Smart warehousing inbound and outbound path planning method and system

Abstract

The present invention discloses a method and system for planning in-and-out warehouse paths for smart warehousing, and relates to the field of artificial intelligence technology. The method comprises: generating an optimal operating path for a waiting truck through a route planning model; predicting whether there is a collision risk between the current truck and all trucks in the warehouse in the time dimension through a time block, and predicting whether there is a collision risk between the current truck and all trucks in the warehouse in the space dimension through a space block; when the time block and the space block simultaneously detect that the current truck has a collision risk at a certain moment, the decision block determines that the current truck has a collision conflict, and determines an optimal avoidance strategy through an action value function; determines to avoid the waiting truck through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting truck, and executes the avoidance strategy: the technical solution of the present invention can solve the problem that the prior art cannot intelligently and accurately perform in-and-out warehouse path planning and collision processing.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and system for planning inbound and outbound paths for smart warehousing. Background Art

[0002] In the field of modern logistics and supply chain management, efficient operation of warehousing is crucial. With the continuous increase in global trade volume and the rapid development of e-commerce business, the scale of warehousing is constantly expanding, the types of goods are becoming more and more complicated, and traditional warehousing management methods are facing huge challenges.

[0003] Traditional warehousing operations rely on manual experience. In terms of order processing, the integration of cargo information and the control of time nodes are not accurate enough, and there is a lack of a systematic operation time prediction mechanism. It is difficult to reasonably arrange the order of tasks according to the processing time limit of the order, which can easily cause order backlogs or delays and reduce customer satisfaction. In terms of storage space allocation, it is impossible to fully utilize detailed information such as cargo type, quantity, weight and size in combination with historical data, resulting in unreasonable storage space planning, low storage space utilization, and chaotic storage of goods, which increases the difficulty and cost of subsequent storage and transportation.

[0004] At the same time, in the transportation process, the current operation path of the transporter lacks scientific planning, and does not comprehensively consider the optimization of multiple factors such as the starting point, the target point, the turning cost and the driving cost, resulting in low transportation efficiency and excessive energy consumption. Moreover, in a warehouse environment where multiple transporters are running at the same time, due to the lack of an effective collision prediction and avoidance mechanism, it is very easy to cause collision accidents between transporters, which not only damages equipment and goods, causing direct economic losses, but also brings the entire warehousing operation to a standstill, seriously affecting the normal operation and overall efficiency of the warehousing system.

[0005] Therefore, enterprises urgently need an intelligent and highly automated smart warehousing in-and-out path planning solution to achieve efficient integration and allocation of warehousing resources, improve the accuracy, timeliness and safety of warehousing operations, reduce operating costs, enhance the core competitiveness of enterprises in the logistics supply chain, and adapt to the development trend of modern logistics that is fast, precise and intelligent. Summary of the invention

[0006] In view of this, the present invention proposes a method and system for smart warehousing inbound and outbound path planning, which can intelligently integrate orders, shorten operation time, optimize storage location allocation, make full use of storage space, accurately plan paths, improve handling efficiency, reduce vehicle collision risks, improve the accuracy and smoothness of overall warehousing operations, reduce costs, and enhance the efficiency of corporate logistics operations.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for planning inbound and outbound paths for smart warehousing, comprising:

[0009] In response to an order initiation request, obtaining goods information and a processing time limit corresponding to the order, wherein the order includes a goods inbound order and a goods outbound order, and the goods information includes goods type information, goods quantity information, goods weight information, and goods size information;

[0010] Forecast the operation time for each order, generate the expected operation time, calculate the order allocation time node based on the processing time limit and the expected operation time, store the pending orders that have not reached the order allocation time node into the pending order pool, and determine whether there are multiple unallocated orders in the pending order pool. If so, analyze the goods information in the multiple unallocated orders;

[0011] When the order is a goods entry order, the location matching degree of each type of goods is calculated based on the goods information, historical goods entry and exit information and the inventory capacity of the location, and the recommended location of each type of goods is determined based on the location matching degree, and the goods of the same or adjacent locations are integrated into the same order within the limited space of the transport vehicle, wherein the adjacent locations are locations whose distance is less than a preset distance value;

[0012] When the order is a goods outbound order, goods in the same or adjacent warehouse locations are integrated into the same order within the limited space of the transport vehicle;

[0013] Constructing a route planning model including a map representation module and a path generation module, storing warehouse map data through the map representation module, determining the starting point and the target point of the transport vehicle to be operated through the path generation module, and generating the optimal operation path of the transport vehicle to be operated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more;

[0014] A collision prediction model including a time block, a space block and a decision block is constructed. The time block is used to predict whether there is a collision risk between the current truck and all the trucks in the warehouse in the time dimension. The space block is used to predict whether there is a collision risk between the current truck and all the trucks in the warehouse in the space dimension. When the time block and the space block simultaneously detect that the current truck has a collision risk at a certain moment, the decision block determines that the current truck has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting truck through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting truck, and executes the avoidance strategy; and replans the optimal operation path for avoiding the waiting truck through the truck route planning model.

[0015] On the basis of the above technical solution, the present invention can also be improved as follows:

[0016] Further, the calculating of the order allocation time node based on the processing time limit and the expected operation time includes:

[0017] Calculate the order allocation time node through formula (1);

[0018] Formula (1);

[0019] In the formula, Assign time nodes to orders, is the current time node, The processing time for the order is The expected operation time for the order.

[0020] Furthermore, the calculation of the storage location matching degree of each type of goods based on the goods information, historical goods in-and-out information and storage location inventory capacity includes:

[0021] The location matching degree of each type of goods is calculated by formula (2);

[0022] Formula (2);

[0023] In the formula, The location matching degree for each type of goods, Match weights for cargo types, Match the weights for the quantity of goods, Match weights for cargo weight, is the maximum load-bearing capacity of the storage location, The weight is the historical frequency matching weight. The weight for matching the remaining capacity of the storage location. is the cargo type matching degree, is the matching degree of goods quantity, is the cargo weight matching degree, For cargo size matching, is the historical in / out frequency matching degree, is the matching degree of the remaining capacity of the storage location, The quantity of the goods type.

