Communication perception control integrated multi-intelligent machine closed-loop control method and device

Abstract

The invention provides a closed-loop control method and device for multiple intelligent machines with integrated communication, perception and control, and relates to the technical field of integrated communication, perception and control, and the method comprises the steps: receiving a first control instruction sent by communication network equipment under the condition that a first mobile terminal meets a first condition, the instruction module is used for instructing to adjust a motion track and a motion speed of the first mobile terminal in a target motion period, and duration of the target motion period is preset duration; and under the condition that the first mobile terminal does not meet the first condition, according to a first motion state observation value sensed by the sensor of the first mobile terminal and / or a second motion state observation value sensed by the sensor of the second mobile terminal, adjusting the motion track and the motion speed of the first mobile terminal in the target motion period, the duration of the target motion period is first duration, the first duration is determined according to control environment state information in the workshop, and the first condition comprises at least one of the following items: the target motion period is located in an assembly area; and a steering area.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated communication, perception and control, and in particular to a closed-loop control method and device for a multi-intelligent machine with integrated communication, perception and control. Background Art

[0002] With the widespread application and development of technologies such as mobile communications and edge computing in the manufacturing industry, smart factories have deployed smart machines and mobile platforms on a large scale, such as automated guided vehicles (AGVs), to release human resources and effectively improve workshop production efficiency. In addition, by introducing a networked control system (NCS), smart factories adopt a centralized management mode to achieve networked closed-loop control of AGVs to efficiently perform industrial tasks such as material transportation and collaborative assembly. Among them, the closed-loop control process includes AGVs periodically uploading sensory data to the edge server on the base station side through a wireless communication network, and receiving and executing control instructions from the programmable logic controller (PLC) built into the edge server. However, with the expansion of factory production scale and the increase in production capacity demand, the demand for the number of controllable AGVs and working speed has increased sharply, and the failure rate and risk of loss of control of large AGV closed-loop control systems have also increased accordingly. Therefore, in order to meet the low-latency and efficient real-time control requirements of the AGV closed-loop control system, it is necessary to collaboratively design the perception, communication and control systems in the AGV closed-loop control process, including perception-control collaboration and communication-control collaboration, aiming to consume the least resources and improve control performance.

[0003] Specifically, the existing closed-loop control scheme of perception-control collaboration only considers the movement of AGVs along fixed paths, and may require a lot of infrastructure modifications to enable AGVs to navigate new paths. In this production mode, the AGV system has high deployment costs, relatively small cumulative control errors, low performance requirements for the communication system, and a relatively simple network structure model. However, with the large-scale deployment of mobile intelligent machines such as AGVs in manufacturing workshops, the number of retransmissions and the risk of congestion collision and loss of control have increased sharply. The complex physical environment and limited communication resources cannot meet the needs of large-scale intelligent terminals to efficiently perform industrial tasks and the problem of real-time control of AGVs.

[0004] Existing closed-loop control schemes for communication-control collaboration focus on studying parameters such as air interface delay and reliability in unidirectional links, ignoring the closed-loop interaction characteristics of closed-loop information flows of actual intelligent machines, making it difficult for intelligent machine networks to meet the requirements of efficient information flow interaction. For small AGV closed-loop control systems, the existing closed-loop control schemes under these simple network structures are already very detailed and can meet the requirements of real-time and efficient control of low-density AGVs in manufacturing workshops. However, with the increase in the number of AGVs and the speed of operation, most of the work ignores the coupling relationship between communication-perception-control, especially the degree of dependence of different industrial tasks on communication performance, control and perception errors. This leads to the transmission of too much redundant data, increased communication overhead, exponential increase in end-to-end delay, and the control system cannot meet the requirements of efficient and stable closed-loop interaction of large-scale intelligent machines.

[0005] In summary, existing mobile terminal (such as AGV) control systems have the problems of poor control performance and high resource consumption. Summary of the invention

[0006] The purpose of the present invention is to provide a multi-intelligent machine closed-loop control method and device with integrated communication, perception and control, so as to solve the problems of poor control performance and high resource consumption of existing mobile terminal control systems.

[0007] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0008] An embodiment of the present invention provides a closed-loop control method for a multi-intelligent machine with integrated communication, perception and control, which is applied to a first mobile terminal. The method includes:

[0009] When the first mobile terminal meets the first condition, receiving a first control instruction sent by a communication network device, wherein the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration;

[0010] In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located;

[0011] The first condition includes at least one of the following:

[0012] The first mobile terminal is located in an assembly area in a workshop;

[0013] The first mobile terminal is located in a turning area on an assembly line in a workshop.

[0014] Optionally, after adjusting the motion trajectory and motion speed of the first mobile terminal within the target motion cycle according to the first motion state observation value sensed by the sensor of the first mobile terminal and / or the second motion state observation value sensed by the sensor of the second mobile terminal, the method further includes:

[0015] Sending a third motion state observation value sensed by a sensor of the first mobile terminal to the communication network device;

[0016] Receive a second control instruction sent by the communication network device, wherein the second control instruction is sent by the communication network device when it is determined that the difference between the motion trajectory of the first mobile terminal and the preset motion trajectory is greater than a preset threshold based on the third motion state observation value sensed by the sensor and the preset motion trajectory, and the second control instruction is used to instruct the first mobile terminal to make adjustments according to the preset motion trajectory.

[0017] Optionally, the method further comprises:

[0018] sensing a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal;

[0019] Determining whether the first mobile terminal meets the first condition according to the first motion state observation value sensed by the sensor and the pre-stored workshop map data;

[0020] If so, determining that the first mobile terminal meets the first condition;

[0021] In a case where the first mobile terminal does not satisfy the first condition, indication information sent by the communication network device is received, where the indication information is used to indicate whether the first mobile terminal satisfies the first condition.

[0022] Optionally, receiving indication information sent by the communication network device includes:

[0023] Perceiving a first motion state observation value perceived by a sensor of the first mobile terminal in the workshop;

[0024] Sending the first motion observation value to the communication network device;

[0025] Receiving the indication information sent by the communication network device;

[0026] Among them, the communication network device is used to receive second sensor data sent by the sensor in the workshop, determine the third sensor data of the first mobile terminal observed in the second sensor data, determine the fourth motion state observation value perceived by the sensor of the first mobile terminal in the third sensor data, and process the fourth motion state observation value perceived by the sensor to obtain the fifth motion state observation value perceived by the sensor, and estimate the target motion state observation value of the first mobile terminal based on the first motion state observation value perceived by the sensor and the fifth motion state observation value perceived by the sensor, and determine whether the first mobile terminal meets the first condition based on the target motion state observation value and pre-stored workshop map data.

[0027] The first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop; the second motion state observation value sensed by the sensor includes a second position of the second mobile terminal in the workshop;

[0028] Adjusting the movement speed of the first mobile terminal in the first movement cycle according to the first movement state observation value sensed by the sensor of the first mobile terminal and / or the second movement state observation value sensed by the sensor of the second mobile terminal includes:

[0029] Obtaining a safety distance according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, a preset collision warning threshold factor and a first radius, wherein the first radius is a radius of an inscribed circle of a projection profile of the first mobile terminal on the ground;

[0030] Obtaining a collision distance according to a preset collision warning threshold factor and the safety distance;

[0031] When the Euclidean distance between the first position and the second position is greater than the safety distance, controlling the first mobile terminal to accelerate within the target motion cycle;

[0032] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, controlling the first mobile terminal to decelerate within the target motion cycle;

[0033] When the Euclidean distance between the first position and the second position is less than the collision distance, the first mobile terminal is controlled to perform emergency braking within the target motion cycle.

[0034] Optionally, receiving a first control instruction sent by the communication network device includes:

[0035] After the first motion cycle ends, sending a first motion state observation value sensed by a sensor of the first mobile terminal to the communication network device, wherein the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop, wherein the first motion cycle is a motion cycle adjacent to the target motion cycle;

[0036] Receiving a first control instruction sent by the communication network device;

[0037] Wherein, when the Euclidean distance between the first position and the second position is greater than the safety distance, the first control instruction is used to control the first mobile terminal to accelerate within the target motion cycle;

[0038] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, the first control instruction is used to control the first mobile terminal to decelerate within the target motion cycle;

[0039] In a case where the Euclidean distance between the first position and the second position is less than the collision distance, the first control instruction is used to control the first mobile terminal to perform emergency braking within the target motion cycle;

[0040] The safety distance is obtained by the communication network device according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, a preset collision warning threshold factor and a first radius, and the first radius is the radius of the inscribed circle of the projection outline of the first mobile terminal on the ground;

[0041] The collision distance is obtained according to a preset collision warning threshold factor and the safety distance;

[0042] The second location is sent by the second mobile terminal to the communication network device.

[0043] Optionally, the first motion state observation value sensed by the sensor includes a first position, a first steering angle, a current speed and a current motion trajectory of the first mobile terminal in the workshop;

[0044] Adjusting the motion trajectory of the first mobile terminal in a first motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal includes:

[0045] The motion trajectory of the first mobile terminal within the target motion cycle is obtained based on the current speed, the acceleration of the first mobile terminal, the duration of the first motion cycle, the first position, the first steering angle, the current motion trajectory and the first duration.

[0046] Optionally, the method further comprises:

[0047] According to the first time length, a time-related stability coefficient is obtained;

[0048] The acceleration of the first mobile terminal is obtained according to the communication delay, the safety distance, the maximum speed of the first mobile terminal and the stability coefficient.

