RGV trolley-based high-density stereoscopic warehouse delivery scheduling and transportation system and method

Abstract

The application discloses a high-density stereoscopic warehouse delivery scheduling and transportation system and method based on an RGV trolley, relates to the technical field of warehouse delivery scheduling and transportation, and comprises a scheduling management center, which is communicatively connected with a virtual model construction module, a task analysis and planning module, a task allocation module and a real-time monitoring intervention module, wherein the modules are connected through electric signals; the virtual model construction module is used for constructing virtual models of the warehouse and the RGV trolley. The application can synchronize the state of a physical entity in real time, accurately simulate the space-time path of multiple RGV trolleys, effectively avoid path conflicts and traffic congestion, significantly improve the efficiency of the delivery operation, greatly reduce the task delay caused by path conflicts, ensure the accuracy and timeliness of the delivery operation, and thus improve the overall logistics efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the warehouse dispatching and transportation technical field, in particular to a high-density stereoscopic warehouse dispatching and transportation system and method based on RGV trolley. BACKGROUND

[0002] Manufacturing industry has an increasing demand for automated and intelligent production mode and modern and intelligent logistics transportation and warehousing system. High-density stereoscopic warehouse is an advanced storage mode adopted to cope with limited storage space. Through stereoscopic stacking and vertical storage technology, the warehouse space can be maximized to improve storage density and reduce floor area. RGV trolley, as one of the core equipment of intelligent logistics, plays an important role in automated stereoscopic warehouse.

[0003] In high-density stereoscopic warehouse, goods are stored at different heights and positions, and the task of RGV trolley is to perform accurate storage and retrieval operation according to the dispatching order. Multiple RGV trolleys may have path conflict and traffic congestion problems during dispatching, which may cause task delay. Therefore, how to use digital twin technology to analyze the optimal execution path of dispatching task and improve the efficiency of dispatching and transportation system is a problem to be solved in the present application. Therefore, the present application proposes a high-density stereoscopic warehouse dispatching and transportation system and method based on RGV trolley. SUMMARY

[0004] The present application aims to provide a high-density stereoscopic warehouse dispatching and transportation system and method based on RGV trolley to solve the problems raised in the background.

[0005] To solve the above technical problems, the technical solution adopted by the present application is as follows:

[0006] In the first aspect, the high-density stereoscopic warehouse dispatching and transportation system based on RGV trolley comprises a dispatching management center, which is communicatively connected with a virtual model construction module, a task analysis and planning module, a task allocation module and a real-time monitoring and intervention module, wherein the modules are electrically connected.

[0007] The virtual model construction module is used to construct a virtual model of the warehouse and RGV trolley, synchronously map the shelf height, channel layout and real-time positioning of RGV trolley, and real-time synchronize the physical state of the warehouse and RGV trolley.

[0008] The task analysis and planning module is used to receive and analyze dispatching orders, generate an initial task queue, and simulate different path schemes of task allocation based on the virtual model of the warehouse and RGV trolley.

[0009] The task allocation module is used for intelligently allocating tasks to corresponding RGVs according to the outbound order priority and the current position of the RGVs.

[0010] The real-time monitoring intervention module is used for visually displaying the real-time track, task progress and abnormal events of the RGV, and supporting manual emergency intervention.

[0011] The further improvement of the technical scheme of the application is that the virtual model construction module specifically comprises:

[0012] The sensor network deployed in the warehouse is used for collecting the physical state data of the shelf height, channel layout, position, speed and load of the RGV in real time, the three-dimensional space information of the warehouse is acquired by using laser scanning and visual recognition technology, and the collected data is preprocessed to remove noise and redundant information.

[0013] According to the preprocessed data, the virtual model of the warehouse and the RGV is constructed by using modeling software, and in the virtual model, the number of layers, height and position of the shelf, and the layout and size of the channel are represented, and the appearance, size and motion characteristics of the RGV are modeled.

[0014] The real-time data communication channel between the virtual model and the physical entity is established, the real-time data collected by the sensor is transmitted to the virtual model by using the Internet of Things technology, and in the virtual model, the goods storage state of the shelf and the position and motion state information of the RGV are updated in real time, so that the virtual model can accurately reflect the real-time changes of the physical entity.

[0015] The further improvement of the technical scheme of the application is that the task analysis and allocation module comprises a task receiving and analyzing unit and a path planning simulation unit.

[0016] The task receiving and analyzing unit is used for receiving the outbound order, analyzing the task demand, determining the outbound order priority, goods position and RGV load state, and generating an initial task queue with a time window.

[0017] The path planning simulation unit is used for simulating the time-space path conflict of multiple RGVs in combination with the virtual model, generating a collision-free track of different path schemes, and optimizing the outbound path.

[0018] The further improvement of the technical scheme of the application is that the task receiving and analyzing unit specifically comprises:

[0019] The outbound order information is received through the interface with the warehouse management system, and the received order data is preliminarily formatted and checked for integrity to ensure the accuracy and integrity of the order information, and the orders are classified and labeled.

