Multi-brand agv same field mixed running management and control method and device, electronic equipment and storage medium

Abstract

The application discloses a multi-brand AGV same-field mixed running management and control method and device, electronic equipment and a storage medium. The method comprises the following steps: establishing a standardized data channel with each brand AGV original scheduling system to obtain real-time running state data and planning path data of each AGV; based on the real-time running state data and the planning path data, spatial intersection judgment is performed on the planning paths of all AGVs, and for the AGV pairs with spatial intersection, a preliminary collision risk judgment is performed based on preset limit running parameters; for the AGV pairs with the preliminary collision risk, motion simulation is performed in a virtual environment based on actual running parameters to output a quantitative collision risk result; based on the quantitative collision risk result, a low-priority AGV in the AGV pair with the collision risk is generated with an avoidance strategy in combination with a preset dynamic priority rule, and a standardized scheduling instruction generated based on the avoidance strategy is issued and executed. The application can realize safe and efficient collaborative operation of multi-brand AGVs in a same-field mixed running environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of AGV management and control, in particular to a multi-brand AGV same-site mixed running management and control method and device, electronic equipment and storage medium. BACKGROUND

[0002] As the core equipment of intelligent factories and smart logistics, AGVs (Automated Guided Vehicles) are increasingly widely used. With the expansion of production scale and the increase of process complexity, a single brand or batch of AGVs often cannot meet all the needs. In the scenarios of production line upgrading and capacity expansion, multi-brand AGVs working collaboratively in the same physical space, i.e., "same-site mixed running", has become a realistic and urgent demand.

[0003] However, multi-brand AGV mixed running faces severe technical challenges. The core pain points are as follows: AGVs produced by different manufacturers usually adopt private underlying communication protocols, independent and closed RCSs (Robot Control Systems), and heterogeneous data standards. This "protocol island" and "system barrier" result in the inability of different brands of AGVs to realize real-time intercommunication and sharing of key information such as device status and planned path. In the complex dynamic environment of same-site mixed running and path intersection, each AGV can only rely on its own limited sensors for local obstacle avoidance, lacking global and forward-looking collaborative planning. This makes the system extremely vulnerable to the risk of path conflict, obstacle avoidance deadlock, and even physical collision at intersections, narrow passages, and other areas due to information opacity and action incoordination, seriously threatening the smoothness, safety, and overall production efficiency of material transportation.

[0004] To cope with the multi-AGV coordination problem, the prior art proposes some solutions. For example, some solutions deploy a central scheduling server to perform centralized task allocation and path planning for AGV groups of a single brand or a unified protocol, which to some extent optimizes system efficiency and reduces conflicts. However, such solutions usually cannot penetrate the underlying protocol barriers of AGVs of different brands, and are difficult to directly apply to multi-brand mixed running scenarios. Some other solutions introduce digital twin technology to monitor and simulate AGV operation by building virtual models, which improves state visualization and debugging flexibility, but their core still focuses on monitoring and post-analysis, or provides convenience in modifying scheduling instructions, and lacks systematic solutions on how to accurately quantify collision risks based on real-time simulation for forward-looking collision risk quantification, and accordingly drive dynamic avoidance scheduling across brands considering production priorities. Specifically, the existing solutions generally have the following defects when implementing multi-brand AGV mixed running management: (1) lack of efficient and universal multi-brand data integration means, which cannot achieve low-cost and rapid unified collection of heterogeneous AGV data and standardized issuance of instructions; (2) conflict prediction methods are relatively simple, either too conservative leading to low efficiency or not accurate enough leading to missed judgment risks, failing to effectively combine the advantages of rapid rough screening and high-precision calculation; (3) avoidance scheduling strategies are usually based on simple "first come, first served" or fixed priority rules, failing to deeply integrate with dynamic business logic of the production system, making it difficult to maximize production efficiency while ensuring safety.

[0005] It should be noted that the above statements are only used to provide background technical information related to the present application, and do not necessarily constitute prior art. SUMMARY

[0006] To solve the above technical problems, embodiments of the present application provide a multi-brand AGV mixed running management method, device, electronic equipment and storage medium, which can overcome the data barriers caused by protocol and system heterogeneity of multi-brand AGVs, and realize safe and efficient coordinated operation of multi-brand AGVs in the same mixed running environment through four-level coordinated management of protocol adaptation, cross-prediction, collision simulation and dynamic scheduling.

[0007] In a first aspect, embodiments of the present application provide a multi-brand AGV mixed running management method, which comprises:

[0008] establishing a standardized data channel with each brand AGV native scheduling system to obtain real-time running state data and planned path data of each AGV;

[0009] based on the real-time running state data and the planned path data, performing spatial intersection judgment on the planned paths of all AGVs, and based on the preset limit running parameters, performing collision risk preliminary judgment on AGV pairs with spatial intersection;

[0010] For the AGV pair with collision risk, motion simulation is performed in the virtual environment based on actual operation parameters to output a quantitative collision risk result; wherein the actual operation parameters are determined based on at least the real-time operation state data;

[0011] Based on the quantitative collision risk result, a low-priority AGV in the AGV pair with collision risk is generated with an avoidance strategy in combination with a preset dynamic priority rule, and a standardized scheduling instruction generated based on the avoidance strategy is issued for execution through the standardized data channel.

[0012] In some embodiments of the present application, the spatial intersection judgment on the planned paths of all AGVs includes:

[0013] In a unified plane coordinate system, the planned path of each AGV is represented as a polyline composed of a series of continuous line segments;

[0014] A line segment intersection detection algorithm based on vector cross product is used to judge whether the planned path line segments of any two AGVs exist intersection points one by one;

[0015] If there is an intersection point, the intersection point is recorded as a spatial intersection point, and the corresponding two AGVs are associated to form the AGV pair with spatial intersection.