[0049] Optionally, receiving a first control instruction sent by the communication network device includes:

[0050] sensing a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal;

[0051] Sending a first motion state observation value sensed by the sensor to the communication network device, where the first motion state observation value sensed by the sensor includes a current motion track of the first mobile terminal;

[0052] Receiving a first control instruction sent by the communication network device, the first control instruction includes a second motion trajectory, the second motion trajectory is obtained by the communication network device according to a trajectory error, and the trajectory error is an error between the current motion trajectory and a preset motion trajectory;

[0053] The motion trajectory of the first mobile terminal within the target motion cycle is adjusted according to the second motion trajectory, the preset duration, the first travel angle and the current trajectory.

[0054] Optionally, the method further comprises:

[0055] The first duration is determined based on the safety distance, the current speed of the first mobile terminal and the closed-loop communication delay.

[0056] An embodiment of the present invention further provides a multi-intelligent machine closed-loop control device integrating communication, perception and control, the device comprising: a control module;

[0057] The control module is used for:

[0058] When the first mobile terminal meets the first condition, receiving a first control instruction sent by the communication network device, wherein the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration;

[0059] In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located;

[0060] The first condition includes at least one of the following:

[0061] The first mobile terminal is located in an assembly area in a workshop;

[0062] The first mobile terminal is located in a turning area on an assembly line in a workshop.

[0063] An embodiment of the present invention also provides a multi-intelligent machine closed-loop control device with integrated communication, perception and control, comprising: a transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; when the processor executes the program or instruction, it implements the steps in the multi-intelligent machine closed-loop control method with integrated communication, perception and control as described in any one of the above.

[0064] An embodiment of the present invention also provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps in the multi-intelligent machine closed-loop control method with integrated communication, perception and control as described in any one of the above are implemented.

[0065] An embodiment of the present invention also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps in the multi-intelligent machine closed-loop control method with integrated communication, perception and control as described in any one of the above.

[0066] The beneficial effects of the above technical solution of the present invention are as follows:

[0067] The communication-perception-control integrated multi-intelligent machine closed-loop control method provided by the solution of the present invention is that when the first mobile terminal meets the first condition (the first mobile terminal is located in the assembly area in the workshop and / or the first mobile terminal is located in the turning area on the assembly line in the workshop), the first mobile terminal receives a first control instruction sent by the communication network device, and the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal in the first motion cycle, that is, the first mobile terminal passively adjusts its motion trajectory and motion speed in the target motion cycle according to the first control instruction. When the first mobile terminal does not meet the first condition, the first mobile terminal adjusts the motion trajectory and motion speed according to the first motion state observation value perceived by the sensor of the first mobile terminal and / or the second motion state observation value perceived by the sensor of the second mobile terminal. Observation values ​​are used to adjust the motion trajectory and speed of the first mobile terminal within the target motion cycle, that is, through the auxiliary control of the first mobile terminal, active decision-making is selected to actively control the motion trajectory and speed of the first mobile terminal. Through the combination of active and passive adjustment, the problem that the complex physical environment and limited communication resources cannot meet the needs of large-scale intelligent terminals to efficiently execute industrial tasks, and the problem of poor control stability caused by the inability of intelligent terminals to respond quickly to the environment is solved. In the active decision-making, the first duration of the target motion cycle is determined according to the control environment status information in the workshop where the first mobile terminal is located, that is, the duration of the motion cycle is adaptively adjusted, the number of times and overhead of the mobile terminal visiting the base station are reduced, the number of mobile terminals that can be controlled by the network is expanded, the production capacity of the workshop is improved, and resource consumption is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A flow chart of a closed-loop control method for a multi-intelligent machine with integrated communication, perception and control provided by an embodiment of the present invention;

[0069] Figure 2 A schematic diagram of the structure of a flexible automobile manufacturing assembly workshop provided by an embodiment of the present invention;

[0070] Figure 3 A schematic diagram of the probability of congestion collision between different AGV spacings provided in an embodiment of the present invention;

[0071] Figure 4 A schematic diagram showing a comparison between the duration of a traditional closed-loop interaction cycle provided in an embodiment of the present invention and the duration of a closed-loop interaction cycle obtained in the present invention;

[0072] Figure 5 A schematic diagram showing the comparison results of the number of AGVs that can be controlled by different solutions provided in the embodiments of the present invention;

[0073] Figure 6 A schematic diagram showing the comparison results of closed-loop communication delays of different solutions provided in an embodiment of the present invention;

[0074] Figure 7 A schematic diagram of the structure of a multi-intelligent machine closed-loop control device with integrated communication, perception and control provided by an embodiment of the present invention;

[0075] Figure 8 A schematic diagram of the structure of a multi-intelligent machine closed-loop control device with integrated communication, perception and control provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0076] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0077] In order to solve the problem that the existing complex physical environment and limited communication resources cannot meet the needs of large-scale intelligent terminals to efficiently execute industrial tasks and have poor control stability, an embodiment of the present invention provides a multi-intelligent machine closed-loop control method and device with integrated communication perception and control.

[0078] like Figure 1 As shown, an embodiment of the present invention provides a multi-intelligent machine closed-loop control method integrating communication, perception and control, which is applied to a first mobile terminal, and the method includes:

[0079] Step 101: When the first mobile terminal meets a first condition, receiving a first control instruction sent by a communication network device, wherein the first control instruction is used to instruct to adjust a motion trajectory and a motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration;

[0080] In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located;

[0081] The first condition includes at least one of the following:

[0082] The first mobile terminal is located in an assembly area in a workshop;

[0083] The first mobile terminal is located in a turning area on an assembly line in a workshop.

[0084] It should be noted that the first mobile terminal is any mobile terminal in the workshop, the mobile terminal can be an AGV or an intelligent AGV, and the communication network device can be a 5G base station, an edge server, or a PLC device on an edge server. In the embodiment of the present invention, the mobile terminal is AGV as an example for explanation, the communication network device is PLC as an example for explanation, and the workshop is an automobile manufacturing assembly workshop as an example for explanation. As an example, the closed-loop control process of AGV in an industrial environment is accurately described. Taking the transportation of materials by AGV in an automobile manufacturing assembly workshop as the scenario, a new type of integrated closed-loop control system model of telecontrol is established. The automobile flexible manufacturing assembly workshop has a structure such as Figure 2 As shown. There are four assembly lines in the workshop, namely the chassis assembly bus, the general assembly bus, the engine assembly line and the cockpit assembly line. At least one AGV is set on each assembly line, and the AGV may run along a circular or rectangular assembly line on the assembly line. The assembly area includes the assembly area on the chassis assembly line, the assembly area on the general assembly bus, the assembly area on the engine assembly line, and the assembly area on the cockpit assembly line. The workshop also includes the loading end area, the turning machine area and the unloading end area. Sensors (multiple) and base stations are also set up in the workshop. The data of multiple sensors are uploaded to the base station. The base station communicates with the PLC on the edge server, and data can be transmitted between them.

[0085] In the AGV closed-loop control process, the perception system, communication network and control system influence each other. The independent redundant design of perception, communication and control will lead to a large waste of resources, resulting in a surge in the number of retransmissions and the risk of congestion collision and loss of control, high communication overhead and exponential increase in end-to-end delay.

[0086] In order to solve the above technical problems, in this step, it is determined whether the first AGV (i.e., the first mobile terminal) is located in the assembly area in the workshop, or whether it is located in the turning area on the assembly line in the workshop, or whether it is located in both the assembly area in the workshop and the turning area on the assembly line in the workshop. Based on the above judgment results, active local control decisions (active closed-loop control) or passive reception of PLC control decisions (passive closed-loop control) are intelligently selected.

[0087] Among them, the target motion cycle is any motion cycle, and the motion cycle is a closed-loop interaction cycle. Within the target interaction cycle, it is determined whether the above-mentioned first condition is met, so as to trigger active closed-loop control and passive closed-loop control. If it is met, for example, the first AGV is located in the assembly area, and the AGV will receive parts assembly tasks from collaborative robotic arms, flip machines and other intelligent machines. The assembly task requires the AGV to frequently communicate with the communication network equipment in a fixed closed-loop interaction cycle, constantly calibrate its own trajectory and positioning information, and perform high-precision control synchronization with the robotic arms, etc., that is, the first AGV receives the first control instruction sent by the communication network equipment, and the duration of the target motion cycle is fixed, which is a preset duration. The preset duration is relatively short, such as a preset duration of 4ms. The first AGV adjusts the motion trajectory and motion speed within the preset duration according to the first control instruction. Slow and high-precision cyclic forward movement within a long period of time ensures efficient operation of the production process; if, for example, the first AGV is located in the linear motion area on the assembly line and in the non-assembly area, the AGVs in this area are all operating in a stable state, and there are no robotic arm assembly stations or other obstacles (such as other AGVs) between any AGVs, therefore, the AGV is mainly actively controlled, and the second AGV (second mobile terminal) is used as the obstacle avoidance target to adjust its own motion trajectory and speed, that is, the motion trajectory and speed of the first AGV are adjusted according to the motion state observation value of the first AGV and the motion state observation value of the second AGV.

[0088] The second AGV is an AGV that is adjacent to the first AGV in front. The adjacent to the first AGV in front can be understood as the second AGV and the first AGV are located on the same assembly line, the second AGV moves before the first AGV, and the second AGV is adjacent to the first AGV.