[0020] Detailed analysis is carried out on each outbound order, key task demand information is extracted, including goods location (shelf number, layer height, channel), outbound quantity, priority and goods type, and combined with real-time inventory information of the warehouse and a three-dimensional space indexing algorithm, the goods space coordinates are generated, and the feasibility of the task is ensured;

[0021] The load parameters of the current idle RGV trolley are acquired, the remaining load capacity and the allocated task amount of the idle RGV trolley are analyzed, a task-equipment matching degree matrix is established, a candidate equipment set meeting the load requirement is screened out, and a time window constraint is allocated to each task;

[0022] Whether the time window constraint overlaps with the existing task queue is verified by a time axis projection method, if there is a conflict, a priority arbitration mechanism is triggered, the current task time window is adjusted, the parsed tasks are sorted in descending order of priority and ascending order of time window urgency, an initial task queue is generated, and the initial task queue comprises a task unique identifier, a goods space coordinate, a candidate RGV trolley list and a time window constraint.

[0023] Further improvement of the technical scheme of the application is that the establishment process of the task-equipment matching degree matrix comprises:

[0024] The load parameters of the current idle RGV trolley are acquired from the control system of the RGV trolley, including the maximum load capacity, the allocated task amount and the current load capacity, the current position and state of each idle RGV trolley are recorded;

[0025] According to the acquired load parameters, the remaining load capacity of each idle RGV trolley is calculated by subtracting the current load capacity from the maximum load capacity of the RGV trolley, and at the same time, the interference of the allocated task amount on the task execution capability of the RGV trolley is analyzed to determine whether the RGV trolley can undertake a new task;

[0026] According to the task weight, the distance between the task position and the current position of the RGV trolley and the remaining load capacity of the RGV trolley, a matching degree value is calculated, a task-equipment matching degree matrix is established, and the matching degree value is arranged in descending order according to the size, so as to quantify the matching degree between each task and each idle RGV trolley;

[0027] Based on the task-equipment matching degree matrix, a candidate equipment set meeting the load requirement is screened out, for each task, according to the weight and urgency of the goods, a RGV trolley with high matching degree and sufficient remaining load capacity is selected from the matching degree matrix as a candidate equipment, and a time window constraint is allocated to each task.

[0028] Further improvement of the technical scheme of the application is that the path planning simulation unit specifically comprises:

[0029] The virtual model of the warehouse and the RGV trolley is acquired from a virtual model construction module, including a shelf layout, channel information and an initial position of the RGV trolley, and an initial task queue is synchronously generated, the space coordinates of goods of the task, a candidate RGV trolley list and time window constraints are acquired, related data of the initial task queue is mapped to the virtual model, and it is ensured that the simulation environment is consistent with the actual warehouse state;

[0030] Based on the virtual model, the space-time path of the multiple RGV trolleys when executing the task is simulated, potential conflicts between the paths of different RGV trolleys are detected through a time axis projection method and a space position analysis method, the positions and motion directions of the RGV trolleys at different time points are analyzed, collision points and congestion areas are identified, and time and position information of the conflicts are recorded to form a conflict report;

[0031] According to the conflict detection result, the path planning strategy of the RGV trolley is adjusted, a genetic algorithm is used to generate a collision-free trajectory of different path schemes, the path scheme is gradually optimized through multiple simulation iterations, the optimal path scheme of each iteration is recorded, path conflicts and task delays are reduced, and the warehouse-out efficiency is improved;

[0032] After multiple iterations, the path scheme with the highest fitness is selected as the optimal path scheme, the optimal path scheme is simulated and verified to ensure that it is collision-free and meets the time window constraints of the task, if new conflicts or non-compliance with the time window constraints are found in the simulation verification, the parameters of the genetic algorithm are adjusted, and optimization iteration is performed again until the generated path scheme is collision-free and meets all task requirements, and then the optimized path scheme is output and fed back to the task allocation module.

[0033] The further improvement of the technical scheme of the application is that the identification process of the collision points and the congestion areas is:

[0034] For each task, the start time and the end time of the task are determined according to the time window constraints of the task, the time axis projection method is used to project the task time window of each RGV trolley onto the time axis to form a time interval, and it is detected whether the task time intervals of different RGV trolleys overlap, if there is an overlapping part between two time intervals, there is a time conflict;

[0035] For each task, the position of the RGV trolley at different time points is calculated according to the space coordinates of the goods and the motion path of the RGV trolley, the space position analysis method is used to detect whether the positions of different RGV trolleys at the same time point overlap, the positions of different RGV trolleys at the same time point are compared, if the positions are the same, there is a space conflict, and the time point and the position of the conflict are recorded;

[0036] For each time point, check whether multiple RGV trolleys exist at the same position or adjacent positions, record the time and position information of all conflict occurrences, identify potential collision points and congestion areas, and form a conflict report, the conflict report including: conflict time, conflict position, RGV trolley number involved and conflict type (time conflict, space conflict or both).

[0037] The further improvement of the technical scheme of the present application is that the task allocation module specifically comprises:

[0038] An initial task queue with a time window is obtained from the task analysis and planning module, which contains the priority of the task, the space coordinates of the goods, and the candidate RGV trolley list information, and at the same time, the current positions, load states and remaining load capacities of all RGV trolleys are obtained from the control system of the RGV trolleys as basic data for task allocation, to ensure the accuracy and real-time performance of task allocation;

[0039] The task allocation efficiency is calculated in combination with the current position of the RGV trolley and the goods position of the task, and then the task is intelligently allocated to the corresponding RGV trolley;

[0040] For each task, the RGV trolley with the highest matching degree is selected for allocation, and a path planning is generated according to the time window constraint of the task and the current position of the RGV trolley.