[0024] Furthermore, the method of generating the optimal operation path of the transport vehicle to be operated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse includes:

[0025] The optimal operation path is calculated by formula (3);

[0026] Formula (3);

[0027] In the formula, is the optimal operation path, For all possible paths, is the number of nodes in the path, is the i-th node in the path, For slave nodes To Node The driving cost, For slave nodes To Node The turning cost, For Node The degree of influence of the real-time status of the warehouse on the path selection.

[0028] Furthermore, the selecting of the avoidance strategy through the action value function includes:

[0029] Calculate the action value function using formula (4);

[0030] Formula (4)

[0031] In the formula, is the action value function, is the time-dependent factor, is the spatial correlation factor, is the path-related factor, is the importance coefficient of the time-related factor, is the importance coefficient of the spatial correlation factor, is the importance coefficient of the path-related factor;

[0032] Will The avoidance strategy with the smallest value is determined as the optimal avoidance strategy.

[0033] Further, the method of determining the avoidance waiting vehicle by using the optimal avoidance strategy includes:

[0034] The avoidance priority of transport vehicle i is determined by formula (5);

[0035] Formula (5);

[0036] In the formula, Transporter The avoidance priority, is the importance coefficient of distance factor to priority, Transporter The distance to the target truck, For transport vehicles speed, is the importance coefficient of speed factor to priority, For transport vehicles The load, is the importance coefficient of the load factor to the priority, The remaining processing time for the transporter. is the importance coefficient of the remaining processing time factor to the priority;

[0037] Prioritize The tallest truck is determined as the yield waiting truck.

[0038] Furthermore, the calculation of the total waiting time and the waiting space position for avoiding the waiting transport vehicle includes:

[0039] The total waiting time for avoiding the waiting vehicle is determined by formula (6):

[0040] Formula (6);

[0041] In the formula, To avoid the total waiting time for the transport vehicle, To avoid the waiting time of the waiting truck based on the safe distance and average speed, Waiting time for warehouse traffic flow;

[0042] When the collision conflict is a head-on collision or a vertical collision predicted by different routes, the waiting space position for avoiding the waiting transport vehicle is determined by formula (7) and formula (8);

[0043] Formula (7);

[0044] Formula (8);

[0045] In the formula, is the horizontal coordinate of the waiting space position, is the ordinate of the waiting space position, is the horizontal coordinate of the collision point, is the ordinate of the collision point, To avoid step length, To avoid waiting for the transporter's driving direction vector The component on the horizontal axis, To avoid waiting for the transporter's driving direction vector The component on the ordinate;

[0046] When the collision conflict is a chasing collision, the waiting space is determined by formula (9) and formula (10);

[0047] Formula (9);

[0048] Formula (10);

[0049] In the formula, To avoid the horizontal coordinate of the current position of the waiting transporter, The vertical coordinate of the current position of the waiting transporter to avoid.

[0050] A smart warehousing inbound and outbound path planning system, comprising:

[0051] An acquisition module, configured to acquire, in response to an order initiation request, goods information and a processing time limit corresponding to an order, wherein the order includes a goods inbound order and a goods outbound order, and the goods information includes goods type information, goods quantity information, goods weight information, and goods size information;

[0052] The order processing module is used to predict the operation time for each order, generate the expected operation time, calculate the order allocation time node based on the processing time limit and the expected operation time, store the to-be-allocated orders that have not reached the order allocation time node into the to-be-allocated order pool, and determine whether there are multiple unallocated orders in the to-be-allocated order pool. If so, analyze the goods information in the multiple unallocated orders;

[0053] An order integration module is used for, when the order is a goods inbound order, calculating the warehouse location matching degree of each type of goods based on the goods information, historical goods inbound and outbound information and warehouse location inventory capacity, determining the recommended warehouse location of each type of goods based on the warehouse location matching degree, integrating the goods of the same or adjacent warehouse locations into the same order within the limited space of the transport vehicle, wherein the adjacent warehouse locations are warehouse locations whose distance is less than a preset distance value; when the order is a goods outbound order, integrating the goods of the same or adjacent warehouse locations into the same order within the limited space of the transport vehicle;

[0054] A first model building module is used to build a route planning model including a map representation module and a path generation module, wherein the map representation module stores warehouse map data, the path generation module determines the starting point and the target point of the transport vehicle to be operated, and the optimal operation path of the transport vehicle to be operated is generated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more;

[0055] The second model building module is used to build a collision prediction model including a time block, a space block and a decision block. The time block is used to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the time dimension. The space block is used to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the space dimension. When the time block and the space block simultaneously detect that the current transport vehicle has a collision risk at a certain moment, the decision block determines that the current transport vehicle has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting transport vehicle through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting transport vehicle, and executes the avoidance strategy;

[0056] The transport vehicle route planning model is also used to re-plan an optimal operation path that avoids waiting transport vehicles.

[0057] An electronic device comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the method when executing the computer program.

[0058] A non-transitory computer-readable storage medium stores a computer program, which implements the steps of the method when executed by a processor.

[0059] The present invention has the following advantages:

[0060] The inbound and outbound path planning method of the smart warehousing in the present invention can accurately grasp the order details, predict the operation time of the order and calculate the allocation time node by responding to the order initiation request and obtaining the goods information and processing time limit, and store the orders that have not arrived at the node into the order pool to be allocated, which can effectively avoid order processing confusion, ensure that the orders are processed in a reasonable order, improve the overall order processing efficiency, reduce the risk of delays, and make the warehousing operation rhythm more controllable.