[0016] In some embodiments of the present application, the establishment of the standardized data channel with each brand AGV native scheduling system includes:

[0017] Standardized specifications for bidirectional data interaction are established, including unified communication protocols, unified data exchange formats, and protocol conversion mechanisms;

[0018] The acquisition of real-time operation state data and planned path data of each AGV includes:

[0019] Through the protocol conversion mechanism, the standard uplink data messages reported by each native scheduling system and obtained through conversion are received;

[0020] From the standard uplink data messages, real-time position and speed fields as the real-time operation state data and planned path coordinate sequence fields as the planned path data are parsed;

[0021] The issuing for execution through the standardized data channel includes:

[0022] The avoidance strategy is encapsulated as a standard downlink instruction message conforming to the standardized specifications;

[0023] Through the protocol conversion mechanism, the standard downlink instruction message is converted into a private instruction recognizable by the corresponding native scheduling system and issued.

[0024] In some embodiments of the present application, the real-time running state data includes a current position, and the preset limit running parameter includes a preset maximum running speed and a preset maximum vehicle body length; the collision risk preliminary judgment based on the preset limit running parameter includes:

[0025] For the AGV pair with a spatial intersection, according to the preset maximum running speed and the preset maximum vehicle body length of each AGV, a time window required for each AGV to reach and completely pass through the spatial intersection point from the current position is calculated;

[0026] Based on the overlapping length of the two time windows, the collision risk of the AGV pair is classified.

[0027] In some embodiments of the present application, the actual running parameter includes a current speed and a current acceleration; for the AGV pair with a preliminary judged collision risk, motion simulation is performed in a virtual environment based on the actual running parameter to output a quantitative collision risk result, including:

[0028] The preset vehicle body geometric dimension and the preset motion constraint condition of each AGV in the AGV pair are obtained;

[0029] In the virtual environment, based on the current speed, the current acceleration, the preset vehicle body geometric dimension and the preset motion constraint condition, the motion trajectory of each AGV is simulated within the overlapping period of the time window;

[0030] The collision probability of the AGV pair is calculated based on the motion trajectory as the quantitative collision risk result for output.

[0031] In some embodiments of the present application, the collision probability of the AGV pair based on the motion trajectory includes:

[0032] By introducing a random disturbance variable conforming to the actual control logic of the AGV for the current speed, the current acceleration and the preset motion constraint condition, multiple independent motion trajectory simulations are performed;

[0033] The number of times of interference of the spatial occupation area of the AGV pair in the multiple independent motion trajectory simulations is counted;

[0034] According to the proportion of the number of times of interference to the total number of simulations, the collision probability is calculated.

[0035] In some embodiments of the present application, after the collision probability of the AGV pair based on the motion trajectory is calculated as the quantitative collision risk result for output, it further includes risk classification according to the collision probability, specifically:

[0036] when the collision probability is zero, determining that there is no collision risk;

[0037] when the collision probability is greater than zero but less than a preset probability threshold, determining that there is a low-risk collision;

[0038] when the collision probability is greater than or equal to the preset probability threshold, determining that there is a high-risk collision.

[0039] In some embodiments of the present application, based on the quantified collision risk result, a low-priority AGV in the AGV pair with a collision risk is generated an avoidance strategy in combination with a preset dynamic priority rule, including:

[0040] According to the collision risk result, the avoidance urgency of each AGV in the AGV pair is determined;

[0041] Based on the preset dynamic priority rule, the real-time task priority of each AGV in the AGV pair is calculated;

[0042] The avoidance urgency and the real-time task priority are integrated to determine the low-priority AGV in the AGV pair that needs to perform an avoidance operation.

[0043] In some embodiments of the present application, the avoidance strategy for the low-priority AGV in the AGV pair with a collision risk further includes:

[0044] According to the environmental conditions around the spatial intersection point, an instruction to control the low-priority AGV to slow down, an instruction to control the low-priority AGV to enter a preset avoidance area for parking, or an instruction to indicate the low-priority AGV to re-plan a path is selectively generated.

[0045] The second aspect of the present application provides a multi-brand AGV mixed running management and control device, the device includes:

[0046] The establishment module is configured to establish a standardized data channel with the original scheduling system of each brand AGV to obtain real-time running state data and planned path data of each AGV;

[0047] The preliminary judgment module is configured to perform spatial intersection judgment on the planned paths of all AGVs based on the real-time running state data and planned path data, and perform collision risk preliminary judgment on the AGV pair with spatial intersection based on preset limit running parameters;

[0048] The simulation module is configured to perform motion simulation in a virtual environment based on actual running parameters for the AGV pair with a collision risk after preliminary judgment, to output a quantified collision risk result; wherein the actual running parameters are determined based on at least the real-time running state data;

[0049] An avoidance module is configured to generate an avoidance strategy for the AGV with low priority in the AGV pair with collision risk based on the quantified collision risk result and in combination with preset dynamic priority rules, and to issue a standardized scheduling instruction generated based on the avoidance strategy for execution through the standardized data channel.

[0050] The third aspect of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the multi-brand AGV mixed running management and control method described in the embodiments of the present application.

[0051] The fourth aspect of the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the multi-brand AGV mixed running management and control method described in the embodiments.