[0041] The further improvement of the technical scheme of the present application is that the real-time monitoring intervention module specifically comprises:

[0042] Real-time data is collected from the control system of the RGV trolley and the task allocation module, including the real-time position, motion speed, task execution state of the RGV trolley and abnormal events in the system, and at the same time, the collected data is preprocessed to remove noise and redundant information, to ensure the accuracy and reliability of the data;

[0043] The collected data is visualized and displayed through a graphical interface to show the real-time trajectory of the RGV trolley, the task progress and the abnormal events;

[0044] The collected data is analyzed in real time to detect whether there is an abnormal event, including RGV trolley deviation from the predetermined trajectory, task delay, equipment failure, etc., and once an abnormal event is detected, an alarm mechanism is triggered immediately to notify the operator through sound, pop-up window or SMS, and at the same time, the abnormal event is marked in detail on the monitoring interface to provide the time, position and type information of the abnormal event;

[0045] Support the operator to carry out manual intervention through the monitoring interface, make emergency adjustment to the operation of the RGV trolley, the operator can manually adjust the running path, speed or task allocation of the RGV trolley according to the real-time monitoring information, in order to solve abnormal events or optimize system operation, at the same time, record the operation log of manual intervention.

[0046] In the second aspect, the high-density stereoscopic warehouse outbound scheduling and transportation method based on the RGV trolley is realized based on the high-density stereoscopic warehouse outbound scheduling and transportation system based on the RGV trolley, and includes the following steps:

[0047] S1, a virtual model of the warehouse and the RGV trolley is constructed, real-time synchronization of the shelf layout, channel information and RGV state is realized, and a digital twin basis is provided for the outbound scheduling;

[0048] S2, receiving the outbound order, analyzing the task demand, determining the priority, the goods position and the RGV trolley load state, and generating an initial task queue with a time window;

[0049] S3, combining the virtual model, simulating the space-time path of multiple RGV trolleys, detecting potential conflicts, generating a collision-free trajectory, optimizing the outbound path, and ensuring efficient execution of the task;

[0050] S4, intelligently assigning tasks according to the task priority and the current position of the RGV trolley, generating path planning, optimizing the path scheme through a genetic algorithm, and reducing conflicts and delays;

[0051] S5, real-time collection of RGV trolley operation data, visual display of the trajectory, task progress and abnormal events, support for manual emergency intervention, and ensuring stable operation of the system.

[0052] Due to the adoption of the above technical solutions, the present application has the following technical progress compared with the prior art:

[0053] The present application provides a high-density stereoscopic warehouse outbound scheduling and transportation system and method based on the RGV trolley, which can realize real-time synchronization of the state of physical entities by constructing a virtual model of the warehouse and the RGV trolley and combining digital twin technology, accurately simulate the space-time path of multiple RGV trolleys, effectively avoid path conflicts and traffic congestion, not only significantly improve the efficiency of outbound operations, but also greatly reduce the task delays caused by path conflicts, ensure the accuracy and timeliness of outbound operations, and thus improve the overall logistics efficiency.

[0054] This invention provides a high-density automated warehouse outbound scheduling and transportation system and method based on RGV carts. According to the priority of outbound orders and the real-time location and load status of RGV carts, tasks are intelligently allocated to the most suitable RGV carts. Based on a multi-dimensional matching degree task allocation strategy, high-priority tasks are executed first, while maximizing the utilization of RGV cart resources, avoiding resource idleness and overload, and improving resource utilization and overall system performance. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0056] Figure 1 This is a schematic diagram of the functional modules of the system of the present invention;

[0057] Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1, such as Figure 1 As shown, the present invention provides a high-density automated warehouse outbound scheduling and transportation system based on RGV vehicles, including a scheduling management center. The scheduling management center is communicatively connected to a virtual model construction module, a task parsing and planning module, a task allocation module, and a real-time monitoring and intervention module, wherein the modules are electrically connected to each other.

