΔL(t+T)，并代入获得终端罚函数G。     Step 4.1: Estimate the input vector
    Figure PCTCN2020122756-appb-100016
    Substituted into the deviation change model of claim 3 to predict the pilot deviation e x1 (t+T) and e θ1 (t+T) at t+T, and the following AGV deviation e x2 (t+T), e θ2 (t +T) and ΔL(t+T), and substituted into them to obtain the terminal penalty function G.
    步骤4.2：由t时刻领航AGV偏差状态e x1(t)及e θ1(t)和跟随AGV相关偏差状态e x2(t)、e θ2(t)及ΔL(t)，结合估计输入向量
    Figure PCTCN2020122756-appb-100017
    获得代价函数L。     Step 4.2: Lead the AGV deviation state e x1 (t) and e θ1 (t) from time t and follow the AGV related deviation state e x2 (t), e θ2 (t) and ΔL(t), combined with the estimated input vector
    Figure PCTCN2020122756-appb-100017
    Obtain the cost function L.
    步骤4.3：设置目标函数H＝G+L，求解H关于参数k i的二次规划问题，将求解获得的k i代入获得领航AGV与跟随AGV的优化输入向量
    Figure PCTCN2020122756-appb-100018
    Step 4.3: Set the objective function H=G+L, solve the quadratic programming problem of H with respect to the parameter k i , and substitute the obtained k i into the optimized input vector of the pilot AGV and the following AGV
    Figure PCTCN2020122756-appb-100018
    步骤4.4：在控制时段(t,t+δ)内，令真实输入向量
    Figure PCTCN2020122756-appb-100019
    其中0<δ≤T；     Step 4.4: In the control period (t, t+δ), let the real input vector
    Figure PCTCN2020122756-appb-100019
    Where 0<δ≤T;
    步骤4.5：当双AGV系统运行到t+δ时刻后，将时刻t更新为t+δ，重复步骤1到步骤5。Step 4.5: When the dual AGV system runs to time t+δ, update time t to t+δ, and repeat steps 1 to 5.
  6. 基于权利要求1所述方法的双AGV协同运载控制系统，其特征在于，包括传感通讯层、数据融合处理层和运动控制层；所述运动控制层包括领航AGV控制器和跟随AGV控制器，领航AGV控制器和跟随AGV控制器分别连接领航AGV和跟随AGV的车轮电机驱动器；所述传感通讯层用于监测双AGV的路径偏差、队形偏差和车轮转速，并将所监测信息传送给数据融合处理层，数据融合处理层对信息进行融合解算处理，得到领航AGV及跟随AGV的速度及角速度输入量，并将领航AGV及跟随AGV的速度及角速度输入量分别发送给领航AGV控制器和跟随AGV控制器；所述领航AGV控制器和跟随AGV控制器分别对所接收的速度及角速度输入量进行反解，得到各车轮转速，并将转速命令发送至电机驱动器。The dual AGV cooperative carrier control system based on the method of claim 1, characterized in that it includes a sensor communication layer, a data fusion processing layer, and a motion control layer; the motion control layer includes a pilot AGV controller and a follower AGV controller, The pilot AGV controller and the follower AGV controller are respectively connected to the pilot AGV and the wheel motor driver of the follower AGV; the sensor communication layer is used to monitor the path deviation, formation deviation and wheel speed of the dual AGV, and transmit the monitored information to The data fusion processing layer, the data fusion processing layer performs fusion solution processing on the information, obtains the speed and angular velocity input of the pilot AGV and the following AGV, and sends the speed and angular velocity input of the pilot AGV and the following AGV to the pilot AGV respectively And the following AGV controller; the pilot AGV controller and the following AGV controller respectively perform inverse analysis on the received speed and angular velocity input to obtain the rotation speed of each wheel, and send the rotation speed command to the motor driver.
  7. 根据权利要求6所述的一种双AGV协同运载控制系统，其特征在于，领航AGV与跟随AGV之间为运载工件；领航AGV与运载工件的连接处安装有角度传感器用于测量领航AGV与工件之间的队形角度偏差，跟随AGV与运载工件的连接处安装有角度传感器与位移传感器测量跟随AGV与工件之间的队形角度偏差与队形距离偏差；领航AGV和跟随AGV车身中部分别安装有垂直向下的视觉识别模块，用于测量路径角度偏差及路径距离偏差；领航AGV和跟随AGV车轮处分别安装有编码器采集车轮转速信息。The dual AGV cooperative carrier control system according to claim 6, characterized in that, between the pilot AGV and the following AGV is the carrying workpiece; the connection between the pilot AGV and the carried workpiece is equipped with an angle sensor for measuring the pilot AGV and the workpiece The angle deviation between the formation angle between the following AGV and the carrying workpiece is installed with an angle sensor and a displacement sensor to measure the formation angle deviation and the formation distance deviation between the following AGV and the workpiece; the pilot AGV and the following AGV are installed separately in the middle of the body There is a vertical downward visual recognition module for measuring path angle deviation and path distance deviation; encoders are installed at the wheels of the pilot AGV and the following AGV to collect wheel speed information.
PCT/CN2020/122756 2019-12-10 2020-10-22 Dual-agv collaborative carrying control system and method WO2021114888A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911255116.9 2019-12-10
CN201911255116.9A CN110989526B (en) 2019-12-10 2019-12-10 double-AGV cooperative carrying control system and method