[0061] The in-and-out path planning method of smart warehousing in the present invention calculates the location matching degree and determines the recommended location based on the cargo information, historical data and the inventory capacity of the location, so as to realize the scientific allocation of the location. It improves the utilization rate of the location, avoids the waste of space caused by the random stacking of goods, facilitates the management and search of goods, reduces the search time of in-and-out operations, and improves the rationality and efficiency of the use of storage space.

[0062] The in-and-out warehousing path planning method of the smart warehousing of the present invention integrates the goods of the same or adjacent storage locations into the same order within the limited space of the transport vehicle. Whether it is in-warehouse or out-warehouse, it can reduce the empty driving rate of the transport vehicle, increase the amount of goods transported in a single time, reduce the transportation cost, and make the transportation process more compact and orderly, thereby accelerating the circulation speed of goods.

[0063] The in-and-out path planning method of the smart warehouse in the present invention constructs a route planning model, which can generate the optimal operation path by using the map representation module to store map data and the path generation module to determine the starting point and the target point by comprehensively considering multiple factors. Considering the turning cost, driving cost and real-time status of the warehouse, the length and time of the driving path of the transport vehicle can be reduced, energy consumption can be reduced, the handling efficiency can be improved, and the dynamic environment changes of the warehouse can be adapted.

[0064] In the in-and-out path planning method of smart warehousing of the present invention, the time block, space block and decision block in the collision prediction model work together to accurately predict the collision risk and determine the conflict. The optimal avoidance strategy is determined by the action value function, the waiting time and position are calculated and the route is replanned, which effectively avoids the collision accidents of transport vehicles, ensures the safety of goods and the integrity of equipment, reduces the interruption and delay of operations caused by collisions, and maintains the continuity and stability of warehousing operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] For the purpose of illustration and not limitation, the present invention is now described in conjunction with the embodiments of the present invention and the accompanying drawings, in which:

[0066] Figure 1 It is a flowchart of a method for planning inbound and outbound paths for smart warehousing in an embodiment of the present invention;

[0067] Figure 2 A schematic diagram of the main components of the inbound and outbound path planning system of the smart warehouse in an embodiment of the present invention;

[0068] Figure 3 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0069] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solution in the implementation of the present invention will be clearly and completely described below in conjunction with the drawings in the implementation of the present invention. Obviously, the described implementation is only a part of the implementation of the present invention, not all of the implementations. Based on the implementation of the present invention, all other implementations obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0070] It should be noted that the terms "first", "second", etc. in the specification of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0071] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0072] Figure 1 FIG. 1 is a flow chart of a method for planning an inbound and outbound path of a smart warehouse in an embodiment of the present invention. Figure 1 As shown, the method for planning inbound and outbound paths for smart warehousing provided by an embodiment of the present invention includes the following steps S101 to S107.

[0073] S101, in response to an order initiation request, obtaining the goods information and processing time limit corresponding to the order.

[0074] Specifically: Orders include two categories: goods inbound orders and goods outbound orders. First, extract the order initiation signal from the order system, and then query and obtain the corresponding goods information and the set processing time limit data in the database according to the order number or related identification. Goods information includes goods type information, such as ordinary daily necessities, precision instruments or fresh food, which is helpful for subsequent classification and processing according to different goods characteristics; goods quantity information can intuitively reflect the size of the order, which has a key reference significance for resource allocation and operation arrangement; goods weight information plays a decisive role in the selection of storage locations and the adaptation of handling equipment. Overweight goods may require special carrying storage locations and handling tools; goods size information can further clarify the storage space required for goods, so as to reasonably arrange them in warehouse planning. For example, when receiving a storage order request, the system automatically retrieves the type of goods to which the order belongs, such as electronic equipment, the quantity is 50 boxes, each box weighs 10 kilograms, and the size is 50cm×40cm×30cm, and obtains the time limit information required by the order to complete the storage processing within 48 hours.

[0075] S102, predict the operation time for each order, generate the expected operation time, calculate the order allocation time node based on the processing time limit and the expected operation time, store the to-be-allocated orders that have not reached the order allocation time node into the to-be-allocated order pool, and determine whether there are multiple unallocated orders in the to-be-allocated order pool. If so, analyze the goods information in the multiple unallocated orders.

[0076] Specifically, when predicting the operation time, a variety of factors are considered, such as the difficulty of loading and unloading goods. If the goods are fragile, the loading and unloading process needs to be extra cautious and time-consuming; the complexity of the goods packaging. Complex packaging may require more time for disassembly and sorting; and the transportation distance. The distance from the warehouse entrance to a specific storage location directly affects the transportation time. Through the precise analysis of these factors, the expected operation time is obtained, and then the order allocation time node is determined in combination with the processing time limit. Orders that have not yet reached the allocation time are uniformly stored in the pool of orders to be allocated. If there are multiple unallocated orders in the pool, the cargo information is analyzed in depth, such as preliminarily classifying orders with the same or similar cargo types, so as to facilitate subsequent overall arrangements and improve overall operation efficiency.

[0077] Calculate the order allocation time node through formula (1);

[0078] Formula (1);

[0079] In the formula, Assign time nodes to orders, is the current time node, The processing time for the order is The expected operation time for the order.

[0080] Formula (1) means that the order allocation time node is equal to the current time plus the order processing time limit minus the expected operation time of the order;

[0081] Assume that the current time is 9:00 AM, the processing time limit for an order is 4 hours, and the expected processing time for the order is 1.5 hours;

[0082] ;

[0083] Finally, convert the result back to time format, 11.5 hours = 11:30 am (11:30), so the allocation time node of this order is 11:30 am, so just make sure that the order allocation is completed before 11:30 am.

[0084] S103, when the order is a goods warehousing order, the location matching degree of each type of goods is calculated based on the goods information, historical goods warehousing information and the inventory capacity of the storage location, and the recommended storage location of each type of goods is determined based on the storage location matching degree, and the goods in the same or adjacent storage locations are integrated into the same order within the limited space of the transport vehicle.