[0052] In the present application, a standardized data channel is established with each brand AGV native scheduling system to obtain real-time running state data and planned path data of each AGV, effectively penetrating the private protocol and data barriers of each brand AGV, realizing the unity and real-time collection of heterogeneous device state and planning information, and laying a data foundation for global collaborative control. In the present application, based on the real-time running state data and the planned path data, spatial intersection judgment is performed on the planned paths of all AGVs, and for AGV pairs with spatial intersection, a preliminary collision risk judgment is performed based on preset limit running parameters. Through rapid and conservative extreme case rough judgment, all potential path conflicts can be identified without omission, overcoming the limitations of local obstacle avoidance and the disadvantages of global information loss in the prior art, and realizing forward-looking risk screening. Then, for the AGV pairs with collision risk after preliminary judgment, motion simulation is performed in a virtual environment based on actual running parameters to output a quantitative collision risk result; wherein the actual running parameters are determined based on at least the real-time running state data; this constitutes accurate simulation and risk quantification, which makes up for the inaccuracy of simple conflict detection methods, and through algorithms such as Monte Carlo simulation, risk assessment is improved from qualitative to quantitative, providing a scientific and accurate basis for intelligent decision-making. Finally, based on the quantitative collision risk result, combined with the preset dynamic priority rule, an avoidance strategy is generated for the low-priority AGV in the AGV pair with collision risk, and the standardized scheduling instruction generated based on the avoidance strategy is issued and executed through the standardized data channel, safely avoiding collision and deeply integrating production optimization, overcoming the negative impact of fixed priority or simple rule scheduling on production efficiency through dynamic priority and intelligent strategy selection, and ensuring reliable execution of control strategies through standardized instruction issuance and monitoring closed loop. In this way, the present application solves the problems of multi-brand AGVs being unable to cooperate due to protocol heterogeneity and system barriers, and inaccurate conflict prediction and rigid scheduling strategies in the prior art, and through a four-level collaborative control architecture of "protocol adaptation, intersection prediction, collision simulation, and dynamic scheduling", the core problem of multi-brand AGVs running together is systematically solved, and finally safe and efficient collaborative operation of multi-brand AGVs in the same environment can be realized.

[0053] The above description is only a summary of the technical solutions of the embodiments of the present application. In order to more clearly understand the technical means of the present application, the embodiments can be implemented in accordance with the contents of the specification, and in order to make the above and other purposes, features and advantages of the embodiments of the present application more obvious and easy to understand, the specific implementation manner of the present application is described below. BRIEF DESCRIPTION OF DRAWINGS

[0054] Various other advantages and benefits will become apparent to those of ordinary skill in the art, upon reading the following detailed description of the preferred embodiment. The accompanying drawings are included to provide a better understanding of the preferred embodiment, and are not intended to constrain the application. Moreover, like reference numerals denote same or similar components throughout the several views of the drawings. In the drawings:

[0055] Figure 1 A step schematic diagram of a multi-brand AGV same-field mixed running management and control method provided by some embodiments of the application;

[0056] Figure 2 A flowchart of a spatial intersection judgment on a planned path of an AGV provided by some embodiments of the application;

[0057] Figure 3 A flowchart of motion simulation in a virtual environment based on actual running parameters provided by some embodiments of the application;

[0058] Figure 4 An AGV dynamic avoidance scheduling flowchart provided by some embodiments of the application;

[0059] Figure 5 A structural schematic diagram of a multi-brand AGV same-field mixed running management and control device provided by some embodiments of the application;

[0060] Figure 6 A structural schematic diagram of an electronic device provided by some embodiments of the application. DETAILED DESCRIPTION

[0061] The embodiments of the technical solutions of the application will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the application, and therefore only serve as examples, and cannot limit the protection scope of the application.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which embodiments of the application belong; the terms used herein are only for the purpose of describing specific embodiments of the application, and are not intended to limit the embodiments of the application; the terms "include" and "have" and any variations thereof in the specification and claims of the application and the above description of drawings are intended to cover non-exclusive inclusion.

[0063] In the description of embodiments of the application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0064] Reference to“an embodiment” herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase“in an embodiment” in various places in the specification are not necessarily referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. As will be apparent to those of ordinary skill in the art, embodiments described herein can be combined with other embodiments.

[0065] In the description of the embodiments of the application, the term“and / or” only means a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character“ / ” herein generally represents that the front and rear associated objects are in an“or” relationship.

[0066] In the related art, the difficulty of multi-brand AGV mixed running management and control lies in how to penetrate the barrier of heterogeneous systems and build a precise, efficient and intelligent collaborative mechanism on top of it. In view of this, the embodiments of the present application provide a four-level collaborative management and control method, which realizes the unified, safe and efficient scheduling of multi-brand AGVs through protocol adaptation, cross prediction, collision simulation and dynamic scheduling in turn. The method can be executed through a central control platform, which is in communication connection with the original robot control system RCS of each brand AGV. The embodiments of the present application will be described in detail below in combination with the method flowchart shown in Figures 1 to 4

[0067] Figure 1 The step schematic diagram of the multi-brand AGV mixed running management and control method provided by some embodiments of the present application is shown in Figure 1 The method mainly includes the following four steps S101-S104 executed in turn.

[0068] S101, a standardized data channel is established between the AGV original scheduling system and each brand AGV to obtain the real-time running state data and planned path data of each AGV.

[0069] ​Specifically, to overcome the "information island" problem caused by the heterogeneous bottom-layer protocols and data structures of multi-brand AGVs, a standardized specification for bidirectional data interaction is established, which includes a unified communication protocol, a unified data exchange format, and a protocol conversion mechanism. The purpose is to establish a unified and reliable data foundation for all subsequent collaborative management functions. For example, the central management platform first establishes a set of standardized specifications for bidirectional data interaction, including a unified communication protocol such as MQTT (Message Queuing Telemetry Transport) and a unified data exchange format such as JSON. By converting key data fields to JSON format, it can avoid misreading information due to non-uniform data formats.

[0070] Further, based on the above standardized specification, real-time running state data and planning path data of each AGV are obtained, including receiving standard uplink data messages obtained by converting the data reported by each native scheduling system through the protocol conversion mechanism; from the standard uplink data messages, the real-time position and speed fields as real-time running state data and the planning path coordinate sequence fields as planning path data are parsed. This process realizes the unified collection and formatting of heterogeneous AGV data. The standardization data channel is executed, including encapsulating the avoidance strategy into a standard downlink instruction message conforming to the standardized specification; through the protocol conversion mechanism, the standard downlink instruction message is converted into a private instruction recognizable by the corresponding native scheduling system and is issued, thereby completing the instruction closed loop from central decision to bottom-layer device execution.