Publications (1)

Publication Number Publication Date
WO2021114888A1 true WO2021114888A1 (en) 2021-06-17

Family

ID=70091520

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/122756 WO2021114888A1 (en) 2019-12-10 2020-10-22 Dual-agv collaborative carrying control system and method

Country Status (2)

Country Link
CN (1) CN110989526B (en)
WO (1) WO2021114888A1 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113534819A (en) * 2021-08-26 2021-10-22 鲁东大学 Method and storage medium for pilot-follow multi-agent formation path planning
CN113658221A (en) * 2021-07-28 2021-11-16 同济大学 Monocular camera-based AGV pedestrian following method
CN113885514A (en) * 2021-10-25 2022-01-04 上海影谱科技有限公司 AGV path tracking method and system based on fuzzy control and geometric tracking
CN113978574A (en) * 2021-12-01 2022-01-28 湖北物资流通技术研究所(湖北物资流通生产力促进中心) A modularization AGV for many scenes
CN114840003A (en) * 2022-03-21 2022-08-02 哈尔滨工程大学 Single-pilot multi-AUV (autonomous underwater vehicle) cooperative positioning and trajectory tracking control method
CN114995405A (en) * 2022-05-19 2022-09-02 同济大学 AGV cooperative handling method based on open dynamic environment multi-target cooperative theory
CN116449850A (en) * 2023-06-12 2023-07-18 南京泛美利机器人科技有限公司 Three-body cooperative transportation method and system based on behavioral cloning and cooperative coefficient
CN118192392A (en) * 2024-05-16 2024-06-14 成都睿芯行科技有限公司 Double-vehicle linkage control method