[0060] A virtual model construction module is configured to construct a virtual model of the warehouse and the RGV trolley, to synchronously map the height of the shelves, the layout of the passageways and the real-time positioning of the RGV trolley, to synchronously reflect the physical state of the warehouse and the RGV trolley in real time, to collect the physical state data of the height of the shelves, the layout of the passageways and the position, speed and load of the RGV trolley in real time through a sensor network deployed in the warehouse, to obtain the three-dimensional space information of the warehouse by using laser scanning and visual recognition technology, and to pre-process the collected data to remove noise and redundant information. The UWB positioning base station (positioning accuracy ±2 cm), the inertial measurement unit (IMU, acceleration / angular velocity), the pressure sensor (load monitoring) and the encoder (speed / odometer) are integrated on the RGV trolley to realize full-state perception. The Kalman filtering algorithm is used to correct the spatio-temporal consistency of the UWB positioning data. The RANSAC algorithm is used to remove outliers in the point cloud. The principal component analysis (PCA) is used to reduce the dimension of the point cloud data. The event-driven sampling strategy is used to reduce the communication load. According to the pre-processed data, the modeling software is used to construct the virtual model of the warehouse and the RGV trolley. In the virtual model, the number of layers, the height and the position of the shelves and the layout and the size of the passageways are represented. The appearance, the size and the motion characteristics of the RGV trolley are modeled so that the RGV trolley can run in the virtual environment according to the actual physical laws. The preliminary correspondence between the virtual model and the physical entity is realized by mapping the collected physical state data to the virtual model. The Poisson reconstruction algorithm is used to generate the surface mesh of the shelves and to extract the topological parameters of the layer height, the column spacing and the position of the load-bearing beam to form a structured model. The multi-body dynamics modeling method is used to decompose the trolley into the components of the trolley body, the driving wheel and the fork, to define the mass, the inertia matrix and the kinematics constraints of each component, to embed the collision detection algorithm (GJK algorithm) in the virtual model to ensure that the RGV trolley follows the Newton's law of motion and has real physical interaction with the shelves and the passageway boundaries, to establish a real-time data communication channel between the virtual model and the physical entity, to transmit the real-time data collected by the sensors to the virtual model through the Internet of Things technology, and to update the goods storage state of the shelves, the position and the motion state information of the RGV trolley in real time in the virtual model to ensure that the virtual model can accurately reflect the real-time changes of the physical entity.

[0061] A task analysis and planning module is configured to receive the analysis of the outbound orders, to generate an initial task queue, and to simulate different path schemes of task allocation based on the virtual model of the warehouse and the RGV trolley. The task analysis and planning module includes a task receiving and analyzing unit and a path planning simulation unit.

[0062] The task receiving and parsing unit receives outbound orders, parses task requirements, clarifies order priorities, cargo locations, and RGV cart load status, generates an initial task queue with time windows, receives outbound order information through an interface with the warehouse management system, and performs preliminary formatting and integrity checks on the received order data to ensure accuracy and completeness. Simultaneously, it categorizes and tags orders, performs detailed parsing on each outbound order, extracts key task requirement information, including cargo location (shelf number, shelf height, aisle), outbound quantity, priority, and cargo type, and combines this with real-time warehouse inventory information and a 3D spatial indexing algorithm to generate... Cargo spatial coordinates are used to ensure task feasibility. Load parameters of currently idle RGV vehicles are obtained, and the remaining load and assigned task volume of idle RGV vehicles are analyzed. A task-equipment matching degree matrix is ​​established, and a set of candidate equipment that meets the load requirements is screened. Time window constraints are assigned to each task, and the time window constraints are verified to see if they overlap with the existing task queue through the time axis projection method. If there is a conflict, a priority arbitration mechanism is triggered to adjust the current task time window. The parsed tasks are sorted in descending order of priority and ascending order of time window urgency to generate an initial task queue. The initial task queue contains a unique task identifier, cargo spatial coordinates, a list of candidate RGV vehicles, and time window constraints.

[0063] The calculation expression for the time-axis projection method is:

[0064] ;

[0065] In the formula, For the new task T new With existing task T exist The length of time window overlap. For the new task T new The start time of the time window For the new task T new End time of the time window For existing task T exist The start time of the time window For existing task T exist The end time of the time window, and the overlap time of the time window are non-negative numbers, when the new task T... new With existing task T exist When the time windows do not overlap, When time windows overlap, The time length of the overlapping portion;

[0066] The process of establishing the task-equipment matching matrix includes:

[0067] The load parameters of the current idle RGV trolley are acquired from the control system of the RGV trolley, including the maximum load, the allocated task amount and the current load amount, the current position and state of each idle RGV trolley are recorded, according to the acquired load parameters, the remaining load capacity of each idle RGV trolley is calculated by subtracting the current load amount from the maximum load of the RGV trolley, at the same time, the interference of the allocated task amount on the task execution ability of the RGV trolley is analyzed, whether the RGV trolley can undertake a new task is judged, the matching degree value is calculated according to the weight of goods of the task, the distance between the position of the task and the current position of the RGV trolley and the remaining load factor of the RGV trolley, a task-equipment matching degree matrix is established, the matching degrees between each task and each idle RGV trolley are quantified according to the descending order of the matching degree values, based on the task-equipment matching degree matrix, a candidate equipment set meeting the load requirement is screened out, for each task, the RGV trolley with higher matching degree and sufficient remaining load is selected as the candidate equipment from the matching degree matrix according to the weight of goods and the emergency degree of the task, and a time window constraint is allocated to each task;

[0068] The calculation process of the matching degree value is as follows:

[0069] The remaining load is obtained by subtracting the current load amount from the maximum load of the RGV trolley, whether the RGV trolley has sufficient load capacity to undertake the weight of goods of the task is determined, the weight of goods of the task is divided by the remaining load capacity of the RGV trolley to calculate the load matching degree, the matching degree between the remaining load capacity of the RGV trolley and the weight of goods of the task is analyzed, if the ratio is close to 1, it means that the remaining load capacity of the RGV trolley is suitable for the task, if the ratio is greater than 1, it means that the remaining load capacity of the RGV trolley is insufficient to undertake the task, if the ratio is significantly less than 1, it means that the remaining load capacity of the RGV trolley exceeds the demand of the task, the distance between the position of the task and the current position of the RGV trolley is analyzed, the influence of the distance factor on the matching degree is calculated, the farther the distance, the smaller the value, indicating that the greater the negative influence of the distance on the matching degree, the closer the distance, the closer the value to 1, indicating that the smaller the influence of the distance on the matching degree, the remaining load ratio is obtained by dividing the remaining load of the RGV trolley by the maximum load, the part of the load capacity of the RGV trolley that is not utilized is analyzed, the greater the ratio, the more sufficient the load capacity of the RGV trolley, the load matching degree, the influence of the distance factor on the matching degree and the remaining load ratio are multiplied to obtain the matching degree value, that is, the matching degree between the task and the RGV trolley is acquired;