[0071] In one possible implementation, for each AGV native RCS of a different brand, a protocol conversion plug-in is deployed or configured. This plug-in has a bidirectional conversion function: on the one hand, it converts the AGV state data such as position, speed, power, task status, and planning path data reported by the native RCS based on the private protocol into standard uplink data messages conforming to the aforementioned standardized specification, and reports them to the central management platform; on the other hand, it converts the standard downlink instruction messages such as avoidance instructions issued by the central management platform into private instruction formats recognizable by the native RCS. In this way, the platform can obtain the real-time positions, speeds, and planning path coordinate sequences of all AGVs in a unified manner, and lay the foundation for subsequent issuance of standardized scheduling instructions.

[0072] In another alternative implementation, the standardization of the data channel can be achieved by constructing a unified Application Programming Interface (API) gateway. In this approach, the central management platform defines and publishes a set of unified RESTful API or WebSocket interface specifications. The native RCS systems of various brand AGVs then need to develop or configure corresponding API client adaptation modules according to these specifications. The adaptation module is responsible for encapsulating the private data model inside the RCS into a request that meets the platform API requirements, i.e., as uplink data, and parsing the API response or instructions issued by the platform, i.e., as downlink data. Compared with the protocol conversion plug-in approach, the API gateway approach focuses more on defining clear data contracts and interaction semantics at the application layer, while the protocol conversion plug-in may focus more on bridging the underlying communication protocols. In actual deployment, the two approaches can be flexibly selected or combined to achieve effective integration with heterogeneous systems, depending on the openness and technical architecture of different brand RCSs.

[0073] Specifically in terms of uplink data interaction, each native RCS periodically converts its internal private AGV state data into standard uplink data messages in JSON format through the protocol conversion plug-in, for example, once every second, and publishes them to the topic subscribed by the central management platform. The choice of a collection frequency of once every second is based on a balance between the conventional running speed of AGVs in the factory, such as 0.5-2 m / s: a too low frequency may miss the key path points when the AGV moves quickly, and a too high frequency will bring unnecessary communication and processing load. The standard uplink data message aims to provide full-dimensional state information of the AGV, and the fields parsed therefrom include not only real-time position and speed, but also AGV fault status code, complete coordinate sequence of the current planned path, and task execution progress percentage, etc., to ensure that the central management platform can fully and timely grasp the running status of each AGV.

[0074] In terms of downlink instruction interaction, the central management platform encapsulates the generated avoidance strategy into a structured standard downlink instruction message. The instruction focuses on avoidance control, with clear action requirements, such as but not limited to pause instruction, path adjustment instruction, deceleration running instruction, etc. At the same time, each downlink instruction message is attached with a high-priority instruction identifier to ensure that the receiving native RCS can prioritize processing of such avoidance instructions, avoiding conflicts with regular task scheduling instructions. After receiving the standard downlink instruction message, the protocol conversion plug-in converts it back to a private instruction format recognizable by the target RCS and issues it for execution, thus completing the closed-loop control.

[0075] S102, based on the real-time running state data and the planning path data, performing spatial intersection judgment on the planning paths of all AGVs, and performing collision risk preliminary judgment on AGV pairs with spatial intersection based on preset limit running parameters.

[0076] In a specific implementation, to comprehensively screen potential conflicts of all AGVs, first, the planning path of each AGV is represented as a polyline composed of a series of continuous line segments in a unified plane coordinate system; then, a line segment intersection detection algorithm based on vector cross product is used to efficiently and accurately judge whether the planning path line segments of any two AGVs have intersection points; if an intersection point is detected, the intersection point is recorded as a spatial intersection point, and the corresponding two AGVs are associated to form the AGV pair with spatial intersection. This spatial judgment step is the geometric basis for subsequent time analysis and risk quantification, ensuring the completeness of conflict detection.

[0077] In a possible implementation, to further evaluate the collision risk from the time dimension after identifying the spatial intersection, the real-time running state data includes the current position, and the preset limit running parameters include the preset maximum running speed and the preset maximum vehicle body length; the collision risk preliminary judgment based on the preset limit running parameters includes: for the AGV pair with spatial intersection, a conservative estimation strategy is used to calculate the time window required for each AGV to reach and completely pass through the spatial intersection point from the current position according to the preset maximum running speed and the preset maximum vehicle body length of each AGV; then, based on the overlap length of the two time windows, a preset threshold is used to classify the collision risk of the AGV pair. This step constitutes a fast and conservative risk preliminary screening layer, aiming to cover potential risks in the most extreme case with the principle of "prefer false positives to false negatives", and provides high-value targets screened for subsequent accurate simulation.

[0078] Here, reference can be made to Figure 2 , Figure 2 A specific step flowchart of AGV running path intersection pre-judgment in the embodiments of the application is shown. As shown in Figure 2 , the step starts with real-time collected AGV state and planning path data, wherein the AGV state includes its real-time position.

[0079] For each identified AGV pair, the platform will perform collision risk preliminary judgment based on preset limit running parameters. These parameters usually include the preset maximum running speed and the preset maximum vehicle body length of the AGV. The specific calculation method is as follows: for each AGV, taking its currently reported position as the starting point and the planning path as the reference, the time window required for it to travel at the highest speed from the front of the vehicle to reach the intersection point to the rear of the vehicle to completely leave the spatial intersection point is calculated.

[0080] Specifically, the time t1 at which the front end of the AGV vehicle body reaches the spatial intersection point is calculated: t1 = path distance (i.e. the distance from the current position to the intersection point) / maximum speed; and the time t2 at which the rear end of the vehicle body completely passes through the spatial intersection point is calculated: t2 = t1 + maximum vehicle body length / maximum speed; a time window (t1, t2) at which the vehicle body completely passes through the spatial intersection point is formed, without introducing acceleration and deceleration characteristics, to avoid risk misjudgment due to a low current speed, and to cover extreme cases in actual operation scenarios.