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110989526B (en) * 2019-12-10 2022-04-08 南京航空航天大学 double-AGV cooperative carrying control system and method
CN111702782B (en) * 2020-06-29 2022-07-19 重庆大学 Electricity-changing robot lifting device structure and control parameter collaborative optimization method
CN112025697B (en) * 2020-07-10 2022-06-17 浙江工业大学 Integral model prediction control method of omnidirectional mobile robot
CN111813122B (en) * 2020-07-14 2022-11-08 南京航空航天大学苏州研究院 Multi-vehicle cooperative transportation rapid queue changing method based on omnidirectional moving AGV
CN112394727A (en) * 2020-10-20 2021-02-23 广东嘉腾机器人自动化有限公司 AGV (automatic guided vehicle) cooperative transportation control method, storage medium and control system
CN113093215B (en) * 2021-04-01 2022-10-14 南京航空航天大学 Mobile platform tracking method based on laser ranging
TWI781619B (en) * 2021-05-14 2022-10-21 東元電機股份有限公司 Automatic guided vehicle support system and method
CN113759919B (en) * 2021-09-10 2024-03-15 华晟智能自动化装备有限公司 Mobile robot track tracking method and system
CN114296460B (en) * 2021-12-30 2023-12-15 杭州海康机器人股份有限公司 Collaborative handling method and device, readable storage medium and electronic equipment
CN114940452B (en) * 2022-04-20 2023-08-15 上海汇聚自动化科技有限公司 Transfer robot and transfer system
CN115826512B (en) * 2022-10-27 2023-07-25 珠海创智科技有限公司 Processing method and system of travel route and electronic equipment
CN116923232B (en) * 2023-09-12 2023-11-21 天津朗誉机器人有限公司 Double-vehicle linkage transportation system and method

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004127010A (en) * 2002-10-03 2004-04-22 Seiko Epson Corp Apparatus, system, method and program for automatic transportation
CN103587869A (en) * 2013-11-05 2014-02-19 无锡普智联科高新技术有限公司 Multi-robot logistics warehousing system based on bus mode and control method thereof
CN203542594U (en) * 2013-11-05 2014-04-16 无锡普智联科高新技术有限公司 Full-automatic carrier based on combination of multiple robots
JP2018163415A (en) * 2017-03-24 2018-10-18 株式会社日立製作所 Conveyance system, conveyance method, and automatic conveyance vehicle
CN109032128A (en) * 2018-06-13 2018-12-18 江南大学 The triangle formation control method of the discrete non-particle system of more AGV
CN109828580A (en) * 2019-02-27 2019-05-31 华南理工大学 A kind of Mobile Robot Formation's tracking and controlling method based on separate type ultrasonic wave
CN109871032A (en) * 2019-03-04 2019-06-11 中科院成都信息技术股份有限公司 A kind of multiple no-manned plane formation cooperative control method based on Model Predictive Control
CN110989526A (en) * 2019-12-10 2020-04-10 南京航空航天大学 double-AGV cooperative carrying control system and method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004127010A (en) * 2002-10-03 2004-04-22 Seiko Epson Corp Apparatus, system, method and program for automatic transportation
CN103587869A (en) * 2013-11-05 2014-02-19 无锡普智联科高新技术有限公司 Multi-robot logistics warehousing system based on bus mode and control method thereof
CN203542594U (en) * 2013-11-05 2014-04-16 无锡普智联科高新技术有限公司 Full-automatic carrier based on combination of multiple robots
JP2018163415A (en) * 2017-03-24 2018-10-18 株式会社日立製作所 Conveyance system, conveyance method, and automatic conveyance vehicle
CN109032128A (en) * 2018-06-13 2018-12-18 江南大学 The triangle formation control method of the discrete non-particle system of more AGV
CN109828580A (en) * 2019-02-27 2019-05-31 华南理工大学 A kind of Mobile Robot Formation's tracking and controlling method based on separate type ultrasonic wave
CN109871032A (en) * 2019-03-04 2019-06-11 中科院成都信息技术股份有限公司 A kind of multiple no-manned plane formation cooperative control method based on Model Predictive Control
CN110989526A (en) * 2019-12-10 2020-04-10 南京航空航天大学 double-AGV cooperative carrying control system and method