[0089] The motion state observation value includes at least one of the following:

[0090] The position of the mobile terminal in the workshop (current position); the position of the mobile terminal after the end of the previous movement cycle of the target movement cycle;

[0091] The current steering angle of the mobile terminal;

[0092] The current travel angle of the mobile terminal;

[0093] The current speed of the mobile terminal;

[0094] The current acceleration of the mobile terminal;

[0095] The current motion trajectory of the mobile terminal (the current motion trajectory is fitted based on the current position, the current steering angle, the steering angle of the mobile terminal after the end of the previous motion cycle of the target motion cycle, the speed of the mobile terminal after the end of the previous motion cycle of the target motion cycle, the acceleration of the mobile terminal after the end of the previous motion cycle of the target motion cycle, the current speed and the current acceleration, which can also be understood as the motion trajectory of the previous motion cycle of the target motion cycle).

[0096] Through the process of active AGV control, the problem of high-speed AGV’s inability to respond quickly to the environment and poor control performance is solved.

[0097] During the active control process, the duration of the target motion cycle is a first duration, and the first duration is determined based on the control environment status information in the workshop where the first mobile terminal is located. Generally speaking, the first duration is greater than the above preset duration.

[0098] The control environment status information (also referred to as real-time closed-loop control status) includes at least one of the following:

[0099] The safety distance between the first AGV and the second AGV;

[0100] The optimal speed of the first AGV;

[0101] cycle time (closed-loop communication delay).

[0102] That is, during the active control process, the first mobile terminal adaptively adjusts the duration of the closed-loop interaction cycle according to the control environment status information, allocates the duration of active and passive control decisions, greatly reduces the number of AGVs visiting the base station and the communication overhead, expands the number of AGVs that can be controlled by the network, and improves workshop production capacity.

[0103] Through the master control process, the problem of large communication overhead and high closed-loop communication delay caused by frequent interactions between large-scale AGVs and base stations is solved.

[0104] In summary, this step solves the problem of poor communication and control performance caused by busy communication of large-scale intelligent machines in traditional closed-loop control schemes.

[0105] Further, after adjusting the motion trajectory and motion speed of the first mobile terminal within the target motion cycle according to the first motion state observation value perceived by the sensor of the first mobile terminal and / or the second motion state observation value perceived by the sensor of the second mobile terminal, the method further includes:

[0106] Sending a third motion state observation value sensed by a sensor of the first mobile terminal to the communication network device;

[0107] Receive a second control instruction sent by the communication network device, the second control instruction is sent by the communication network device when it is determined that the difference between the motion trajectory of the first mobile terminal and the preset motion trajectory is greater than a preset threshold value based on the third motion state observation value sensed by the sensor and the preset motion trajectory, and the second control instruction is used to instruct the first mobile terminal to adjust according to the preset motion trajectory. The preset motion trajectory can be a motion trajectory pre-stored in the PLC device, which can be understood as the normal motion trajectory of the AGV.

[0108] The difference between the motion trajectory of the first mobile terminal and the preset motion trajectory can be understood as the average distance between several trajectory points on the motion trajectory of the first mobile terminal and several corresponding trajectory points on the preset motion trajectory, or the maximum distance corresponding to each trajectory point.

[0109] That is, after each round of active control, it communicates with the communication network equipment, passively receives the control decisions of the PLC equipment, synchronizes and verifies the driving trajectory, and achieves rapid environmental response with low communication overhead.

[0110] It should also be noted that the above-mentioned motion state observation values ​​can be sent by the mobile terminal to the 5G base station, and then sent by the 5G base station to the PLC device.

[0111] The following is a detailed description of the process of determining whether the AGV is in the assembly area:

[0112] In an optional embodiment, the method further includes:

[0113] Perceiving a first motion state observation value perceived by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal, that is, the first mobile terminal obtains its own sensor data, and obtains a first motion state observation value perceived by the sensor of the first mobile terminal in the workshop according to the sensor data;

[0114] According to the first motion state observation value sensed by the sensor and the pre-stored workshop map data, it is determined whether the first mobile terminal meets the first condition, that is, the first mobile terminal matches the position of the mobile terminal in the workshop in the first motion state observation value sensed by the sensor with the pre-stored workshop map data, and determines whether the current position is an assembly area or a turning area. If so, it is determined that the first mobile terminal meets the first condition, which is helpful for the selection of subsequent control decisions. If not, the indication of the communication network device (that is, the second indication information) is received, and the communication network device determines whether the first condition is met.

[0115] Further, receiving the indication information sent by the communication network device includes:

[0116] Perceiving a first motion state observation value perceived by a sensor of the first mobile terminal in the workshop;

[0117] Sending first sensor data and the first motion observation value on the first mobile terminal to the communication network device;

[0118] Receiving the indication information sent by the communication network device;

[0119] Among them, the communication network device is used to receive second sensor data sent by the sensor in the workshop, determine the third sensor data of the first mobile terminal observed in the second sensor data, determine the fourth motion state observation value perceived by the sensor of the first mobile terminal in the third sensor data, and process the fourth motion state observation value perceived by the sensor to obtain the fifth motion state observation value perceived by the sensor, and estimate the target motion state observation value of the first mobile terminal based on the first motion state observation value perceived by the sensor and the fifth motion state observation value perceived by the sensor, and determine whether the first mobile terminal meets the first condition based on the target motion state observation value and pre-stored workshop map data.

[0120] Specifically, the 5G base station receives sensor data uploaded by the AGV (i.e., the first sensor data) and sensor data uploaded by sensors in the workshop through optical fiber (i.e., the second sensor data). The above sensor data is sent by the 5G base station to the PLC device to provide prior information for the PLC's control decision. Afterwards, the PLC device receives the first sensor device and the second sensor data, integrates them, estimates the motion state of the first mobile terminal, and then specifies the control decision.

[0121] The communication network device determines the third sensor data of the current motion state of the first mobile terminal observed by M (M is an integer greater than or equal to 1) target sensors in the second sensor data, which is recorded as {S1, S2, ..., S M}, determine the observed motion state X of the first mobile terminal in the third sensor data n (i.e., the fourth motion state observation value sensed by the sensor), the PLC device processes the fourth state observation value to obtain the fifth motion state observation value sensed by the sensor satisfy:

[0122]

[0123] in, is the Gaussian white noise of the observation value, N n ~N(0,σ 2 ).

[0124] The state observation value perceived by the AGV itself is (The first motion state observation value perceived by the sensor). The state of the AGV is estimated by the maximum likelihood estimation method based on the state observation values ​​from different sensors and the state values ​​perceived by the AGV itself. According to the above assumptions, X n (The X n The estimated value of the first motion state observation value sensed by the sensor and the fifth motion state observation value sensed by the sensor) (Observation value of target motion state) can be shown as:

[0125]

[0126] That is, the first motion state observation value sensed by the sensor and the fifth motion state observation value sensed by the sensor are estimated using the maximum likelihood estimation method to obtain the target motion state observation value.

[0127] The mean square error can be calculated as follows:

[0128]

[0129] Specifically, due to the influence of the number and accuracy of perception samples, there is a deviation between the actual position of the perceived AGV and the perceived position. At the same time, due to the dynamic movement of the AGV, the perceived AGV position and environmental information have a lag, which leads to an increase in perception errors, thereby affecting the calibration of the control trajectory and resulting in poor control system stability. Therefore, the above-mentioned target motion state observation value is calculated in combination with the sensor data, which can enable the communication network equipment to more accurately determine whether the first mobile terminal meets the first condition based on the target motion state observation value.

[0130] The process of adjusting the movement speed of the first mobile terminal is described in detail below:

[0131] First, considering that the assembly line is arranged in a circular and parallel manner, all AGVs move in a circular and step-by-step manner. At this time, when an AGV collides, the default obstacle is other AGVs. In addition, due to the different sizes of AGVs and the different speeds of their working states, their safety area range can be adaptively changed. The probability of congestion collision between different AGV spacings is different, such as Figure 3 shown.

[0132] When the AGV is traveling on the reference track (preset track), the collision between the AGV and obstacles can be divided into the following three situations. Figure 3 As shown in the first case, when the Euclidean distance (distance) H0 between the AGV and the obstacle (the AGV adjacent to the AGV) is greater than the safety distance H cIn the second case, when the distance is in the warning area, ω0H c ≤H0≤H c , there is a probability of collision. In the third case, when the distance is in the danger zone, R n ≤H0<ω0H c , there is a high probability of collision congestion. And, there is another situation, that is, H0<R n When , a collision will inevitably occur. Among them, ω0 represents the collision warning threshold factor of AGV, which is the preset value, R n Indicates the radius of the inscribed circle of the AGV's projection on the ground.

[0133] In an optional embodiment, the first motion state observation value sensed by the sensor includes a first position (current position) of the first mobile terminal in the workshop; the second motion state observation value sensed by the sensor includes a second position (current position) of the second mobile terminal in the workshop;

[0134] The first position of the first mobile terminal can be represented by the center point of the first AGV, and the second position of the second mobile terminal can be represented by the center point of the second AGV.

[0135] Adjusting the movement speed of the first mobile terminal in the first movement cycle according to the first movement state observation value sensed by the sensor of the first mobile terminal and / or the second movement state observation value sensed by the sensor of the second mobile terminal includes:

[0136] According to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, the preset collision warning threshold factor ω0 and the first radius R n , get the safe distance H c , wherein the first radius is the radius of the inscribed circle of the projection profile of the first mobile terminal on the ground, which is specifically expressed as follows:

[0137]

[0138] Among them, a max is the maximum acceleration of AGV, v max is the maximum speed of the AGV.