[0081] After obtaining the time windows of the two AGVs, as shown in the flow of Figure 2 , the overlap duration of the two time windows is compared, and the risk is classified according to the degree of overlap. If the overlap duration is less than a preset threshold, for example, 2 seconds, it is determined that there is a mild overlap, and the AGV pair is marked as “potential risk”. The system pushes the relevant device number, intersection point coordinates and time window information to the subsequent collision simulation link for accurate simulation calculation.

[0082] If the overlap duration is greater than or equal to the preset threshold, for example, 2 seconds, it is determined that there is a serious overlap, and the AGV pair is marked as “high-priority risk”. In addition to pushing the basic information to the collision simulation link, as shown in Figure 2 , the system triggers the priority pre-calculation process in the dynamic avoidance scheduling simultaneously, to calculate the dynamic priority for the related AGVs in advance, thereby shortening the total response time from risk identification to final scheduling decision.

[0083] S103, for the AGV pair that is initially judged to have a collision risk, motion simulation is performed in a virtual environment based on actual operation parameters to output a quantitative collision risk result; wherein the actual operation parameters are determined based on at least the real-time operation state data.

[0084] Specifically, for the AGV pair that is initially judged to have a collision risk and is pushed from step S102, the central control platform will start a high-precision motion simulation process. First, the platform needs to obtain the preset vehicle body geometric dimensions of each AGV in the AGV pair, such as the actual vehicle length, vehicle width, and preset motion constraint conditions such as maximum acceleration, maximum deceleration, and minimum turning radius. At the same time, based on the real-time operation state data obtained in step S101, the current speed and current acceleration of each AGV are determined as dynamic inputs for simulation. The preset vehicle body geometric dimensions and the preset motion constraint conditions are usually pre-stored in the AGV model library of the system.

[0085] Then, in the virtual environment, based on the current speed, the current acceleration, the preset vehicle body geometry and the preset motion constraints, the motion trajectory of each AGV is simulated within the overlapping time window. This simulation process strictly follows the kinematic model of the AGV, accurately restoring its position and attitude changes within the overlapping time window calculated in step S102 in the future period of time.

[0086] The core goal of the simulation is to calculate the collision probability of the AGV pair based on the motion trajectory, which is output as the quantitative collision risk result. To achieve high-precision probability calculation, in a preferred embodiment, the Monte Carlo simulation algorithm is used. Specifically, the platform introduces random disturbance variables conforming to the actual control logic of the AGV for the current speed, current acceleration and the preset motion constraints, such as simulating sensor measurement errors, communication delays, motor response fluctuations and other real uncertainties, and performs multiple, for example, 1000, independent motion trajectory simulations. In each simulation, the system will recalculate the accurate position and attitude of the two AGVs at each simulation time step based on the introduced random disturbance, and judge whether the spatial occupancy area, usually represented by the circumscribed polygon or bounding box of the AGV geometry model, interferes. Count the number of times the spatial occupancy area of the AGV pair interferes in the multiple independent motion trajectory simulations. Finally, according to the ratio of the number of times of interference to the total number of simulations, the collision probability is calculated. This calculation process can refer to the collision simulation step shown in Figure 3 The collision simulation step shown in, which explicitly includes the core steps of simulating AGV position changes based on actual data, calculating spatial occupancy area and calculating collision probability using Monte Carlo algorithm. This statistical-based method can effectively evaluate the possibility of collision under the influence of uncertain factors, and output a quantitative risk indicator.

[0087] After calculating the collision probability, it also includes risk classification according to the collision probability. As Figure 3As shown, after calculating the collision probability, the system will make a logical judgment and result distribution based on the probability value. Specifically: when the collision probability is zero, it is determined that there is no collision risk; when the collision probability is greater than zero but less than a preset probability threshold, for example, 30%, it is determined that there is a low-risk collision; and when the collision probability is greater than or equal to the preset probability threshold, it is determined that there is a high-risk collision. It is determined whether the collision probability is less than or equal to 0, and if so, it is determined that there is no collision and a log is recorded; if not, it is further determined whether it is less than 30%, thereby leading to two different processing paths of "low-risk collision" and "high-risk collision", and high risk is given priority to processing and low risk is given regular processing. The classification results of no risk, low risk and high risk are the final output of the quantified collision risk results of step S103. The high-risk collision result will directly and high-priority trigger the subsequent dynamic avoidance scheduling. This step introduces actual running parameters and high-fidelity Monte Carlo simulation to convert the qualitative / ranking risk judgment of S102 into an accurate, quantifiable collision probability and risk level, significantly improving the accuracy of risk assessment, avoiding unnecessary scheduling intervention due to over-conservative rough judgment, and realizing the logical complementarity of rough screening and precise calculation.

[0088] As an optional optimization implementation, to improve the accuracy of collision probability evaluation, the risk quantification model can be further refined in the Monte Carlo simulation. For example, when interference in the space occupation area is detected in each simulation, not only the event of "interference occurring" is recorded, but also the overlapping area or volume of the interference area in the proportion of the AGV's own occupation area is calculated. Subsequently, a mapping relationship model between the overlap ratio and the probability of actually colliding can be established based on a large amount of factory historical operation data or experimental data. When calculating the final collision probability, the "interference occurrence frequency" obtained by statistical calculation and the "interference severity" weighted based on the mapping relationship can be integrated, thereby outputting a more accurate risk quantification value.

[0089] S104, based on the quantified collision risk result, combining the preset dynamic priority rule, generating an avoidance strategy for the low-priority AGV in the AGV pair with collision risk, and issuing the standardized scheduling instruction generated based on the avoidance strategy through the standardized data channel for execution.