[0085] Specifically, when calculating the matching degree of storage locations, the weight and size in the cargo information are combined with the storage habits and frequency of similar cargo in the historical cargo in and out information, as well as the current storage capacity, to ensure that the storage location can accommodate the cargo and maintain the rationality of the inventory structure. For example, for small cargo that is frequently in and out of the warehouse, priority is given to storage locations that are close to the entrance and exit and easy to access. For the definition of adjacent storage locations, by setting an accurate preset distance value, such as 5 meters, the storage locations within a distance of 5 meters are regarded as adjacent storage locations, and their cargo is integrated into the same order, which can effectively reduce the driving path and number of stops of the transport vehicle in the warehouse and improve the handling efficiency.

[0086] The location matching degree of each type of goods is calculated by formula (2);

[0087] Formula (2);

[0088] In the formula, The location matching degree for each type of goods, Match weights for cargo types, Match the weights for the quantity of goods, Match weights for cargo weight, is the maximum load-bearing capacity of the storage location, The weight is the historical frequency matching weight. The weight for matching the remaining capacity of the storage location. is the cargo type matching degree, is the matching degree of goods quantity, is the cargo weight matching degree, For cargo size matching, is the historical in / out frequency matching degree, is the matching degree of the remaining capacity of the storage location, The quantity of the goods type.

[0089] Assume we have two types of goods ( =2), respectively, cargo type A and cargo type B.

[0090] Set the cargo type matching weight =0.1, cargo quantity matching weight =0.2, cargo weight matching weight =0.3, cargo size matching 0.1, historical in / out frequency matching weight =0.2, the remaining capacity of the storage location matches the weight =0.1;

[0091] For cargo type A:

[0092] set up =0.8, =0.7, =0.9, =0.6, =0.8, =0.7;

[0093] For cargo type B;

[0094] set up =0.8, =0.6, =0.8, =0.7, =0.7, =0.8;

[0095] when =2 hours; ;

[0096] For cargo type Part A,

[0097] ;

[0098] For cargo type B,

[0099] ;

[0100]

[0101] Therefore, the comprehensive location matching This value is about 0.845. This value can be used to evaluate the comprehensive adaptability of the storage location to these two types of goods. The higher the value, the more adaptable it is.

[0102] S104, when the order is a goods outbound order, goods in the same or adjacent storage locations are integrated into the same order within the limited space of the transport vehicle.

[0103] Specifically, when processing orders for goods out of the warehouse, the limited space of the transport truck is also taken into consideration. By accurately mastering the warehouse inventory layout, the goods in the same or adjacent storage locations can be integrated. For example, if multiple storage locations in a certain area store goods from the same batch or the same customer, integrating them into the same order can enable the transport truck to load as much goods as possible during one trip, reduce the empty driving rate, speed up the outbound delivery of goods, increase the throughput of warehousing operations, and better meet the customer's pick-up needs and improve customer satisfaction.

[0104] S105, constructing a route planning model including a map representation module and a path generation module, storing warehouse map data through the map representation module, determining the starting point and target point of the transport vehicle to be operated through the path generation module, and generating the optimal operation path of the transport vehicle to be operated based on the starting point, target point, turning cost, driving cost and real-time status of the warehouse.

[0105] Specifically, the map representation module uses advanced digital technology to accurately convert warehouse layout, channels, storage locations and other information into electronic map data, including detailed information such as the size, shape and access restrictions of each area. The path generation module is based on these map data, combined with the starting point of the transporter, such as the current charging area, loading and unloading area or temporary stop, and the target point (the number of target points is one or more, which may be a pickup point for a single storage location or a centralized loading and unloading point after the goods of multiple storage locations are integrated). When calculating the optimal operation path, the turning cost is fully considered. For example, large transporters need more time and space to adjust when turning in narrow channels, and their turning costs are higher; the driving cost covers factors such as distance and road conditions. For example, the uneven ground of some channels may limit the driving speed and increase costs; at the same time, combined with the real-time status of the warehouse, such as whether there are other transporters working on a certain section of the road, whether there are temporary piles of goods blocking the passage, etc., a complex algorithm is used to calculate the most time-saving, labor-saving and safe optimal operation path.

[0106] The optimal operation path is calculated by formula (3);

[0107] Formula (3);

[0108] In the formula, is the optimal operation path, For all possible paths, is the number of nodes in the path, is the i-th node in the path, For slave nodes To Node The driving cost, For slave nodes To Node The turning cost, For Node The degree of influence of the real-time status of the warehouse on the path selection.

[0109] Suppose we have two possible paths and :

[0110] path : Go directly from S to T (through 2 nodes, );

[0111] Driving cost: , (from S to the first intermediate node), (from the first intermediate node to T),

[0112] ;

[0113] Turning cost: Because there is no turning, ;

[0114] Warehouse real-time status impact: Assuming there is no other interference on the path, ,

[0115] Total cost: ;

[0116] path : Move down one square from S, then move right to (pass through 3 nodes, );

[0117] Driving cost: (from S to the first intermediate node), (from the first intermediate node to the second intermediate node), (from the second intermediate node to T),

[0118] ;

[0119] Turn cost: There is one turn, , ;

[0120] Impact of warehouse real-time status: Assume there are other transport vehicles at the turn. , other nodes have no effect, ;

[0121] Total cost: ;

[0122] Compare the total cost of the two paths, The total cost is 2, The total cost is 9.

[0123] Therefore, the optimal operation path , that is, the path directly from S to T to the right.

[0124] S106, construct a collision prediction model including a time block, a space block and a decision block, and predict whether there is a collision risk between the current truck and all the trucks in the warehouse in the time dimension through the time block, and predict whether there is a collision risk between the current truck and all the trucks in the warehouse in the space dimension through the space block. When the time block and the space block simultaneously detect that the current truck has a collision risk at a certain moment, the decision block determines that the current truck has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting truck through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting truck, and executes the avoidance strategy.