[0139] According to the preset collision warning threshold factor ω0 and the safety distance H c , and the collision distance ω0H is obtained c (Also known as warning distance).

[0140] When the Euclidean distance between the first position and the second position is greater than the safety distance, the first mobile terminal is controlled to accelerate within the first movement cycle, that is, when in the first case, it is in the safety area, and the control decision is to execute the acceleration command to ensure the production efficiency of the assembly line, so as to ensure that the distance is near the safety distance;

[0141] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, the first mobile terminal is controlled to decelerate in the first movement cycle, that is, in the second case, in order to avoid collision and to prevent the carried parts from tipping over due to inertia, the control decision issues a deceleration command;

[0142] When the Euclidean distance between the first position and the second position is less than the collision distance, the first mobile terminal is controlled to perform emergency braking within the target motion cycle, that is, in the third case mentioned above, when the distance is in a dangerous area, the control instruction issues an emergency braking command.

[0143] In the embodiment of the present invention, the congestion probability function model of the third situation and the above situation where a collision is bound to occur may be as follows:

[0144]

[0145] Where, exp(1-(H0 / R n )) is the collision probability function constructed by the Euclidean distance between the AGV and the obstacle (the second AGV). Assume that when the AGV n (The first mobile terminal) moves at a maximum speed v max Driving, the distance between the front vehicle and the front vehicle is reduced to ω0H c When the emergency brake command is executed, the AGV n The speed will be reduced from the maximum speed to 0, and the distance from the vehicle in front will be greater than or equal to R after stopping. n .

[0146] It should also be noted that, considering the visual blind spots when the AGV turns, the PLC needs to control the AGV based on real-time global information to prevent the risk of emergency collisions that may occur outside the line of sight. Therefore, it is assumed that when the AGV turns, completes or receives a task, it will send a control request to the base station based on a typical business cycle, and passively receive PLC control decisions to adjust its own motion trajectory.

[0147] In an optional embodiment, receiving a first control instruction sent by a communication network device includes:

[0148] After the first motion cycle ends, sending a first motion state observation value sensed by a sensor of the first mobile terminal to the communication network device, wherein the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop, wherein the first motion cycle is a motion cycle adjacent to the target motion cycle;

[0149] Receiving a first control instruction sent by the communication network device;

[0150] Wherein, when the Euclidean distance between the first position and the second position is greater than the safety distance, the first control instruction is used to control the first mobile terminal to accelerate within the first movement cycle;

[0151] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, the first control instruction is used to control the first mobile terminal to decelerate within the first movement cycle;

[0152] When the Euclidean distance between the first position and the second position is less than the collision distance, the first control instruction is used to control the first mobile terminal to perform emergency braking;

[0153] The safety distance is obtained by the communication network device according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, a preset collision warning threshold factor and a first radius, and the first radius is the radius of the inscribed circle of the projection outline of the first mobile terminal on the ground;

[0154] The collision distance is obtained according to a preset collision warning threshold factor and the safety distance;

[0155] The second location is sent by the second mobile terminal to the communication network device.

[0156] It should also be noted that the safety distance calculation process, the meaning of each parameter and the control content in this optional embodiment are consistent with the calculation process, the meaning of each parameter and the control content executed by the first mobile terminal in the above-mentioned optional embodiment. The only difference is that it is executed by the communication network device.

[0157] In this optional embodiment, the first motion state observation value sensed by the sensor of the first mobile terminal can be sensed by the first mobile terminal and sent to the communication network device, and the communication network device calculates the first position of the first mobile terminal based on the first motion state observation value sensed by the sensor. The second position of the second mobile terminal can be sensed by the second terminal itself, or the second mobile terminal can sense the second motion state observation value sensed by its own sensor and send it to the communication network device, and the communication network device calculates the second position based on the second motion state observation value sensed by the sensor.

[0158] The process of adjusting the motion trajectory of the first mobile terminal is described in detail below:

[0159] In the dynamic environment of the non-assembly area, the AGV will run in a stable state along the circular assembly line, and there are no robot assembly stations and other obstacles between any AGVs. Therefore, the AGV in this area mainly uses the active local control decision-making method, takes other AGVs as obstacle avoidance targets, and optimizes its own trajectory.

[0160] In an optional embodiment, the first motion state observation value sensed by the sensor includes a first position, a first steering angle, a current speed and a current motion trajectory of the first mobile terminal in the workshop;

[0161] Adjusting the motion trajectory of the first mobile terminal in a first motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal includes:

[0162] The motion trajectory of the first mobile terminal within the target motion cycle is obtained based on the current speed, the acceleration of the first mobile terminal, the duration of the first motion cycle, the first position, the first steering angle, the current motion trajectory and the first duration.

[0163] That is, when the AGV uploads the perception information to the PLC device next time (i.e., before the start of the target motion cycle to the end of the target motion cycle), the motion trajectory expression of the AGV active closed-loop control is:

[0164] X n (t+T′ ns )=X n (t)+T′ ns ·J'(t)·V n (t)

[0165] in, are binary discriminant factors of the changes in the XY coordinates (i.e., the first position) and steering angle (i.e., the first steering angle) trajectories of the AGV, T n ' s For AGV nThe closed-loop interaction period (i.e., the duration of the target motion period), V n (t) = [v n (t) T′ ns / 2·(dv n (t) / dt)0] -1 , where v n (t) is the speed of AGVn at time t (which can be understood as the current time), where T′ ns / 2·(dv n (t) / dt) is a function related to acceleration.

[0166] The following is a detailed description of the optimization of the motion trajectory, that is, the method for determining the acceleration:

[0167] Furthermore, the method further comprises:

[0168] According to the first time length, a time-related stability coefficient is obtained;

[0169] The acceleration of the first mobile terminal is obtained according to the communication delay, the safety distance, the maximum speed of the first mobile terminal and the stability coefficient.

[0170] It should be noted that during the active closed-loop control of AGV, due to the lack of global information of PLC for trajectory calibration, it is necessary to design a trajectory optimization function based on the closed-loop interaction cycle to calculate the control instructions and achieve rapid environmental response with low communication overhead.

[0171] Its trajectory optimization equation is:

[0172]

[0173] in, is the end-to-end delay (i.e., communication delay) during the AGV closed-loop communication process, κ = 1 / T ns ' is the time-dependent stability coefficient, T ns ' represents the duration of the target motion cycle, V(Δx n-1 (t)) is the velocity function, that is, V(Δx n-1 (t)) = (Vmax / 2) · [tanh(Δx n-1 (t)-H c )+tanh(H c )]. Where H c is the safe distance between AGVs, and Vmax represents the maximum moving speed of AGV. It is assumed that the acceleration remains constant during the closed-loop interaction cycle.

[0174] Expanding the equation and combining similar terms, we can get the final form of the AGV trajectory optimization function under the influence of the closed-loop interaction cycle:

[0175]

[0176] In the formula,

[0177]

[0178] If the AGV can run smoothly under offline active control and wait for the next closed-loop interaction cycle, upload the perception data to the PLC, receive control decisions through the downlink, adjust the motion state to run smoothly, the networked multi-node control with active and passive control coordination has been completed.

[0179] It should be noted that when the AGV senses that it is in the assembly area, it will also choose to actively receive PLC control decisions to adjust its own motion trajectory and synchronize with the robot arm with high-precision control. Therefore, in this area or in the turning area, the AGV mainly adopts the mode of passively receiving PLC control decisions, and moves slowly and highly accurately in a cyclic step-by-step manner to ensure efficient operation of the production process.

[0180] In an optional embodiment, receiving a first control instruction sent by a communication network device includes:

[0181] According to the first sensor data on the first mobile terminal, the first motion state observation value sensed by the sensor of the first mobile terminal in the workshop is sensed; specifically, at the tth moment, the AGV n By transmitting the sensed data (first sensed data) in an uplink, obtaining a first observation value according to the first sensed data, and transmitting the first observation value to the PLC device of the edge server, that is, sending the first motion state observation value sensed by the sensor to the communication network device, wherein the first motion state observation value sensed by the sensor includes the current motion trajectory of the first mobile terminal;

[0182] Receive a first control instruction sent by the communication network device, the first control instruction includes a second motion trajectory, the second motion trajectory is obtained by the communication network device according to a trajectory error, and the trajectory error is an error between the current motion trajectory and a preset motion trajectory; specifically, the PLC device is based on the current AGV n The motion trajectory and the reference trajectory (preset motion trajectory or pre-stored motion trajectory) are used to calculate the trajectory error. Calibrate and compensate for the motion trajectory deviation and generate a new round of vehicle trajectory decision X n (t), which is the second motion trajectory. Then the result is packaged into the first control instruction u n (t) is sent to AGV via downlink n .

[0183] The motion trajectory of the first mobile terminal within the target motion cycle is adjusted according to the second motion trajectory, the preset duration, the first travel angle and the current trajectory.