[0090] In a specific implementation, first, the avoidance urgency of each AGV in the AGV pair is determined according to the collision risk result, for example, the risk level or specific collision probability value output by step S103. Then, based on the preset dynamic priority rule, the real-time task priority of each AGV in the AGV pair is calculated.

[0091] The dynamic priority rule is a multi-factor weight calculation model that considers at least three core factors, namely, task urgency, load state, and task execution state. For example, the weight of task urgency is the highest, such as 50%-70%, the weight of load state is the second, such as 20%-40%, the weight of heavy load is higher than that of empty load, and the weight of task execution state is the lowest, such as 5%-15%, the weight of executed task is higher than that of to-be-started task. As a specific example, the following weight distribution can be adopted: the weight of task urgency is 60%, because it directly affects the production progress; the weight of load state is 30%; and the weight of execution state is 10%. Among them, the task urgency can be further classified, for example: the production breakpoint material transfer is the highest priority of level 1 task, the urgent order material is the second priority of level 2 task, the regular production material is the third priority of level 3 task, and the empty car return is the fourth priority of level 4 task in turn. It can be understood that in the load state, the priority of the heavy load AGV is higher than that of the empty load AGV. In the execution state, the priority of the AGV that has started the task is higher than that of the AGV that is to start the task. Finally, the low-priority AGV that needs to perform the avoidance operation is determined by comprehensively considering the avoidance urgency and the real-time task priority of the AGV. The high-priority AGV maintains its original path and speed to continue running to ensure that the key production task is not affected.

[0092] Further, after determining the low-priority AGV, a specific avoidance strategy needs to be generated for it. In the avoidance strategy generation link, the original path and original speed of the high-priority AGV are prioritized to run, and for the low-priority AGV, a specific strategy is developed and selected according to the physical environment around the intersection. This includes selectively generating control instructions according to the environmental conditions around the space intersection. The specific decision logic can be flexibly developed according to the risk level and time urgency, which can refer to the AGV dynamic avoidance scheduling flowchart shown in Figure 4 As shown in Figure 4 After the platform obtains the collision simulation result and determines the AGV priority, it first determines whether it is a high-priority AGV, and if so, it runs according to its original path and original speed; if it is a low-priority AGV, it further makes decisions based on time urgency and environmental conditions. If the risk level is high and the collision time is extremely urgent, a control instruction is generated to slow down the low-priority AGV. For example, when the predicted collision time window is close, such as only 2 seconds left, a speed reduction instruction can be generated, which specifies a target speed, such as usually set to 50% of the original speed, to pull apart the time difference by reducing the speed until the risk is removed.

[0093] If the time is relatively sufficient and there is a preset available avoidance area near the intersection, a control instruction is generated to control the low-priority AGV to enter the preset avoidance area for parking. At this time, as shown in Figure 4As shown, after determining that the time window is greater than 2 seconds and the avoidance area is available, the RCS of the low-priority AGV is issued with an early stop avoidance instruction. The instruction can specify a specific stop position, such as 5-8 meters away from the intersection, and a waiting time, such as covering the entire time window of the high-priority AGV passing through the intersection, and can set a buffer, such as stopping 1 second in advance and starting 1 second later, to ensure complete yielding.

[0094] If there is no available avoidance area or the detour is more efficient, an instruction is generated to indicate that the low-priority AGV re-plans the path, requiring its native RCS to plan an alternative path that does not pass through the current conflict point. As shown, if the low-priority AGV is not in the conflict point, the system will issue a re-planning path instruction. At this time, the new path can be required to meet the shortest path priority, while ensuring that no new intersection risk is introduced, thereby providing a safe and efficient alternative route for the AGV. Figure 4

[0095] In addition, after executing the deceleration strategy, the system continuously determines whether the high-priority AGV has passed through the intersection. If not, the low-priority AGV remains in the deceleration or waiting state; if it has passed, the low-priority AGV can resume normal driving, thereby forming a dynamic and adaptive control closed loop.

[0096] Finally, the standardized scheduling instructions generated based on the avoidance strategy are executed through the standardized data channel established in step S101. The central control platform encapsulates the strategy as a standard downlink instruction packet, converts it into a private instruction recognizable by the target AGV's native RCS through a protocol conversion mechanism or API gateway, and issues it. To ensure the effective execution of the instruction, a complete risk handling closed loop is formed, and instruction execution state monitoring is started. For example, the response and motion state of the low-priority AGV is continuously monitored, and if it is not executed within the expected time or the execution effect is not consistent, the instruction is reissued or manual alarm is triggered.

[0097] ​Specifically, in the instruction issuing and feedback link, the avoidance instruction is issued in real time to the native RCS scheduling system of the low-priority AGV in a standardized format, and at the same time, the central control platform starts the instruction monitoring mechanism, for example, collecting the response state of the AGV once every second. If the AGV feeds back "the instruction has been received and executed" within the preset time, for example, 1 second, the platform continues to monitor whether the subsequent motion state such as position and speed meets the requirements of the instruction; if no feedback is received or the feedback is "execution failed" after the preset time, the reissuing mechanism can be started, for example, the instruction is repeatedly issued every 0.5 seconds, and the maximum number of repetitions is 3. If it is still invalid after reissuing, the audible and light warning system is triggered, and the operation and maintenance personnel are notified to investigate on site, so as to ensure that the avoidance instruction is not missed and ineffective. This step realizes the decision upgrade from safe collision avoidance to efficient cooperation by fusing quantitative risk and dynamic production priority, and ensures the reliable landing of the control strategy through standardized instruction issuing and closed-loop monitoring, and finally achieves the balance between safety and efficiency in the mixed running scene of multiple brand AGVs.

[0098] In some embodiments of the present application, a multi-brand AGV mixed running control device is also provided, as shown in Figure 5 The device comprises:

[0099] A building module is configured to establish a standardized data channel with the native scheduling system of each brand AGV to obtain real-time running state data and planned path data of each AGV. Through protocol conversion or API adaptation, this module effectively solves the information barrier problem caused by heterogeneous underlying protocols and data formats of multi-brand AGVs, and provides a unified and real-time data basis for subsequent collaborative control.