[0125] Specifically, the time block accurately analyzes the driving plan and speed of each truck, predicts whether they will meet in the same channel or intersection at a certain time in the future, and thus determines the collision risk in the time dimension. The space block uses the warehouse map data and the real-time location information of the truck to build a three-dimensional space model, analyze the spatial distance and movement trajectory between the trucks, and determine whether there is a possibility of collision in the spatial dimension. When the time and space blocks detect risks at the same time, the decision block intervenes quickly and determines the optimal avoidance strategy based on the action value function, which comprehensively considers multiple factors such as the difficulty of avoidance, the impact on the overall operation process, and the waiting time and space occupation. For example, let the truck with a shorter driving path and less urgent cargo avoid, determine it as an avoidance waiting truck, and accurately calculate its total waiting time, including the time required to wait for other trucks to pass and the additional driving time after the route is replanned due to avoidance. At the same time, determine the waiting space location, such as a temporary stop in a specific safe avoidance area or channel, and strictly implement the avoidance strategy to ensure the safety and efficiency of warehouse operations.

[0126] Calculate the action value function using formula (4);

[0127] Formula (4)

[0128] In the formula, is the action value function, is the time-dependent factor, is the spatial correlation factor, is the path-related factor, is the importance coefficient of the time-related factor, is the importance coefficient of the spatial correlation factor, is the importance coefficient of the path-related factor; The avoidance strategy with the smallest value is determined as the optimal avoidance strategy.

[0129] Suppose there are two trucks in a warehouse, truck A and truck B, which may collide during driving. We need to determine the optimal avoidance strategy for truck A through the action value function.

[0130] Time-related factors : Assume that truck A is expected to meet truck B in 5 seconds (the smaller the time value, the higher the risk of collision), then .

[0131] Spatial correlation factor : Assuming that the current distance between the two cars is 10 meters (the smaller the distance, the higher the risk of collision), after normalization, (The values ​​here are just examples and may need to be determined based on specific normalization methods).

[0132] Path-related factors : Assume that the intersection angle between the current driving path of truck A and the driving path of truck B is 30 degrees (the smaller the angle, the higher the risk of collision). After normalization, .

[0133] Importance coefficient of time-related factors ;

[0134] Importance coefficient of spatial correlation factor ;

[0135] Importance coefficient of path-related factors ;

[0136] ;

[0137] Assuming that we have other possible avoidance strategies, by calculating their respective action-value functions, we select the strategy with the smallest Q value as the optimal avoidance strategy. For example, if there is another avoidance strategy with a calculated Q value of 12, then that strategy will be determined as the optimal avoidance strategy.

[0138] The avoidance priority of transport vehicle i is determined by formula (5);

[0139] Formula (5);

[0140] In the formula, Transporter The avoidance priority, is the importance coefficient of distance factor to priority, Transporter The distance to the target truck, For transport vehicles speed, is the importance coefficient of speed factor to priority, For transport vehicles The load, is the importance coefficient of the load factor to the priority, The remaining processing time for the transporter. is the importance coefficient of the remaining processing time factor to the priority; The tallest truck is determined as the yield waiting truck.

[0141] Assume that there are three trucks in a warehouse: truck A, truck B, and truck C. They may collide while driving, and the avoidance priority of each truck needs to be determined by formula (5).

[0142] The distance between truck A and the target truck (assuming it is truck B) rice;

[0143] The distance between truck B and the target truck (assuming it is truck C) rice;

[0144] The distance between truck C and the target truck (assuming it is truck A) rice;

[0145] Speed ​​of truck A = 2 m / s, the speed of truck B = 3 m / s, the speed of truck C =1 m / s; load of truck A = 500 kg, load of truck B = 300 kg, load of truck C = 400 kg; remaining handling time for truck A = 30 minutes, the remaining processing time of truck B = 20 minutes, the remaining processing time of truck C =25 minutes; the importance coefficient of distance factor to priority , the speed factor's importance to the priority , the importance coefficient of load factor to priority , the importance coefficient of the remaining processing time limit factor to the priority ;

[0146] For transport vehicle A:

[0147] ;

[0148] ;

[0149] For truck B:

[0150] ;

[0151] ;

[0152] For transporter C:

[0153] ;

[0154] ;

[0155] Compare , and ,because The smallest, so truck A has the highest avoidance priority, and truck A should be determined as the avoidance waiting truck.

[0156] The total waiting time for avoiding the waiting vehicle is determined by formula (6):

[0157] Formula (6);

[0158] In the formula, To avoid the total waiting time for the transport vehicle, To avoid the waiting time of the waiting truck based on the safe distance and average speed, Waiting time for warehouse traffic flow;

[0159] Assume that when a transport truck is driving in a warehouse, it needs to wait for This may be because there are other trucks loading or unloading goods ahead, and the truck needs to wait at a safe distance.

[0160] Due to the heavy traffic of other trucks in the warehouse, this truck needs to wait extra For example, at a certain channel intersection, many transport vehicles intersect, causing a brief congestion.

[0161] Substituting the values ​​into the equation, we get: , the total waiting time for the transporter to complete a transport task For 8 minutes.

[0162] When the collision conflict is a head-on collision or a vertical collision predicted by different routes, the waiting space position for avoiding the waiting transport vehicle is determined by formula (7) and formula (8);

[0163] Formula (7);

[0164] Formula (8);

[0165] In the formula, is the horizontal coordinate of the waiting space position, is the ordinate of the waiting space position, is the horizontal coordinate of the collision point, is the ordinate of the collision point, To avoid step length, To avoid waiting for the transporter's driving direction vector The component on the horizontal axis, To avoid waiting for the transporter's driving direction vector The component on the ordinate;

[0166] Assume that there are two trucks A and B in a warehouse and they are expected to collide with each other while driving. We need to use formulas (7) and (8) to determine the waiting space position of truck A (assuming it is avoiding the waiting truck).