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113658221A (en) * 2021-07-28 2021-11-16 同济大学 Monocular camera-based AGV pedestrian following method
CN113658221B (en) * 2021-07-28 2024-04-26 同济大学 AGV pedestrian following method based on monocular camera
CN113534819A (en) * 2021-08-26 2021-10-22 鲁东大学 Method and storage medium for pilot-follow multi-agent formation path planning
CN113534819B (en) * 2021-08-26 2024-03-15 鲁东大学 Method and storage medium for pilot following type multi-agent formation path planning
CN113885514A (en) * 2021-10-25 2022-01-04 上海影谱科技有限公司 AGV path tracking method and system based on fuzzy control and geometric tracking
CN113885514B (en) * 2021-10-25 2024-05-07 上海影谱科技有限公司 AGV path tracking method and system based on fuzzy control and geometric tracking
CN113978574A (en) * 2021-12-01 2022-01-28 湖北物资流通技术研究所(湖北物资流通生产力促进中心) A modularization AGV for many scenes
CN114840003A (en) * 2022-03-21 2022-08-02 哈尔滨工程大学 Single-pilot multi-AUV (autonomous underwater vehicle) cooperative positioning and trajectory tracking control method
CN114995405A (en) * 2022-05-19 2022-09-02 同济大学 AGV cooperative handling method based on open dynamic environment multi-target cooperative theory
CN116449850A (en) * 2023-06-12 2023-07-18 南京泛美利机器人科技有限公司 Three-body cooperative transportation method and system based on behavioral cloning and cooperative coefficient
CN116449850B (en) * 2023-06-12 2023-09-15 南京泛美利机器人科技有限公司 Three-body cooperative transportation method and system based on behavioral cloning and cooperative coefficient
CN118192392A (en) * 2024-05-16 2024-06-14 成都睿芯行科技有限公司 Double-vehicle linkage control method

Also Published As

Publication number Publication date
CN110989526B (en) 2022-04-08
CN110989526A (en) 2020-04-10

Similar Documents

Publication Publication Date Title
WO2021114888A1 (en) Dual-agv collaborative carrying control system and method
CN101817182B (en) Intelligent moving mechanical arm control system
CN102231233B (en) Automatic guiding vehicle distributed autonomous cooperation control system and control method
WO2017004943A1 (en) Smart mobile detection platform for greenhouse
CN205375196U (en) Group robot control device for wind power plant inspection
CN107097241A (en) A kind of service robot and its control method
CN103978474B (en) A kind of job that requires special skills robot towards extreme environment
CN103901889A (en) Multi-robot formation control path tracking method based on Bluetooth communications
CN110162103A (en) A kind of unmanned plane independently cooperates with transportation system and method with intelligent vehicle group
CN108415460B (en) Combined and separated rotor wing and foot type mobile operation robot centralized-distributed control method
CN104597912A (en) Tracking flying control system and method of six-rotor unmanned helicopter
CN104950885A (en) UAV (unmanned aerial vehicle) fleet bilateral remote control system and method thereof based on vision and force sense feedback
WO2022179179A1 (en) Multi-agent collaborative autonomous transfer system for large equipment having heterogeneous characteristic
CN107102641A (en) A kind of original place driftage spinning solution based on laser aiming two-wheel differential AGV
CN111367285B (en) Wheeled mobile trolley cooperative formation and path planning method
CN106774352A (en) The robot control system of automatical pilot transportation vehicle and single two-way automatical pilot transportation vehicle of drive
WO2023045760A1 (en) Automatic train uncoupling robot and system
Lee et al. Artificial intelligence and Internet of Things for robotic disaster response
He et al. Intelligent logistics system of steel bar warehouse based on ubiquitous information
CN214846390U (en) Dynamic environment obstacle avoidance system based on automatic guided vehicle
CN206639040U (en) A kind of single two-way automatical pilot transportation vehicle of drive
Tong et al. Experimental and theoretical analysis on truss construction robot: automatic grasping and hoisting of concrete composite floor slab
CN116859849A (en) AGVS virtual-real fusion intelligent management and control system based on digital twin
CN110297489A (en) The control method and control system of automatic driving vehicle
CN112731914A (en) Cloud AGV application system of 5G smart factory

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: 20899538

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: 20899538

Country of ref document: EP

Kind code of ref document: A1