[0100] A preliminary judgment module is configured to perform spatial intersection judgment on the planned paths of all AGVs based on the real-time running state data and planned path data, and perform collision risk preliminary judgment on AGV pairs with spatial intersection based on preset limit running parameters. This module uses a fast time window calculation and grading method based on limit parameters to realize efficient and conservative preliminary screening of potential collision risks, ensuring full coverage of risk identification, and selecting high-value targets for subsequent accurate simulation.

[0101] A simulation module is configured to perform motion simulation in a virtual environment based on actual running parameters for AGV pairs that have been preliminarily judged to have collision risks, to output quantitative collision risk results. The actual running parameters are determined based on at least the real-time running state data. This module introduces actual running parameters and uses simulation algorithms such as Monte Carlo to perform high-fidelity and quantitative evaluation of the preliminary judgment risk, improving the risk judgment from qualitative grading to accurate probability calculation, effectively avoiding invalid scheduling that may be caused by overly conservative preliminary judgment, and improving the scientificity and accuracy of the decision.

[0102] The avoidance module is configured to generate an avoidance strategy for the AGV with low priority in the AGV pair with collision risk based on the quantified collision risk result and the preset dynamic priority rule, and to issue a standardized scheduling instruction generated based on the avoidance strategy through the standardized data channel for execution. The module deeply integrates the quantified risk and real-time production priority, generates an optimal avoidance strategy based on the risk urgency and environmental conditions, and forms a closed-loop control through instruction issuing, state monitoring and re-alarming mechanism, thereby ensuring the running safety and maximizing the overall logistics efficiency and production continuity.

[0103] In some embodiments of the present application, an electronic device is also provided to execute the above-mentioned multi-brand AGV mixed running management method. Please refer to Figure 6 which shows a schematic diagram of an electronic device provided by some embodiments of the present application. As Figure 6 shown, the electronic device 9 comprises a processor 901, a memory 902, a bus 903 and a communication interface 904, the processor 901, the communication interface 904 and the memory 902 are connected through the bus 903; the memory 902 stores a computer program executable on the processor 901, and the processor 901 executes the computer program to perform the multi-brand AGV mixed running management method provided by any of the preceding embodiments of the present application.

[0104] The memory 902 can include a high-speed random access memory (RAM) and can also include a non-volatile memory such as at least one disk memory. The communication between the device network element and at least one other network element is realized through at least one communication interface 904 (which can be wired or wireless), and the Internet, wide area network, local network, metropolitan area network, etc. can be used.

[0105] The bus 903 can be an ISA bus, a PCI bus or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 902 is used to store programs, and the processor 901 executes the programs after receiving the execution instructions. The multi-brand AGV mixed running management method disclosed in any of the preceding embodiments of the present application can be applied to the processor 901 or realized by the processor 901.

[0106] The processor 901 can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware or the instruction in the form of software in the processor 901. The processor 901 described above can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a ready programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read only memory, a programmable read only memory or an electrically erasable programmable memory, a register or other mature storage medium in the art. The storage medium is located in the memory 902, and the processor 901 reads the information in the memory 902, and combines the hardware to complete the steps of the above method.

[0107] The electronic device provided by the embodiments of the present application and the multi-brand AGV same field mixed running management and control method provided by the embodiments of the present application have the same beneficial effects as the methods they adopt, run or implement.

[0108] The present application also provides a computer readable storage medium corresponding to the multi-brand AGV same field mixed running management and control method provided by the preceding embodiments, which stores a computer program (i.e. program product). When the computer program is run by a processor, it will execute the multi-brand AGV same field mixed running management and control method provided by any of the preceding embodiments.

[0109] It should be noted that the computer readable storage medium can include but is not limited to phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory, optical disc or other optical, magnetic storage medium, etc., which will not be described one by one here.

[0110] The computer readable storage medium provided by the embodiments of the present application and the multi-brand AGV same field mixed running management and control method provided by the embodiments of the present application have the same beneficial effects as the methods they store and the application programs they run or implement.

[0111] The application also provides a computer program product corresponding to the multi-brand AGV same-field mixed running management and control method provided by the foregoing embodiments, comprising a computer program executed by a processor to implement the multi-brand AGV same-field mixed running management and control method.

[0112] The computer program product provided by the application embodiment has the same beneficial effects as the method implemented by the computer program thereof, based on the same inventive concept as the multi-brand AGV same-field mixed running management and control method provided by the application embodiment.

[0113] It can be understood that the above description of each embodiment tends to emphasize the differences between each embodiment, and the same or similar parts can be referred to each other, and will not be described herein for brevity.

[0114] Those skilled in the art can understand that in the above method of the specific embodiment, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible inherent logic.

[0115] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the application, but not to limit it; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the application, and they should be covered in the scope of the claims and the specification of the application. Especially, as long as there is no structural conflict, each technical feature mentioned in each embodiment can be combined in any way. The application is not limited to the specific embodiments disclosed in the text, but includes all technical solutions falling within the scope of the claims.