[0167] Collision point location : Assume that the coordinates of the collision point are (10,10) (unit: meter).

[0168] Avoidance step length S: Assume that the avoidance step length is 3 meters.

[0169] Avoid the driving direction vector of the waiting transporter A : Assume that the moving direction vector of truck A is (2,2), which means that it has the same moving tendency in both x and y directions.

[0170] Avoid waiting for the transporter A's driving direction vector model :

[0171] According to the calculation formula of vector modulus, .

[0172] Calculate the horizontal coordinate of the waiting space position according to formula (7) :

[0173] Substituting the values: (rice);

[0174] Calculate the ordinate of the waiting space position according to formula (8): :

[0175] Substituting the values: (rice);

[0176] Therefore, the waiting space position for avoiding and waiting truck A is approximately (7.88,7.88) meters. This position is calculated after considering the collision point position, avoidance step length, and driving direction vector, which can help truck A avoid truck B safely.

[0177] When the collision conflict is a chasing collision, the waiting space is determined by formula (9) and formula (10);

[0178] Formula (9);

[0179] Formula (10);

[0180] In the formula, To avoid the horizontal coordinate of the current position of the waiting transporter, The vertical coordinate of the current position of the waiting transporter to avoid.

[0181] Assume that there are two trucks in a warehouse, truck A and truck B. Truck B is chasing truck A. To avoid collision, the waiting space of truck A (avoiding the waiting truck) needs to be determined according to formula (9) and formula (10).

[0182] Assume that the horizontal coordinate of the current position of truck A is =20 meters, ordinate =30 meters.

[0183] Assume that the avoidance step length S = 5 meters.

[0184] Assume that the driving direction vector of truck A is ,but .

[0185] The ordinate of the waiting space position is calculated according to formula (9):

[0186] Substituting the values:

[0187] ;

[0188] Calculate the horizontal coordinate of the waiting space position according to formula (10);

[0189] Substituting the values:

[0190] ;

[0191] Therefore, when a chase-and-collision conflict occurs, the waiting space position of the avoidance waiting truck A is (17, 26) meters. This position is calculated by considering the current position of truck A, the avoidance step length, and the driving direction vector, which can help truck A safely avoid truck B.

[0192] S107, replanning an optimal operation path for avoiding the waiting transport vehicles through the transport vehicle route planning model.

[0193] Specifically, after the avoidance waiting truck and its related waiting information are determined, the truck route planning model is started again. It re-evaluates the real-time status of the warehouse, including the new positions of other trucks, the traffic conditions of the passages, and the status of the goods at the target point. Taking the current waiting space position of the avoidance waiting truck as the new starting point, combined with the original target point or the target point that may change during the waiting period, recalculate the turning cost, driving cost, and consider the latest real-time status of the warehouse. Through a complex path planning algorithm, a new optimal operation path is generated for the avoidance waiting truck to ensure that it can continue to perform its tasks in the most efficient way after completing the avoidance, minimizing the impact of the avoidance operation on the efficiency of the entire warehouse operation.

[0194] Figure 2Schematic diagram of the main components of the smart warehouse in-and-out path planning system in the embodiment of the present invention. Figure 2 As shown, the smart warehousing inbound and outbound path planning system 1 provided in an embodiment of the present invention includes an acquisition module 10, an order processing module 20, an order integration module 30, a first model building module 40, and a second model building module 50.

[0195] The acquisition module 10 is used to respond to an order initiation request and obtain the goods information and processing time limit corresponding to the order, wherein the order includes a goods inbound order and a goods outbound order, and the goods information includes goods type information, goods quantity information, goods weight information and goods size information.

[0196] The order processing module 20 is used to predict the operation time for each order, generate the expected operation time, calculate the order allocation time node based on the processing time limit and the expected operation time, store the to-be-allocated orders that have not reached the order allocation time node into the to-be-allocated order pool, and determine whether there are multiple unallocated orders in the to-be-allocated order pool. If so, analyze the goods information in the multiple unallocated orders.

[0197] The order integration module 30 is used to calculate the location matching degree of each type of goods based on the goods information, historical goods in-and-out information and the inventory capacity of the location when the order is a goods inbound order, determine the recommended location of each type of goods based on the location matching degree, and integrate the goods of the same or adjacent locations into the same order within the limited space of the transport vehicle, wherein the adjacent locations are locations whose distance is less than a preset distance value; when the order is a goods outbound order, integrate the goods of the same or adjacent locations into the same order within the limited space of the transport vehicle;

[0198] The first model building module 40 is used to build a route planning model including a map representation module and a path generation module, wherein the map representation module stores warehouse map data, the path generation module determines the starting point and the target point of the transport vehicle to be operated, and generates the optimal operation path of the transport vehicle to be operated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more;

[0199] The second model building module 50 is used to build a collision prediction model including a time block, a space block and a decision block. The time block is used to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the time dimension. The space block is used to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the space dimension. When the time block and the space block simultaneously detect that the current transport vehicle has a collision risk at a certain moment, the decision block determines that the current transport vehicle has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting transport vehicle through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting transport vehicle, and executes the avoidance strategy;

[0200] The transport vehicle route planning model is also used to re-plan an optimal operation path that avoids waiting transport vehicles.