[0070] The calculation expression of the matching degree value is as follows:

[0071] ;

[0072] In the formula, M i,jis the matching degree value of task i and RGV car j, indicating the matching degree between the task and the RGV car, W task,i is the weight of the goods of task i, is the maximum load of RGV car j, is the current load of RGV car j, D i,j is the distance between the goods position of task i and the current position of RGV car j, when approaches to or D i,j is very large, M i,j will approach to 0, only when and D i,j = 0, M i,j = 1;

[0073] The path planning simulation unit is used to simulate the space-time path conflict of multiple RGVs in combination with the virtual model, generate collision-free trajectories of different path schemes, and optimize the outbound path. The virtual model of the warehouse and the RGV car is obtained from the virtual model construction module, including the rack layout, channel information and the initial position of the RGV car, and the initial task queue is generated synchronously. The space coordinates of the task, the candidate RGV car list and the time window constraint are obtained, the relevant data of the initial task queue are mapped into the virtual model, ensuring that the simulation environment is consistent with the actual warehouse state. Based on the virtual model, the space-time path of multiple RGV cars in executing tasks is simulated. Through time axis projection method and space position analysis, the potential conflict between different RGV car paths is detected, the position and motion direction of each RGV car at different time points are analyzed, the collision points and congestion areas are identified, and the time and position information of the conflict occurrence are recorded to form a conflict report. According to the conflict detection result, the path planning strategy of the RGV car is adjusted, the genetic algorithm is used to generate collision-free trajectories of different path schemes, and through multiple simulation iterations, the path scheme is gradually optimized. The optimal path scheme of each iteration is recorded to reduce path conflict and task delay, improve the outbound efficiency, and select the path scheme with the highest fitness as the optimal path scheme after multiple iterations. The optimal path scheme is simulated and verified to ensure that it is collision-free and meets the time window constraint of the task. If new conflicts or time window constraints are found in the simulation verification, the parameters of the genetic algorithm are adjusted and the optimization iteration is performed again until the generated path scheme has no conflict in the simulation and meets all task requirements. Then the optimized path scheme is output, including the motion path of each RGV car, task allocation and time arrangement. The optimized path scheme is fed back to the task allocation module to provide guidance for the actual operation of the RGV car.

[0074] In addition, the identification process of the collision points and congestion areas is as follows:

[0075] For each task, determine the start time and end time of the task according to the time window constraint of the task, project the time window of each RGV car's task onto the time axis using the time axis projection method to form a time interval, detect whether the task time intervals of different RGV cars overlap, if there is an overlapping part between two time intervals, there is a time conflict, for each task, calculate the position of the RGV car at different time points according to the spatial coordinates of the goods and the motion path of the RGV car, use the spatial position analysis method to detect whether the positions of different RGV cars at the same time point overlap, compare the positions of different RGV cars at the same time point, if the positions are the same, there is a spatial conflict, record the time point and position of the conflict, for each time point, check whether there are multiple RGV cars at the same position or adjacent positions, record all the time and position information of the conflict, identify the potential collision points and congestion areas, form a conflict report, the conflict report includes: conflict time, conflict position, RGV car number involved and conflict type (time conflict, spatial conflict or both);

[0076] The optimization process of the path scheme is:

[0077] According to the conflict detection result, analyze the potential conflict points and conflict types of each task, set the optimization goals including reducing path conflict, shortening task completion time and reducing path length, determine the priority and time window constraint of each task, define the parameters of genetic algorithm, including population size, crossover rate, mutation rate and iteration number, initialize the population, generate multiple initial path schemes, each path scheme is represented as a chromosome, the genes in the chromosome represent the position and motion direction of the RGV car at different time points, encode each path scheme to ensure that the path scheme meets the time window constraint of the task and the spatial coordinates of the goods, design a fitness function to evaluate the fitness of each path scheme, calculate its fitness value, the higher the fitness value, the better the path scheme, according to the fitness value, select the path scheme with higher fitness into the next generation population, randomly select two parent path schemes, generate new child path schemes through crossover operation, randomly change the value of some genes through mutation operation to increase the diversity of the population, evaluate the fitness of the newly generated child path scheme, calculate its fitness value, repeat the selection, crossover, mutation and fitness evaluation operations to gradually optimize the path scheme, record the optimal path scheme of each iteration;

[0078] The expression of the fitness function is:

[0079] ;

[0080] In the formula, Fitness(P) is the fitness value output by the fitness function, Conflict(P) is the number of conflicts of the path scheme P, Time(P) is the task completion time of the path scheme P, Length(P) is the total path length of the path scheme P, and α, β, and γ are weight coefficients which are adjusted according to the optimization target;

[0081] The task allocation module is configured to intelligently allocate tasks to corresponding RGVs according to the priorities of the outbound orders and the current positions of the RGVs.