Claims

1. A multi-brand AGV same field mixed running management method, characterized in that, The method comprises: establishing a standardized data channel with each brand AGV native scheduling system to obtain real-time running state data and planned path data of each AGV; based on the real-time running state data and the planned path data, judging the spatial intersection of the planned paths of all AGVs, and based on the preset limit running parameters, preliminarily judging the collision risk of the AGV pair with spatial intersection; for the AGV pair preliminarily judged to have a collision risk, motion simulation is performed in a virtual environment based on actual running parameters to output a quantitative collision risk result; wherein the actual running parameters are determined based on at least the real-time running state data; based on the quantitative collision risk result, combined with the preset dynamic priority rule, an avoidance strategy is generated for the low-priority AGV in the AGV pair with a collision risk, and a standardized scheduling instruction generated based on the avoidance strategy is issued for execution through the standardized data channel; the establishment of the standardized data channel with each brand AGV native scheduling system comprises: establishing a standardized specification for bidirectional data interaction, which includes a unified communication protocol, a unified data exchange format and a protocol conversion mechanism; the real-time running state data includes the current position, and the preset limit running parameters include the preset maximum running speed and the preset maximum vehicle body length; the preliminary collision risk judgment based on the preset limit running parameters comprises: for the AGV pair with spatial intersection, according to the preset maximum running speed and the preset maximum vehicle body length of each AGV, the time window required for each AGV to reach and completely pass through the spatial intersection point from the current position is calculated; based on the overlapping length of the two time windows, the collision risk of the AGV pair is classified.

2. The method of claim 1, wherein, The spatial intersection judgment of the planned paths of all AGVs comprises: in a unified plane coordinate system, the planned path of each AGV is represented as a polyline composed of a series of continuous line segments; using a line segment intersection detection algorithm based on vector cross product, it is judged one by one whether the planned path line segments of any two AGVs exist intersection points; if there is an intersection point, the intersection point is recorded as a spatial intersection point, and the corresponding two AGVs are associated to form the AGV pair with spatial intersection.

3. The method of claim 1, wherein, The real-time running state data and the planned path data of each AGV are obtained, comprising: through the protocol conversion mechanism, receiving the standard uplink data message reported by each native scheduling system and converted; from the standard uplink data message, the real-time position and speed fields as the real-time running state data and the planned path coordinate sequence field as the planned path data are parsed; the issuing and execution through the standardized data channel comprises: the avoidance strategy is packaged into a standard downlink instruction message conforming to the standardized specification; through the protocol conversion mechanism, the standard downlink instruction message is converted into a private instruction recognizable by the corresponding native scheduling system and is issued.

4. The method of claim 1, wherein, The actual operation parameters include a current speed and a current acceleration; the AGV pair preliminarily judged to have a collision risk is subjected to motion simulation in a virtual environment based on the actual operation parameters to output a quantitative collision risk result, including: obtaining preset vehicle body geometric dimensions and preset motion constraints of each AGV in the AGV pair; in the virtual environment, simulating motion trajectories of the AGVs in an overlapping time window based on the current speed, the current acceleration, the preset vehicle body geometric dimensions and the preset motion constraints; calculating a collision probability of the AGV pair based on the motion trajectories as the quantitative collision risk result.

5. The method of claim 4, wherein, The collision probability of the AGV pair based on the motion trajectories includes: introducing random disturbance variables conforming to the actual control logic of the AGV for the current speed, the current acceleration and the preset motion constraints, and performing multiple independent motion trajectory simulations; counting a number of times of interference of a space occupation area of the AGV pair in the multiple independent motion trajectory simulations; calculating the collision probability according to a proportion of the number of times of interference to a total number of simulations.

6. The method of claim 4, wherein, After the collision probability of the AGV pair is calculated based on the motion trajectories as the quantitative collision risk result, further including risk grading according to the collision probability, specifically: when the collision probability is zero, determining no collision risk; when the collision probability is greater than zero but less than a preset probability threshold, determining a low-risk collision; when the collision probability is greater than or equal to the preset probability threshold, determining a high-risk collision.

7. The method of claim 1, wherein, Based on the quantitative collision risk result, a low-priority AGV in the AGV pair having a collision risk is generated an avoidance strategy in combination with a preset dynamic priority rule, including: determining an avoidance urgency of each AGV in the AGV pair according to the collision risk result; calculating a real-time task priority of each AGV in the AGV pair based on the preset dynamic priority rule; determining a low-priority AGV in the AGV pair that needs to perform an avoidance operation by comprehensively considering the avoidance urgency and the real-time task priority.

8. The method of claim 2, wherein, The avoidance strategy for the low-priority AGV in the AGV pair having a collision risk further includes: selectively generating an instruction to control the low-priority AGV to decelerate, an instruction to control the low-priority AGV to drive into a preset avoidance area to stop, or an instruction to instruct the low-priority AGV to re-plan a path according to environmental conditions around the space intersection.

9. A multi-brand AGV same field mixed running management and control device, characterized in that, The device includes: an establishment module configured to establish a standardized data channel with a native scheduling system of each brand AGV to obtain real-time operation state data and planned path data of each AGV; a preliminary judgment module configured to perform spatial intersection judgment on planned paths of all AGVs based on the real-time operation state data and the planned path data, and perform collision risk preliminary judgment on AGV pairs having spatial intersections based on preset limit operation parameters; The simulation module is configured to perform motion simulation in a virtual environment based on actual operation parameters to output a quantitative collision risk result for the AGV pair determined to have a collision risk; wherein the actual operation parameters are determined based on at least the real-time operation state data; The avoidance module is configured to generate an avoidance strategy for the low-priority AGV in the AGV pair determined to have a collision risk based on the quantitative collision risk result and in combination with a preset dynamic priority rule, and to issue a standardized scheduling instruction generated based on the avoidance strategy through the standardized data channel for execution; The establishment of the standardized data channel with each brand AGV native scheduling system includes: Establishing a standardized specification for bidirectional data interaction, which includes a unified communication protocol, a unified data exchange format, and a protocol conversion mechanism; The real-time operation state data includes a current position, and the preset limit operation parameters include a preset maximum running speed and a preset maximum vehicle body length; the collision risk preliminary determination based on the preset limit operation parameters includes: For the AGV pair having a spatial intersection, a time window required for each AGV to reach and completely pass through the spatial intersection point from the current position is calculated according to the preset maximum running speed and the preset maximum vehicle body length of each AGV; Based on the overlap length of the two time windows, the collision risk of the AGV pair is classified.

10. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor runs the computer program to implement the method of any one of claims 1-8.

11. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-8.

Images