[0184] The first sensing data or the first motion state observation value sensed by the sensor includes the coordinates (x n (t),y n (t)), direction of travel θ n (t) (or called travel angle), etc. Considering the influence of perception error on control parameters, the trajectory error calculated by the PLC device This includes perception errors. n After receiving the new first control instruction, the new driving trajectory is X n (t+1), that is

[0185] X n (t+1)=X n (t)+T s ·J·u n (t)

[0186] Among them, T s It is the duration of the interval between two PLC receiving control instructions, that is, the duration of the target motion cycle. The AGV is calculated by PLC based on the reference trajectory and trajectory error. n Trajectory control function. J is the transposed matrix, and the expression is as follows:

[0187]

[0188] Afterwards, the first mobile terminal travels within the target motion cycle according to the new form trajectory.

[0189] The following specifically describes the matching process of the upper bound of the closed-loop interaction period provided by the embodiment of the present invention:

[0190] In an optional embodiment, the method further includes:

[0191] The first duration is determined based on the safety distance, the current speed of the first mobile terminal and the closed-loop communication delay.

[0192] It should be noted that in the closed-loop interaction process of intelligent machines, it was found through analysis that the ability of the wireless network to respond to AGV motion control requests efficiently with low latency and the ability of the control system to safely and stably control intelligent machines are mutually influential. For example, when the AGV in a stable system state is moving in a straight line along a flat assembly line, if the AGV continuously uploads K times of perception and task-related data, it will not change the control decision of the PLC downlink transmission. At this time, the AGV system is still in a stable state. Then the data packets within these interaction cycles are worthless, and there is no need to transmit them to increase communication overhead and closed-loop delay. Based on this concept, the present invention ensures that the system is in a stable state by adaptively increasing the closed-loop interaction cycle of the AGV and reducing the number of business requests per unit time of the AGV to support networked control of more AGVs.

[0193] Generally speaking, the stability of the AGV control system refers to whether the motion state of the AGV in a system equilibrium state can automatically adjust back to the initial equilibrium state after being disturbed. If the system is unstable, the disturbance will spread and eventually evolve into AGV blockage, collision and other phenomena. If the system is stable, the disturbance will gradually decrease and disappear during the propagation process, and the vehicle spacing will continue to maintain the optimal distance after fluctuating for a period of time.

[0194] Therefore, the present invention uses the perturbation method to analyze the stability of the AGV control system under the influence of external interference. Assume that at time t, each group of AGVs travels along a straight line at a constant safety distance H0, and the travel speed is the optimal speed V(H0). At this time, each AGV in the AGV control system is in a completely stable state. At time t, AGV n The initial position is get:

[0195]

[0196] At the start time, add a small disturbance y to the nth AGV on any assembly line within the base station range. n (t), which causes the AGV trajectory to fluctuate and propagate backwards. The trajectory deviation is recorded as:

[0197] x n (t) = n·H′0 + V(H′0)t + y n (t)

[0198]

[0199] Expand the disturbance term, that is, y n (t) = exp(nki + zt), substitute it into the above formula, simplify and combine like terms to get the polynomial about z:

[0200] (1-γe (ik) )2 +[α-β(e (ik) -1)]z-αV'(H0)(e (ik) -1)=0

[0201] E ik Taylor expansion is:

[0202] e ik =1+ik+0.5×(ik) 2

[0203] z=z1(ik)+z2(ik) 2 Substitute e ik , the square term coefficient of ik is obtained as:

[0204]

[0205] If z2≥0, the AGV control system will be in a stable state.

[0206] Based on the stability coefficient (i.e. the time-dependent stability coefficient mentioned above) κ = 1 / T n ' s , the stability condition of the unified model is obtained as:

[0207]

[0208] in, It is represented by the reference end-to-end delay (i.e., closed-loop communication delay) when AGV performs industrial tasks with different cycle times in the 3GPP standard, and V represents the current speed of the first mobile terminal (i.e., current speed). Let z2 = 0, and obtain the critical stability condition for the model to maintain stable operation under disturbance. Since the control feedback delay of AGV is much lower than the closed-loop delay and can be ignored, the upper bound of the closed-loop interaction period in the stable state is:

[0209]

[0210] It can be seen that During this time, the AGV system can remain stable, and the perception data uploaded by the AGV multiple times during this time will not affect the control decision of the edge server. Therefore, the upper bound of the closed-loop interaction period in the above formula is It represents the maximum closed-loop interaction period that the AGV can tolerate under the condition of system stability. That is, the AGV can ensure the maximum duration of control system stability without sending perception data and receiving control data requests to the server, which is the first duration.

[0211] Finally, the comparison between the duration of the traditional closed-loop interaction cycle and the duration of the closed-loop interaction cycle obtained by the present invention (i.e., the first duration) is as follows: Figure 4 As shown. Figure 4 It can be seen that compared with the transmission mechanism of traditional WIFI, Zigbee and other 802.11P protocols, the closed-loop interaction cycle of the present invention is much longer than the fixed closed-loop interaction cycle in the traditional transmission mechanism. This is because the transmission method analyzes the influence of cycle time, safety interval distance, driving speed and other parameters on control stability, and designs a more flexible and less redundant adaptive closed-loop interaction cycle transmission scheme. Especially during the movement of AGV in the non-assembly area, the closed-loop interaction cycle of the system's stable state communication with the base station is significantly increased, which reduces the number of AGV visits to the base station and the communication overhead, and expands the number of controllable AGVs in the network.

[0212] The following is a comparative description of the performance of the multi-intelligent machine closed-loop control method with integrated communication, perception and control in an embodiment of the present invention:

[0213] The comparison results of the number of AGVs that can be controlled by different solutions are as follows: Figure 5 As shown, Figure 5 The performance of the multi-AGV control system of the traditional AGV control scheme based on Nmax outage tolerance (Traditional algorithm with Nmax Outage tolerance, TANOT), the traditional AGV control scheme with integrated interawareness (Classic ISAC algorithm, CISACA), the traditional networked control AGV scheme (Traditional algorithm, TA) and the design scheme of the present invention is compared. As can be seen from the figure, when the AGV performs an industrial task with a reference interaction cycle of 4ms, the maximum number of AGVs supported by the design scheme of the present invention is 100, which is much larger than the maximum number of AGVs supported by TANOT, CISACA and TA {60, 23 and 20}. This is because TANOT ensures the stability of the AGV control system by calculating the number of AGV interruption retransmissions under stable conditions, aiming to reduce the data access of AGV to the base station. However, it ignores the impact of perception and control errors on communication delay during each interruption and reconnection process, and has a weak effect on network load relief. Although CISACA can reduce perception and control errors through active perception, it cannot effectively solve the problems of data packet conflicts and large queuing delays caused by large-scale AGVs accessing base stations at high frequencies, and cannot optimize the number of AGVs. Due to the limited preamble code, TA cannot support large-scale AGV access to the base station, resulting in problems such as packet conflict and large queuing delay, and the number of AGVs supported is also small. In contrast, the design scheme of the present invention comprehensively considers the impact of perception error and control error on communication transmission, calculates the upper limit of the closed-loop interaction cycle in the stable state of the AGV control system, and adaptively adjusts the closed-loop interaction cycle to enable more AGVs to participate in industrial tasks, effectively reducing network load pressure.

[0214] Table 1 below shows a comparison of the production efficiency performance and the maximum supported travel speed of traditional solutions such as TANOT, CISACA and TA and the design of the present invention.

[0215] Table 1 Comparison of production efficiency performance

[0216]

[0217] As can be seen from the table, the network performance of the design scheme of the present invention is better, and the number and speed of AGVs that can be controlled by the wireless network are much greater than other schemes. This is because TANOT focuses on analyzing the impact of wireless network link interruption on control stability. Although it can reduce the number of AGV visits to the base station to increase the number of AGVs that the network can support, it ignores the safety spacing and emergency braking emergencies. Once the AGV is disturbed by the outside world during each link interruption, the AGV speed will be out of control, which will increase the probability of collision congestion, thereby limiting the number of AGVs and travel speed that the system can support. CISACA can reduce latency and errors through active perception, but it cannot effectively solve the problems of packet conflicts and high queuing delays caused by a large number of AGVs visiting the base station at a high frequency. It cannot optimize the number of AGVs, and the system design mechanism does not pay attention to the impact of AGV speed on communication and control systems, and cannot further effectively optimize the speed of AGVs. At this time, the too slow travel speed leads to reduced production capacity. Similarly, the limited preamble code under the transmission mechanism of the fixed closed-loop interaction cycle cannot support a large number of AGVs to access the base station, which leads to problems such as packet conflicts and high queuing delays. The number of AGVs that can be supported is small, and the AGV speed cannot be optimized. However, the design scheme of the present invention comprehensively considers the impact of the communication-sensing-control closed-loop process on network performance, adaptively adjusts the closed-loop interaction cycle duration, and reduces the number of AGV visits to the base station and communication overhead. At the same time, the number of AGVs that can be controlled by the network is increased to ensure that the wireless communication system can control large-scale AGVs in a networked manner. On the other hand, the impact of parameters such as safety spacing and closed-loop interaction cycle on AGV control performance is comprehensively considered, and a networked multi-node control method with active and passive control coordination is designed to assist AGVs in different assembly areas, intelligently select active local decisions and passively receive PLC control decisions for trajectory control, reduce the probability of collision congestion, and ensure the safety and stability of the AGV control system. In addition, it can be seen from the table that the capacity of the design scheme of the present invention is 15.33JPH, while the capacities of TANOT, CISACA and TA methods are 9.43JPH, 3.41JPH and 2.90JPH respectively. The capacity of the design scheme of the present invention is improved by at least 62.6% compared with the latter, verifying the effectiveness of the scheme of this article. In summary, the simulation performance analysis proves that the speed and number of AGVs are the key parameters affecting the capacity of the general assembly workshop, and the capacity can be effectively improved by adjusting the speed and number of AGVs.