[0082] The real-time monitoring intervention module is configured to visually display the real-time trajectories of the RGVs, the task progress, and abnormal events, and support manual emergency intervention.

[0083] In Embodiment 2, as shown in the figure, on the basis of Embodiment 1, the application provides a technical solution: Figure 1 Preferably, the task allocation module specifically comprises:

[0084] The initial task queue with time windows is obtained from the task analysis and planning module, which contains the priorities of the tasks, the space coordinates of the goods, and the information of the candidate RGVs, and the positions, load states, and residual load capacities of all the RGVs are obtained from the control system of the RGVs as the basic data for task allocation, so as to ensure the accuracy and real-time performance of the task allocation, calculate the task allocation efficiency in combination with the current positions of the RGVs and the positions of the goods of the tasks, and then intelligently allocate tasks to corresponding RGVs.

[0085] The calculation process of the task allocation efficiency is as follows:

[0086] The distance between the goods position of the task and the current position of the RGV trolley is analyzed, the influence of the distance factor on the distribution efficiency is calculated, the farther the distance, the smaller the value, indicating that the greater the negative influence of the distance on the distribution efficiency, the closer the distance, the closer the value to 1, indicating that the smaller the influence of the distance on the distribution efficiency, the weight of the goods of the task is divided by the remaining carrying capacity of the RGV trolley, and the carrying matching degree is calculated, the ratio reflects the matching degree between the remaining carrying capacity of the RGV trolley and the weight of the task goods, if the ratio is close to 1, it means that the remaining carrying capacity of the RGV trolley is just suitable for the task, if the ratio is greater than 1, it means that the remaining carrying capacity of the RGV trolley is insufficient to undertake the task, if the ratio is less than 1, it means that the remaining carrying capacity of the RGV trolley exceeds the demand of the task, the ratio between the length of the time window of the task and the total available time of the RGV trolley is analyzed, and the time window matching degree is calculated, the ratio reflects the matching degree between the time window of the task and the available time of the RGV trolley, if the ratio is close to 1, it means that the time window of the task and the available time of the RGV trolley just match, if the ratio is greater than 1, it means that the time window of the task exceeds the available time of the RGV trolley, if the ratio is less than 1, it means that the time window of the task is less than the available time of the RGV trolley, the influence of the distance factor on the distribution efficiency, the carrying matching degree and the time window matching degree are multiplied to calculate the task distribution efficiency, the efficiency of the task distribution to the RGV trolley is analyzed;

[0087] The calculation expression of the task distribution efficiency is:

[0088] ;

[0089] In the formula, E i,j is the efficiency of the task i distributed to the RGV trolley j, represents the comprehensive efficiency of the task i distributed to the RGV trolley j, D i,j is the distance between the goods position of the task i and the current position of the RGV trolley j, W task,i is the weight of the goods of the task i, is the maximum carrying capacity of the RGV trolley j, is the current load of the RGV trolley j, is the end time of the time window of the task i, is the start time of the time window of the task i, is the total available time of the RGV trolley j, when D i,j is very large or tends to or is very small or is very large, E i,j tends to 0, only when D i,j =0 and and , E i,j =1;

[0090] The real-time monitoring intervention module specifically comprises:

[0091] Real-time data is collected from the control system and task allocation module of the RGV trolley, including the real-time position, motion speed, task execution state of the RGV trolley, and abnormal events in the system. At the same time, the collected data is preprocessed to remove noise and redundant information, ensuring the accuracy and reliability of the data. The collected data is visualized and displayed through a graphical interface, showing the real-time trajectory, task progress, and abnormal events of the RGV trolley. The real-time trajectory is displayed in the form of a dynamic graph, showing the motion path of the RGV trolley in the warehouse. The task progress is displayed in the form of a progress bar or a timeline, showing the completion of each task. Abnormal events are displayed in a prominent icon or color marker at the corresponding location. The graphical interface is automatically updated regularly to ensure that the operator can monitor the system's running state in real time and promptly identify potential problems. Real-time analysis of collected data detects whether there are abnormal events, including RGV trolley deviation from the predetermined trajectory, task delay, equipment failure, etc. Once an abnormal event is detected, an alarm mechanism is triggered immediately to notify the operator through sound, pop-up window, or SMS. At the same time, the abnormal event is marked in detail on the monitoring interface, providing information on the time, location, and type of the abnormal event to help the operator quickly locate and handle the problem, ensuring the stable operation of the system. The operator can manually adjust the RGV trolley's running path, speed, or task allocation through the monitoring interface to address abnormal events or optimize system operation. At the same time, the operator's intervention log is recorded.