[0201] Figure 3 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as Figure 3 As shown, the electronic device 60 includes: a processor 601 (processor), a memory 602 (memory) and a bus 603;

[0202] The processor 601 and the memory 602 communicate with each other via the bus 603;

[0203] The processor 601 is used to call the program instructions in the memory 602 to execute the methods provided by the above-mentioned method embodiments, for example, including: in response to an order initiation request, obtaining the cargo information and processing time limit corresponding to the order, wherein the order includes a cargo inbound order and a cargo outbound order, and the cargo information includes cargo type information, cargo quantity information, cargo weight information and cargo size information; predicting the operation time for each order, generating an expected operation time, calculating the order allocation time node based on the processing time limit and the expected operation time, storing the to-be-allocated orders that have not reached the order allocation time node into the to-be-allocated order pool, determining whether there are multiple unallocated orders in the to-be-allocated order pool, and if so, analyzing the cargo information in the multiple unallocated orders; when the order is a cargo inbound order, calculating the location matching degree of each cargo type based on the cargo information, historical cargo inbound and outbound information and the inventory capacity of the location, determining the recommended location of each cargo type based on the location matching degree, integrating the cargo of the same or adjacent locations into the same order within the limited space of the transport vehicle, wherein the adjacent locations are locations whose distance is less than a preset distance value; when the order is a cargo outbound order, The invention integrates the goods of the same or adjacent storage locations into the same order within the limited space of the vehicle; constructs a route planning model including a map representation module and a path generation module, stores the warehouse map data through the map representation module, determines the starting point and the target point of the waiting transport vehicle through the path generation module, and generates the optimal operation path of the waiting transport vehicle based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more; constructs a collision prediction model including a time block, a space block and a decision block, predicts whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the time dimension through the time block, and predicts whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the space dimension through the space block. When the time block and the space block simultaneously detect that the current transport vehicle has a collision risk at a certain moment, the decision block determines that the current transport vehicle has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting transport vehicle through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting transport vehicle, and executes the avoidance strategy; replans the optimal operation path of the route to avoid the waiting transport vehicle through the transport vehicle route planning model.

[0204] The present embodiment provides a non-transitory computer-readable storage medium, which stores computer instructions. The computer instructions enable a computer to execute the methods provided by the above-mentioned method embodiments, for example, including: in response to an order initiation request, obtaining the cargo information and processing time limit corresponding to the order, wherein the order includes a cargo inbound order and a cargo outbound order, and the cargo information includes cargo type information, cargo quantity information, cargo weight information, and cargo size information; performing operation time prediction for each order, generating an expected operation time, calculating the order allocation time node based on the processing time limit and the expected operation time, storing the to-be-allocated orders that have not reached the order allocation time node into a to-be-allocated order pool, determining whether there are multiple unallocated orders in the to-be-allocated order pool, and if so, analyzing the cargo information in the multiple unallocated orders; when the order is a cargo inbound order, calculating the storage location matching degree of each cargo type based on the cargo information, historical cargo inbound and outbound information, and storage location inventory capacity, determining the recommended storage location of each cargo type based on the storage location matching degree, and integrating the cargo of the same or adjacent storage locations into the same order within the limited space of the transport vehicle, wherein the adjacent storage locations are storage locations with a distance less than a preset distance value; When the order is a goods outbound order, the goods in the same or adjacent warehouses are integrated into the same order within the limited space of the transport vehicle; a route planning model including a map representation module and a path generation module is constructed, the warehouse map data is stored through the map representation module, the starting point and the target point of the transport vehicle to be operated are determined through the path generation module, and the optimal operation path of the transport vehicle to be operated is generated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more; a collision prediction model including a time block, a space block and a decision block is constructed, and the time block is used to predict the collision between the current transport vehicle and the Whether there is a collision risk among all the trucks in the warehouse in the time dimension, the spatial block is used to predict whether there is a collision risk between the current truck and all the trucks in the warehouse in the spatial dimension. When the time block and the spatial block simultaneously detect that the current truck has a collision risk at a certain moment, the decision block determines that the current truck has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting truck through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting truck, and executes the avoidance strategy; replans the optimal operation path for avoiding the waiting truck through the truck route planning model.

[0205] A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various storage media that can store program codes.

[0206] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for planning inbound and outbound paths for smart warehousing, characterized in that: include: In response to an order initiation request, obtaining goods information and a processing time limit corresponding to the order, wherein the order includes a goods inbound order and a goods outbound order, and the goods information includes goods type information, goods quantity information, goods weight information, and goods size information; Forecast the operation time for each order, generate the expected operation time, calculate the order allocation time node based on the processing time limit and the expected operation time, store the pending orders that have not reached the order allocation time node into the pending order pool, and determine whether there are multiple unallocated orders in the pending order pool. If so, analyze the goods information in the multiple unallocated orders; When the order is a goods entry order, the location matching degree of each type of goods is calculated based on the goods information, historical goods entry and exit information and the inventory capacity of the location, and the recommended location of each type of goods is determined based on the location matching degree, and the goods of the same or adjacent locations are integrated into the same order within the limited space of the transport vehicle, wherein the adjacent locations are locations whose distance is less than a preset distance value; When the order is a goods outbound order, goods in the same or adjacent warehouse locations are integrated into the same order within the limited space of the transport vehicle; Constructing a route planning model including a map representation module and a path generation module, storing warehouse map data through the map representation module, determining the starting point and the target point of the transport vehicle to be operated through the path generation module, and generating the optimal operation path of the transport vehicle to be operated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more; Construct a collision prediction model including a time block, a space block and a decision block. Use the time block to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the time dimension. Use the space block to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the space dimension. When the time block and the space block simultaneously detect that the current transport vehicle has a collision risk at a certain moment, the decision block determines that the current transport vehicle has a collision conflict, and determines the optimal avoidance strategy through the action value function; determine to avoid the waiting transport vehicle through the optimal avoidance strategy, calculate the total waiting time and waiting space position of the waiting transport vehicle, and execute the avoidance strategy; The optimal operation path for avoiding the waiting transport vehicles is replanned through the transport vehicle route planning model.

2. The method for planning inbound and outbound paths of smart warehousing according to claim 1 is characterized in that: The calculation of the order allocation time node based on the processing time limit and the expected operation time includes: Calculate the order allocation time node through formula (1); Formula (1): In the formula, Assign time nodes to orders. is the current time node, The processing time for the order is The expected operation time for the order.