[0218] Figure 6 The closed-loop communication delays of the four schemes are compared. Figure 6 It can be seen that the closed-loop delay of the design scheme of the present invention can converge at 600ms and keep the system stable, which is much lower than 1600ms of TANOT, 2600ms of CISACA and 2000ms of TA. This is because TANOT only optimizes the data retransmission process based on the communication parameter SNR threshold, and reduces the closed-loop delay by slightly increasing the control instruction step size. It cannot effectively deal with a series of error changes caused by external interference, resulting in a slow convergence speed. CISACA can reduce some errors through active perception, but cannot quickly perceive and reduce the new control errors generated by the AGV system under interference, resulting in a slow convergence speed. Similarly, TA cannot cope with a series of error changes caused by external interference, resulting in a slow convergence speed. However, when all AGVs are disturbed, the design scheme of the present invention can quickly optimize the trajectory locally on the AGV, adaptively adjust the closed-loop interaction cycle, reduce the number of user visits to the base station per unit time, reduce data packet conflicts and queuing delays, further reduce the impact of closed-loop delay on the stability coefficient, and accelerate the stability convergence speed, verifying that the scheme proposed in this paper has better anti-interference performance.

[0219] In addition, from Figure 6 It can be seen that the closed-loop delay of the design of the present invention is much lower than that of the other three schemes. This is because this scheme calculates the upper limit of the closed-loop interaction cycle from the communication level by designing a cooperative system of inter-sensing control, and adopts a more flexible transmission scheme of adaptive closed-loop interaction cycle. Then the number of retransmissions and queuing delays are improved, and from the level of motion state control and environmental information perception, active local decisions and passive reception of PLC control decisions are intelligently selected to reduce closed-loop delays with low communication overhead. The other three schemes respectively integrate communication control, communication perception, and a single communication perspective, and separately design redundant and more demanding optimization schemes, and also ignore the mutual influence between the number of AGVs and the travel speed and the communication delay, and thus cannot effectively reduce the closed-loop delay. In addition, it can be seen from the figure that the closed-loop delay of the design scheme of the present invention is 1.3952ms, while the closed-loop delays of TANOT, CISACA and TA are 1.8811ms, 4.4413ms and 7.8727ms respectively. Compared with the performance of other classic schemes, the closed-loop delay of this scheme is improved by 25.83%, 68.59% and 82.28% respectively. It further verifies that the closed-loop delay performance improvement effect of the design scheme of the present invention is the best, and the low-latency real-time control of AGV is realized.

[0220] In summary, the present invention can be applied to the field of closed-loop control, which refers to the closed-loop control business with highly differentiated time and space such as the closed-loop control of intelligent machine control and the closed-loop management of the whole life cycle of the production line in the intelligent manufacturing workshop. The process involves a large number of communication, perception and control processes between sensing devices, edge servers and control devices, which requires flexible and efficient closed-loop control. It can also be applied to the field of integrated sensing, communication, and control (ISCC), which refers to the unified design of communication, perception and control functions through the joint design of linear / nonlinear controllers, industrial wireless network protocols, networked collaborative perception and other means, so that the wireless network can realize high-precision and refined perception functions while performing high-quality communication interaction, as well as real-time and efficient control performance, and realize the improvement of the overall performance and business capabilities of the control system of the manufacturing workshop. It can also be applied to the field of wireless network control, which refers to combining wireless communication and automatic control technology, using wireless networks to realize real-time monitoring and control of production lines, robots and automatic guided vehicles (AGVs), and improve the production efficiency of intelligent manufacturing.

[0221] The multi-intelligent machine closed-loop control method with integrated communication, perception and control provided by the present invention is a closed-loop control solution that can stably and efficiently control large-scale AGV high-speed movement through networking, and ensure the quality of communication services while optimizing control performance. The protection point is to integrate perception, communication and control functions, jointly design a closed-loop control method for multiple AGVs in a flexible manufacturing workshop, reduce communication overhead, and ensure control system stability and production capacity.

[0222] like Figure 7 As shown, the embodiment of the present invention also provides a multi-intelligent machine closed-loop control device with integrated communication, perception and control, the device comprising: a control module 701;

[0223] The control module 701 is used for:

[0224] When the first mobile terminal meets the first condition, receiving a first control instruction sent by the communication network device, wherein the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration;

[0225] In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located;

[0226] The first condition includes at least one of the following:

[0227] The first mobile terminal is located in an assembly area in a workshop;

[0228] The first mobile terminal is located in a turning area on an assembly line in a workshop.

[0229] Optionally, the device further comprises:

[0230] A first sending module, configured to send a third motion state observation value sensed by a sensor of the first mobile terminal to the communication network device;

[0231] The first receiving module is used to receive a second control instruction sent by the communication network device, the second control instruction is sent by the communication network device when it is determined that the difference between the motion trajectory of the first mobile terminal and the preset motion trajectory is greater than a preset threshold based on the third motion state observation value sensed by the sensor and the preset motion trajectory, and the second control instruction is used to instruct the first mobile terminal to make adjustments according to the preset motion trajectory.

[0232] Optionally, the device further comprises:

[0233] A first processing module, configured to sense a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal;

[0234] A second processing module, configured to determine whether the first mobile terminal satisfies the first condition according to the first motion state observation value sensed by the sensor and pre-stored workshop map data;

[0235] A third processing module, configured to determine that the first mobile terminal satisfies the first condition if the condition is satisfied;

[0236] The second receiving module is used to receive indication information sent by the communication network device when the first mobile terminal does not meet the first condition, and the indication information is used to indicate whether the first mobile terminal meets the first condition.

[0237] Optionally, the second receiving module includes:

[0238] A first processing unit, configured to sense a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop;

[0239] A first sending unit, configured to send the first motion observation value to the communication network device;

[0240] A first receiving unit, configured to receive the indication information sent by the communication network device;

[0241] Among them, the communication network device is used to receive second sensor data sent by the sensor in the workshop, determine the third sensor data of the first mobile terminal observed in the second sensor data, determine the fourth motion state observation value perceived by the sensor of the first mobile terminal in the third sensor data, and process the fourth motion state observation value perceived by the sensor to obtain the fifth motion state observation value perceived by the sensor, and estimate the target motion state observation value of the first mobile terminal based on the first motion state observation value perceived by the sensor and the fifth motion state observation value perceived by the sensor, and determine whether the first mobile terminal meets the first condition based on the target motion state observation value and pre-stored workshop map data.

[0242] Optionally, the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop and a collision warning threshold factor; the second motion state observation value sensed by the sensor includes a second position of the second mobile terminal in the workshop;

[0243] The control module 701 includes:

[0244] a second processing unit, configured to obtain a safety distance according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, the collision warning threshold factor, and a first radius, wherein the first radius is a radius of an inscribed circle of a projection profile of the first mobile terminal on the ground;

[0245] a third processing unit, configured to obtain a collision distance according to the collision warning threshold factor and the safety distance;

[0246] a fourth processing unit, configured to control the first mobile terminal to accelerate within the target motion cycle when the Euclidean distance between the first position and the second position is greater than the safety distance;

[0247] a fifth processing unit, configured to control the first mobile terminal to decelerate within the target motion cycle when the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance;

[0248] The sixth processing unit is used to control the first mobile terminal to perform emergency braking within the target motion cycle when the Euclidean distance between the first position and the second position is less than the collision distance.

[0249] Optionally, the control module 701 includes:

[0250] a second sending unit, configured to send, after a first motion cycle ends, a first motion state observation value sensed by a sensor of the first mobile terminal to the communication network device, wherein the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop and a collision warning threshold factor, wherein the first motion cycle is a motion cycle adjacent to a motion cycle before the target motion cycle;

[0251] A second receiving unit, configured to receive a first control instruction sent by the communication network device;

[0252] Wherein, when the Euclidean distance between the first position and the second position is greater than the safety distance, the first control instruction is used to control the first mobile terminal to accelerate within the target motion cycle;

[0253] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, the first control instruction is used to control the first mobile terminal to decelerate within the target motion cycle;

[0254] In a case where the Euclidean distance between the first position and the second position is less than the collision distance, the first control instruction is used to control the first mobile terminal to perform emergency braking within the target motion cycle;

[0255] The safety distance is obtained by the communication network device according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, the collision warning threshold factor and a first radius, and the first radius is the radius of the inscribed circle of the projection outline of the first mobile terminal on the ground;

[0256] The collision distance is obtained according to the collision warning threshold factor and the safety distance;

[0257] The second location is sent by the second mobile terminal to the communication network device.

[0258] Optionally, the first motion state observation value sensed by the sensor includes a first position, a first steering angle, a current speed and a current motion trajectory of the first mobile terminal in the workshop;

[0259] The control module 701 includes:

[0260] The seventh processing unit is used to obtain the motion trajectory of the first mobile terminal within the target motion cycle based on the current speed, the acceleration of the first mobile terminal, the duration of the first motion cycle, the first position, the first steering angle, the current motion trajectory and the first duration.

[0261] Optionally, the control module 701 further includes:

[0262] an eighth processing unit, configured to obtain a time-related stability coefficient according to the first duration;

[0263] The ninth processing unit is used to obtain the acceleration of the first mobile terminal according to the communication delay, the safety distance, the maximum speed of the first mobile terminal and the stability coefficient.