[0092] In Example 3, as shown in Figure 2 Based on Examples 1-2, the application further provides a high-density stereoscopic warehouse outbound scheduling and transportation method based on RGV trolleys, which is implemented based on a high-density stereoscopic warehouse outbound scheduling and transportation system based on RGV trolleys, and comprises the following steps:

[0093] S1, a virtual model of the warehouse and the RGV trolley is constructed, real-time synchronization of the shelf layout, channel information, and RGV state is performed, and a digital twin basis is provided for outbound scheduling;

[0094] S2, receiving an outbound order, analyzing task requirements, determining priority, goods location, and RGV trolley load state, and generating an initial task queue with a time window;

[0095] S3, combining the virtual model, simulating the space-time path of multiple RGV trolleys, detecting potential conflicts, generating a collision-free trajectory, optimizing the outbound path, and ensuring efficient task execution;

[0096] S4, according to the task priority and the current position of the RGV trolley, intelligently allocate tasks, generate path planning, and optimize the path scheme through genetic algorithm to reduce conflicts and delays;

[0097] S5, real-time collection of RGV trolley operation data, visual display of trajectory, task progress and abnormal events, support for manual emergency intervention, and ensure stable operation of the system.

[0098] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A high-density automated warehouse outbound scheduling and transportation system based on RGV carts, including a scheduling management center, characterized in that: The scheduling and management center has communication connections to a virtual model construction module, a task parsing and planning module, a task allocation module, and a real-time monitoring and intervention module, wherein the modules are connected by electrical signals. The virtual model construction module is used to construct virtual models of the warehouse and RGV vehicles, synchronously map the shelf height, aisle layout and real-time positioning of the RGV vehicles, and synchronize the physical status of the warehouse and RGV vehicles in real time. The task parsing and planning module is used to receive and parse outbound orders, generate an initial task queue, and simulate different path schemes for task allocation based on the warehouse and RGV vehicle virtual model. The task parsing and allocation module includes a task receiving and parsing unit and a path planning simulation unit. The task receiving and parsing unit is used to receive outbound orders, parse task requirements, determine the priority of outbound orders, the location of goods and the load status of RGV carts, and generate an initial task queue with a time window, specifically including: The system receives outbound order information through an interface with the warehouse management system, performs preliminary formatting and integrity checks on the received order data, and categorizes and tags the orders. Each outbound order is analyzed in detail to extract key task requirements information, including the location of goods, the quantity of goods being shipped, priority, and type of goods. Combined with real-time inventory information of the warehouse and a three-dimensional spatial indexing algorithm, the spatial coordinates of the goods are generated. Obtain the load parameters of the currently idle RGV vehicles, analyze the remaining load and assigned tasks of the idle RGV vehicles, establish a task-equipment matching degree matrix, filter the candidate equipment set that meets the load requirements, and assign time window constraints to each task. The time axis projection method is used to verify whether the time window constraint overlaps with the existing task queue. If there is a conflict, the priority arbitration mechanism is triggered to adjust the current task time window. The parsed tasks are sorted in descending order of priority and ascending order of time window urgency to generate an initial task queue. The initial task queue contains a unique task identifier, cargo space coordinates, a candidate RGV vehicle list and time window constraints. The process of establishing the task-device matching degree matrix includes: Obtain the load parameters of the currently idle RGVs from the control system of the RGVs, including the maximum load, the number of assigned tasks and the current load, and record the current position and status of each idle RGV. Based on the obtained load parameters, the remaining load capacity of each idle RGV is calculated by subtracting the current load from the maximum load of the RGV. At the same time, the interference of the allocated task load on the task execution capability of the RGV is analyzed to determine whether the RGV can undertake new tasks. The matching degree value is calculated based on the weight of the cargo in the task, the distance between the task location and the current position of the RGV, and the remaining load factor of the RGV. A task-equipment matching degree matrix is ​​established and sorted in descending order according to the matching degree value. Based on the task-equipment matching degree matrix, a set of candidate equipment that meets the load requirements is selected. For each task, according to its cargo weight and urgency requirements, RGV trolleys with high matching degree and sufficient remaining load are selected from the matching degree matrix as candidate equipment, and time window constraints are assigned to each task. The path planning simulation unit is used to simulate spatiotemporal path conflicts of multiple RGVs in conjunction with a virtual model, generate collision-free trajectories for different path schemes, and optimize the outbound path, specifically including: The virtual model of the warehouse and RGV carts is obtained from the virtual model building module, including the shelf layout, aisle information and the initial position of the RGV carts. The initial task queue is generated synchronously, and the spatial coordinates of the goods, the list of candidate RGV carts and the time window constraints of the tasks are obtained. The relevant data of the initial task queue are mapped into the virtual model. Based on a virtual model, the spatiotemporal paths of multiple RGV vehicles are simulated when performing tasks. By using the time axis projection method and spatial position analysis, potential conflicts between different RGV vehicle paths are detected. The positions and directions of movement of each RGV vehicle at different time points are analyzed, collision points and congestion areas are identified, and the time and location information of the conflict are recorded to generate a conflict report. Based on the collision detection results, the path planning strategy of the RGV vehicle is adjusted, and a genetic algorithm is used to generate collision-free trajectories with different path schemes. Through multiple simulation iterations, the path schemes are gradually optimized, and the optimal path scheme of each iteration is recorded. After multiple iterations, the path scheme with the highest fitness is selected as the optimal path scheme. The optimal path scheme is then simulated to verify that it is collision-free and meets the time window constraints of the task. If the simulation verification finds new conflicts or does not meet the time window constraints, the parameters of the genetic algorithm are adjusted and the optimization iteration is repeated until the generated path scheme is collision-free in the simulation and meets all task requirements. Then, the optimized path scheme is output and fed back to the task allocation module. The expression for calculating the matching degree is: ； In the formula, M i,j W represents the matching degree between task i and RGV vehicle j, indicating the degree of matching between the task and the RGV vehicle. task,i Let be the weight of the goods for task i. The maximum load capacity of RGV trolley j is... Let D be the current load of RGV vehicle j. i,j Let be the distance between the location of the goods in task i and the current location of RGV vehicle j; The task allocation module is used to intelligently allocate tasks to the corresponding RGV vehicles for execution based on the priority of outbound orders and the current position of the RGV vehicles. The real-time monitoring and intervention module is used to visualize the RGV's real-time trajectory, task progress, and abnormal events, and supports manual emergency intervention.