3. The method for planning inbound and outbound paths of smart warehousing according to claim 1 is characterized in that: The calculating of the storage location matching degree of each type of goods based on the goods information, historical goods in-and-out storage information and storage location inventory capacity includes: The location matching degree of each type of goods is calculated by formula (2); Formula (2): In the formula, The location matching degree for each type of goods, Match weights for cargo types, Match the weights for the quantity of goods, Match weights for cargo weight, is the maximum load-bearing capacity of the storage location, The weight is the historical frequency matching weight. Match the weight of the remaining capacity of the storage location. is the cargo type matching degree, is the goods quantity matching degree, is the cargo weight matching degree, For cargo size matching, is the historical in / out frequency matching degree, is the matching degree of the remaining capacity of the storage location, The quantity of the goods type.

4. The method for planning inbound and outbound paths of smart warehousing according to claim 1 is characterized in that: The method of generating the optimal operation path of the transport vehicle to be operated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse includes: The optimal operation path is calculated by formula (3); Formula (3): In the formula, is the optimal operation path, For all possible paths, is the number of nodes in the path, is the i-th node in the path, For slave nodes To Node The driving cost, For slave nodes To Node The turning cost, For Node The degree of influence of the real-time status of the warehouse on the path selection.

5. The method for planning inbound and outbound paths of smart warehousing according to claim 1 is characterized in that: The step of selecting an avoidance strategy through an action value function includes: Calculate the action value function using formula (4); Formula (4) In the formula, is the action value function, is the time-dependent factor, is the spatial correlation factor, is the path-related factor, is the importance coefficient of the time-related factor, is the importance coefficient of the spatial correlation factor, is the importance coefficient of the path-related factor; Will The avoidance strategy with the smallest value is determined as the optimal avoidance strategy.

6. The method for planning inbound and outbound paths of smart warehousing according to claim 1 is characterized in that: The method of determining the avoidance waiting vehicle by using the optimal avoidance strategy includes: The avoidance priority of transport vehicle i is determined by formula (5); Formula (5): In the formula, Transporter The avoidance priority, is the importance coefficient of distance factor to priority, Transporter The distance to the target truck, For transport vehicles speed, is the importance coefficient of speed factor to priority, For transport vehicles The load, is the importance coefficient of the load factor to the priority, The remaining processing time for the transporter. is the importance coefficient of the remaining processing time factor to the priority; Prioritize The tallest truck is determined as the yield waiting truck.

7. The method for planning inbound and outbound paths of smart warehousing according to claim 6, characterized in that: The calculation of the total waiting time and waiting space position of the waiting transport vehicle includes: The total waiting time for avoiding the waiting vehicle is determined by formula (6): Formula (6): In the formula, To avoid the total waiting time for the transport vehicle, To avoid the waiting time of the waiting truck based on the safe distance and average speed, Waiting time for warehouse traffic flow; When the collision conflict is a head-on collision or a vertical collision predicted by different routes, the waiting space position for avoiding the waiting transport vehicle is determined by formula (7) and formula (8); Formula (7): Formula (8): In the formula, is the horizontal coordinate of the waiting space position, is the ordinate of the waiting space position, is the horizontal coordinate of the collision point, is the ordinate of the collision point, To avoid step length, To avoid waiting for the transporter's driving direction vector The component on the horizontal axis, To avoid waiting for the transporter's driving direction vector The component on the ordinate; When the collision conflict is a chasing collision, the waiting space is determined by formula (9) and formula (10); Formula (9): Formula (10): In the formula, To avoid the horizontal coordinate of the current position of the waiting transporter, The vertical coordinate of the current position of the waiting transporter to avoid.

8. A system for planning inbound and outbound routes for smart warehousing, characterized in that: include: An acquisition module, configured to acquire, in response to an order initiation request, goods information and a processing time limit corresponding to an order, wherein the order includes a goods inbound order and a goods outbound order, and the goods information includes goods type information, goods quantity information, goods weight information, and goods size information; The order processing module is used to predict the operation time for each order, generate the expected operation time, calculate the order allocation time node based on the processing time limit and the expected operation time, store the to-be-allocated orders that have not reached the order allocation time node into the to-be-allocated order pool, and determine whether there are multiple unallocated orders in the to-be-allocated order pool. If so, analyze the goods information in the multiple unallocated orders; An order integration module is used for, when the order is a goods inbound order, calculating the warehouse location matching degree of each type of goods based on the goods information, historical goods inbound and outbound information and warehouse location inventory capacity, determining the recommended warehouse location of each type of goods based on the warehouse location matching degree, integrating the goods of the same or adjacent warehouse locations into the same order within the limited space of the transport vehicle, wherein the adjacent warehouse locations are warehouse locations whose distance is less than a preset distance value; when the order is a goods outbound order, integrating the goods of the same or adjacent warehouse locations into the same order within the limited space of the transport vehicle; A first model building module is used to build a route planning model including a map representation module and a path generation module, wherein the map representation module stores warehouse map data, the path generation module determines the starting point and the target point of the transport vehicle to be operated, and the optimal operation path of the transport vehicle to be operated is generated based on the starting point, the target point, the turning cost, the driving cost and the real-time status of the warehouse, wherein the number of the target points is one or more; The second model building module is used to build a collision prediction model including a time block, a space block and a decision block. The time block is used to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the time dimension. The space block is used to predict whether there is a collision risk between the current transport vehicle and all the transport vehicles in the warehouse in the space dimension. When the time block and the space block simultaneously detect that the current transport vehicle has a collision risk at a certain moment, the decision block determines that the current transport vehicle has a collision conflict, and determines the optimal avoidance strategy through the action value function; determines to avoid the waiting transport vehicle through the optimal avoidance strategy, calculates the total waiting time and waiting space position of the waiting transport vehicle, and executes the avoidance strategy; The transport vehicle route planning model is also used to re-plan an optimal operation path that avoids waiting transport vehicles.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A non-transitory computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images