[0264] Optionally, the control module 701 includes:

[0265] A tenth processing unit, configured to sense a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal;

[0266] A third sending unit is used to send the first motion state observation value sensed by the sensor to the communication network device, where the first motion state observation value sensed by the sensor includes a current motion track of the first mobile terminal;

[0267] A third receiving unit is used to receive a first control instruction sent by the communication network device, where the first control instruction includes a second motion trajectory, where the second motion trajectory is obtained by the communication network device according to a trajectory error, and the trajectory error is an error between the current motion trajectory and a preset motion trajectory;

[0268] An eleventh processing unit is used to adjust the motion trajectory of the first mobile terminal within the target motion cycle according to the second motion trajectory, the preset duration, the first travel angle and the current trajectory.

[0269] Optionally, the device further comprises:

[0270] The fourth processing module is used to determine the first duration according to the safety distance, the current speed of the first mobile terminal and the closed-loop communication delay.

[0271] It should be noted that the embodiments of the present invention provide Figure 7 The multi-intelligent machine closed-loop control device with integrated communication perception control shown is a device capable of executing the above-mentioned multi-intelligent machine closed-loop control method with integrated communication perception control. All embodiments of the above-mentioned multi-intelligent machine closed-loop control method with integrated communication perception control are applicable to the device and can achieve the same or similar technical effects.

[0272] like Figure 8 As shown, an embodiment of the present invention also provides a multi-intelligent machine closed-loop control device with integrated communication, perception and control, including: a processor 801; and a memory 803 connected to the processor 801 via a bus interface 802, wherein the memory 803 is used to store programs and data used by the processor 801 when performing operations, and the processor 801 calls and executes the programs and data stored in the memory 803.

[0273] The transceiver 804 is connected to the bus interface 802 and is used to receive and send data under the control of the processor 801. Specifically, the processor 801 is used to read the program in the memory 803, and the transceiver 804 performs the following process:

[0274] When the first mobile terminal meets the first condition, receiving a first control instruction sent by the communication network device, wherein the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration;

[0275] The processor 801 performs the following processes:

[0276] In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located;

[0277] The first condition includes at least one of the following:

[0278] The first mobile terminal is located in an assembly area in a workshop;

[0279] The first mobile terminal is located in a turning area on an assembly line in a workshop.

[0280] Optionally, the transceiver 804 is further configured to:

[0281] Sending a third motion state observation value sensed by a sensor of the first mobile terminal to the communication network device;

[0282] Receive a second control instruction sent by the communication network device, wherein the second control instruction is sent by the communication network device when it is determined that the difference between the motion trajectory of the first mobile terminal and the preset motion trajectory is greater than a preset threshold based on the third motion state observation value sensed by the sensor and the preset motion trajectory, and the second control instruction is used to instruct the first mobile terminal to make adjustments according to the preset motion trajectory.

[0283] Optionally, the processor 801 is further configured to:

[0284] sensing a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal;

[0285] Determining whether the first mobile terminal meets the first condition according to the first motion state observation value sensed by the sensor and the pre-stored workshop map data;

[0286] If so, determining that the first mobile terminal meets the first condition;

[0287] The transceiver 804 is also used for:

[0288] In a case where the first mobile terminal does not satisfy the first condition, indication information sent by the communication network device is received, where the indication information is used to indicate whether the first mobile terminal satisfies the first condition.

[0289] Optionally, the processor 801 is specifically configured to:

[0290] Perceiving a first motion state observation value perceived by a sensor of the first mobile terminal in the workshop;

[0291] The transceiver 804 is specifically used for:

[0292] Sending the first motion observation value to the communication network device;

[0293] Receiving the indication information sent by the communication network device;

[0294] Among them, the communication network device is used to receive second sensor data sent by the sensor in the workshop, determine the third sensor data of the first mobile terminal observed in the second sensor data, determine the fourth motion state observation value perceived by the sensor of the first mobile terminal in the third sensor data, and process the fourth motion state observation value perceived by the sensor to obtain the fifth motion state observation value perceived by the sensor, and estimate the target motion state observation value of the first mobile terminal based on the first motion state observation value perceived by the sensor and the fifth motion state observation value perceived by the sensor, and determine whether the first mobile terminal meets the first condition based on the target motion state observation value and pre-stored workshop map data.

[0295] Optionally, the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop and a collision warning threshold factor; the second motion state observation value sensed by the sensor includes a second position of the second mobile terminal in the workshop;

[0296] The processor 801 is specifically configured to:

[0297] Obtaining a safety distance according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, the collision warning threshold factor and a first radius, wherein the first radius is a radius of an inscribed circle of a projection profile of the first mobile terminal on the ground;

[0298] Obtaining a collision distance according to the collision warning threshold factor and the safety distance;

[0299] When the Euclidean distance between the first position and the second position is greater than the safety distance, controlling the first mobile terminal to accelerate within the target motion cycle;

[0300] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, controlling the first mobile terminal to decelerate within the target motion cycle;

[0301] When the Euclidean distance between the first position and the second position is less than the collision distance, the first mobile terminal is controlled to perform emergency braking within the target motion cycle.

[0302] Optionally, the transceiver 804 is specifically configured to:

[0303] After the first motion cycle ends, sending a first motion state observation value sensed by a sensor of the first mobile terminal to the communication network device, wherein the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop and a collision warning threshold factor, wherein the first motion cycle is a motion cycle adjacent to the target motion cycle;

[0304] Receiving a first control instruction sent by the communication network device;

[0305] Wherein, when the Euclidean distance between the first position and the second position is greater than the safety distance, the first control instruction is used to control the first mobile terminal to accelerate within the target motion cycle;

[0306] When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, the first control instruction is used to control the first mobile terminal to decelerate within the target motion cycle;

[0307] In a case where the Euclidean distance between the first position and the second position is less than the collision distance, the first control instruction is used to control the first mobile terminal to perform emergency braking within the target motion cycle;

[0308] The safety distance is obtained by the communication network device according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, the collision warning threshold factor and a first radius, and the first radius is the radius of the inscribed circle of the projection outline of the first mobile terminal on the ground;

[0309] The collision distance is obtained according to the collision warning threshold factor and the safety distance;

[0310] The second location is sent by the second mobile terminal to the communication network device.

[0311] Optionally, the first motion state observation value sensed by the sensor includes a first position, a first steering angle, a current speed and a current motion trajectory of the first mobile terminal in the workshop;

[0312] The processor 801 is specifically configured to:

[0313] The motion trajectory of the first mobile terminal within the target motion cycle is obtained based on the current speed, the acceleration of the first mobile terminal, the duration of the first motion cycle, the first position, the first steering angle, the current motion trajectory and the first duration.

[0314] Optionally, the processor 801 is further configured to:

[0315] According to the first time length, a time-related stability coefficient is obtained;

[0316] The acceleration of the first mobile terminal is obtained according to the communication delay, the safety distance, the maximum speed of the first mobile terminal and the stability coefficient.

[0317] Optionally, the processor 801 is configured to:

[0318] sensing a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal;

[0319] The transceiver 804 is used for:

[0320] Sending a first motion state observation value sensed by the sensor to the communication network device, where the first motion state observation value sensed by the sensor includes a current motion track of the first mobile terminal;

[0321] Receiving a first control instruction sent by the communication network device, the first control instruction includes a second motion trajectory, the second motion trajectory is obtained by the communication network device according to a trajectory error, and the trajectory error is an error between the current motion trajectory and a preset motion trajectory;

[0322] The processor 801 is used for:

[0323] The motion trajectory of the first mobile terminal within the target motion cycle is adjusted according to the second motion trajectory, the preset duration, the first travel angle and the current trajectory.

[0324] Optionally, the processor 801 is further configured to:

[0325] The first duration is determined according to the safety distance, the current speed of the first mobile terminal and the closed-loop communication delay.

[0326] Among them, Figure 8 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 801 and memory represented by memory 803. The bus architecture may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides a user interface 805. The transceiver 804 may be a plurality of components, i.e., including a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. The processor 801 is responsible for managing the bus architecture and general processing, and the memory 803 may store data used by the processor 801 when performing operations.

[0327] A readable storage medium of an embodiment of the present invention stores a program or instruction thereon. When the program or instruction is executed by a processor, the steps in the multi-intelligent machine closed-loop control method with integrated communication, perception and control as described above are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0328] In the embodiment of the present invention, module can be implemented with software so that it can be executed by various types of processors. For example, an executable code module of an identification can include one or more physical or logical blocks of computer instructions, for example, it can be constructed as an object, process or function. Nevertheless, the executable code of the identified module does not need to be physically located together, but can include different instructions stored in different positions, and when these instructions are logically combined together, it constitutes a module and realizes the specified purpose of the module.

[0329] In fact, executable code module can be a single instruction or many instructions, and can even be distributed on a plurality of different code segments, distributed among different programs, and distributed across a plurality of memory devices. Similarly, operating data can be identified in the module, and can be implemented and organized in the data structure of any appropriate type according to any appropriate form. The operating data can be collected as a single data set, or can be distributed in different locations (including on different storage devices), and can only be present on a system or network as an electronic signal at least in part.