2. The high-density automated warehouse outbound scheduling and transportation system based on RGV vehicles according to claim 1, characterized in that: The virtual model construction module specifically includes: By deploying a sensor network in the warehouse, real-time physical status data such as shelf height, aisle layout, and the position, speed, and load of RGV carts are collected. Laser scanning and visual recognition technologies are used to obtain the three-dimensional spatial information of the warehouse, and the collected data is preprocessed to remove noise and redundant information. Based on the preprocessed data, a virtual model of the warehouse and RGV carts is constructed using modeling software. The virtual model represents the number of shelves, height and location, as well as the layout and size of the aisles. At the same time, the appearance, size and motion characteristics of the RGV carts are modeled. Establish a real-time data communication channel between the virtual model and the physical entity. Use IoT technology to transmit real-time data collected by sensors to the virtual model. In the virtual model, update the storage status of goods on the shelf and the position and movement status of the RGV cart in real time.

3. The high-density automated warehouse outbound scheduling and transportation system based on RGV vehicles according to claim 1, characterized in that: The process for identifying the collision point and congestion area is as follows: For each task, the start and end times are determined according to the task's time window constraints. The time axis projection method is used to project the task time window of each RGV vehicle onto the time axis to form a time interval. The task time intervals of different RGV vehicles are checked for overlap. If two time intervals overlap, there is a time conflict. For each task, based on the spatial coordinates of the cargo and the movement path of the RGV, calculate the position of the RGV at different time points. Use spatial position analysis methods to detect whether the positions of different RGVs overlap at the same time point. Compare the positions of different RGVs at the same time point. If the positions are the same, there is a spatial conflict. Record the time and location of the conflict. For each time point, check whether multiple RGV vehicles are in the same or adjacent locations, record the time and location information of all conflicts, identify potential collision points and congestion areas, and generate a conflict report. The conflict report includes: conflict time, conflict location, RGV vehicle number involved, and conflict type.

4. The high-density automated warehouse outbound scheduling and transportation system based on RGV vehicles according to claim 1, characterized in that: The task allocation module specifically includes: The task parsing and planning module obtains an initial task queue with time windows, which includes task priority, cargo spatial coordinates, and candidate RGV vehicle list information. At the same time, the module obtains the current position, load status, and remaining load capacity of all RGV vehicles from the RGV vehicle control system. By combining the current position of the RGV vehicle and the position of the goods in the task, the task allocation efficiency is calculated, and then the task is intelligently allocated to the corresponding RGV vehicle. For each task, the RGV vehicle with the highest matching degree is selected for assignment, and a path plan is generated based on the task's time window constraints and the current position of the RGV vehicle.

5. The high-density automated warehouse outbound scheduling and transportation system based on RGV vehicles according to claim 4, characterized in that: The real-time monitoring and intervention module specifically includes: Data is collected in real time from the RGV vehicle's control system and task allocation module, including the RGV vehicle's real-time position, speed, task execution status, and abnormal events in the system. At the same time, the collected data is preprocessed to remove noise and redundant information. The collected data is visualized and displayed through a graphical interface, showing the real-time trajectory, task progress, and abnormal events of the RGV vehicle. The system analyzes the collected data in real time to detect any abnormal events. Once an abnormal event is detected, an alarm mechanism is immediately triggered to notify the operator via sound, pop-up window, or SMS. At the same time, the abnormal event is marked in detail on the monitoring interface, providing information on the time, location, and type of the abnormality. It supports operators to manually intervene through the monitoring interface to make emergency adjustments to the operation of the RGV, and at the same time, records the operation log of manual intervention.

6. A method for outbound scheduling and transportation of a high-density automated warehouse based on RGV carts, implemented based on the high-density automated warehouse outbound scheduling and transportation system based on RGV carts as described in any one of claims 1-5, characterized in that... Includes the following steps: S1. Construct a virtual model of the warehouse and RGV carts, and synchronize the shelf layout, aisle information and RGV status in real time; S2. Receive outbound orders, parse task requirements, determine priorities, goods locations and RGV cart load status, and generate an initial task queue with time windows. S3. Combine virtual models to simulate the spatiotemporal paths of multiple RGV vehicles, detect potential conflicts, generate collision-free trajectories, and optimize outbound paths. S4. Based on task priority and the current position of the RGV vehicle, intelligently allocate tasks and generate path plans; S5: Real-time collection of RGV vehicle operation data, visualization of trajectory, task progress and abnormal events, and support for manual emergency intervention.

Images