[0330] When a module can be implemented by software, considering the level of existing hardware technology, a person skilled in the art can build a corresponding hardware circuit to implement the corresponding function of the module that can be implemented by software without considering the cost. The hardware circuit includes a conventional very large scale integration (VLSI) circuit or gate array and existing semiconductors such as logic chips, transistors, or other discrete components. The module can also be implemented by a programmable hardware device, such as a field programmable gate array, a programmable array logic, a programmable logic device, etc.

[0331] The above exemplary embodiments are described with reference to the accompanying drawings, and many different forms and embodiments are feasible without departing from the spirit and teachings of the present invention. Therefore, the present invention should not be constructed as a limitation of the exemplary embodiments proposed herein. More specifically, these exemplary embodiments are provided so that the present invention will be perfect and complete, and the scope of the present invention will be conveyed to those who are familiar with the technology. In these figures, the component sizes and relative sizes may be exaggerated for clarity. The terms used here are only based on the purpose of describing specific exemplary embodiments and are not intended to be limiting. As used herein, unless the text clearly indicates otherwise, the singular forms "one", "an" and "the" are intended to include these multiple forms. It will be further understood that the terms "including" and / or "comprising" when used in this specification indicate the presence of the features, integers, steps, operations, components and / or components, but do not exclude the presence or increase of one or more other features, integers, steps, operations, components, components and / or their groups. Unless otherwise indicated, when stated, a range of values ​​includes the upper and lower limits of that range and any subranges therebetween.

[0332] A specific embodiment of the present invention further provides a computer program product, including computer instructions, which, when executed by a processor, implement the above Figure 1 The various processes of the method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0333] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A closed-loop control method for multi-intelligent machines integrating communication, perception and control, characterized in that: Applied to a first mobile terminal, the method includes: When the first mobile terminal meets the first condition, receiving a first control instruction sent by a communication network device, wherein the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration; In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located; The first condition includes at least one of the following: The first mobile terminal is located in an assembly area in a workshop; The first mobile terminal is located in a turning area on an assembly line in a workshop.

2. The method according to claim 1, characterized in that After adjusting the motion trajectory and motion speed of the first mobile terminal within the target motion cycle according to the first motion state observation value sensed by the sensor of the first mobile terminal and / or the second motion state observation value sensed by the sensor of the second mobile terminal, the method further includes: Sending a third motion state observation value sensed by a sensor of the first mobile terminal to the communication network device; Receive a second control instruction sent by the communication network device, wherein the second control instruction is sent by the communication network device when it is determined that the difference between the motion trajectory of the first mobile terminal and the preset motion trajectory is greater than a preset threshold based on the third motion state observation value sensed by the sensor and the preset motion trajectory, and the second control instruction is used to instruct the first mobile terminal to make adjustments according to the preset motion trajectory.

3. The method according to claim 1, characterized in that The method further comprises: sensing a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal; Determining whether the first mobile terminal meets the first condition according to the first motion state observation value sensed by the sensor and the pre-stored workshop map data; If so, determining that the first mobile terminal meets the first condition; In a case where the first mobile terminal does not satisfy the first condition, indication information sent by the communication network device is received, where the indication information is used to indicate whether the first mobile terminal satisfies the first condition.

4. The method according to claim 3, characterized in that Receiving indication information sent by the communication network device, comprising: Perceiving a first motion state observation value perceived by a sensor of the first mobile terminal in the workshop; Sending the first motion observation value to the communication network device; Receiving the indication information sent by the communication network device; Among them, the communication network device is used to receive second sensor data sent by the sensor in the workshop, determine the third sensor data of the first mobile terminal observed in the second sensor data, determine the fourth motion state observation value perceived by the sensor of the first mobile terminal in the third sensor data, and process the fourth motion state observation value perceived by the sensor to obtain the fifth motion state observation value perceived by the sensor, and estimate the target motion state observation value of the first mobile terminal based on the first motion state observation value perceived by the sensor and the fifth motion state observation value perceived by the sensor, and determine whether the first mobile terminal meets the first condition based on the target motion state observation value and pre-stored workshop map data.

5. The method according to claim 1, characterized in that The first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop; the second motion state observation value sensed by the sensor includes a second position of the second mobile terminal in the workshop; Adjusting the movement speed of the first mobile terminal in the first movement cycle according to the first movement state observation value sensed by the sensor of the first mobile terminal and / or the second movement state observation value sensed by the sensor of the second mobile terminal includes: Obtaining a safety distance according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, a preset collision warning threshold factor and a first radius, wherein the first radius is a radius of an inscribed circle of a projection profile of the first mobile terminal on the ground; Obtaining a collision distance according to a preset collision warning threshold factor and the safety distance; When the Euclidean distance between the first position and the second position is greater than the safety distance, controlling the first mobile terminal to accelerate within the target motion cycle; When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, controlling the first mobile terminal to decelerate within the target motion cycle; When the Euclidean distance between the first position and the second position is less than the collision distance, the first mobile terminal is controlled to perform emergency braking within the target motion cycle.

6. The method according to claim 1, characterized in that Receiving a first control instruction sent by a communication network device includes: After the first motion cycle ends, sending a first motion state observation value sensed by a sensor of the first mobile terminal to the communication network device, wherein the first motion state observation value sensed by the sensor includes a first position of the first mobile terminal in the workshop, wherein the first motion cycle is a motion cycle adjacent to the target motion cycle; Receiving a first control instruction sent by the communication network device; Wherein, when the Euclidean distance between the first position and the second position is greater than the safety distance, the first control instruction is used to control the first mobile terminal to accelerate within the target motion cycle; When the Euclidean distance between the first position and the second position is greater than or equal to the collision distance and less than the safety distance, the first control instruction is used to control the first mobile terminal to decelerate within the target motion cycle; In a case where the Euclidean distance between the first position and the second position is less than the collision distance, the first control instruction is used to control the first mobile terminal to perform emergency braking within the target motion cycle; The safety distance is obtained by the communication network device according to the maximum speed of the first mobile terminal, the maximum acceleration of the first mobile terminal, a preset collision warning threshold factor and a first radius, and the first radius is the radius of the inscribed circle of the projection outline of the first mobile terminal on the ground; The collision distance is obtained according to a preset collision warning threshold factor and the safety distance; The second location is sent by the second mobile terminal to the communication network device.

7. The method according to claim 1, characterized in that The first motion state observation value sensed by the sensor includes a first position, a first steering angle, a current speed, and a current motion trajectory of the first mobile terminal in the workshop; Adjusting the motion trajectory of the first mobile terminal in a first motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal includes: The motion trajectory of the first mobile terminal within the target motion cycle is obtained based on the current speed, the acceleration of the first mobile terminal, the duration of the first motion cycle, the first position, the first steering angle, the current motion trajectory and the first duration.

8. The method according to claim 7, characterized in that The method further comprises: According to the first time length, a time-related stability coefficient is obtained; The acceleration of the first mobile terminal is obtained according to the communication delay, the safety distance, the maximum speed of the first mobile terminal and the stability coefficient.

9. The method according to claim 1, characterized in that: Receiving a first control instruction sent by a communication network device includes: sensing a first motion state observation value sensed by a sensor of the first mobile terminal in the workshop according to first sensor data on the first mobile terminal; Sending a first motion state observation value sensed by the sensor to the communication network device, where the first motion state observation value sensed by the sensor includes a current motion track of the first mobile terminal; Receiving a first control instruction sent by the communication network device, the first control instruction includes a second motion trajectory, the second motion trajectory is obtained by the communication network device according to a trajectory error, and the trajectory error is an error between the current motion trajectory and a preset motion trajectory; The motion trajectory of the first mobile terminal within the target motion cycle is adjusted according to the second motion trajectory, the preset duration, the first travel angle and the current trajectory.

10. The method according to claim 1, characterized in that The method further comprises: The first duration is determined based on the safety distance, the current speed of the first mobile terminal and the closed-loop communication delay.

11. A multi-intelligent machine closed-loop control device integrating communication, perception and control, characterized in that: The device comprises: a control module; The control module is used for: When the first mobile terminal meets the first condition, receiving a first control instruction sent by the communication network device, wherein the first control instruction is used to instruct to adjust the motion trajectory and motion speed of the first mobile terminal within a target motion cycle, wherein the duration of the target motion cycle is a preset duration; In the case where the first mobile terminal does not meet the first condition, adjusting the motion trajectory and motion speed of the first mobile terminal within a target motion cycle according to a first motion state observation value sensed by a sensor of the first mobile terminal and / or a second motion state observation value sensed by a sensor of a second mobile terminal, wherein the second mobile terminal is a motion terminal adjacent to the first mobile terminal, wherein the duration of the target motion cycle is a first duration, and the first duration is determined according to the control environment state information in the workshop where the first mobile terminal is located; The first condition includes at least one of the following: The first mobile terminal is located in an assembly area in a workshop; The first mobile terminal is located in a turning area on an assembly line in a workshop.

12. A multi-intelligent machine closed-loop control device integrating communication, perception and control, comprising: A transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; characterized in that when the processor executes the program or instruction, the steps in the closed-loop control method of a multi-intelligent machine with integrated communication, perception and control as described in any one of claims 1 to 10 are implemented.

13. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by the processor, the steps in the multi-intelligent machine closed-loop control method with integrated communication, perception and control as described in any one of claims 1 to 10 are implemented.

14. A computer program product, characterized in that It includes computer instructions, which, when executed by a processor, implement the steps in the multi-intelligent machine closed-loop control method with integrated communication, perception and control as described in any one of claims 1 to 10.

Images