Multi-Intelligent Vehicle Collaborative Delivery Method and System

Abstract

This invention discloses a method and system for collaborative material delivery using multiple intelligent vehicles. The method includes: a task management device responding to an emergency task delivery request; obtaining a set of delivery tasks and task information based on the emergency task delivery request; sending the task information to the material management device and each intelligent vehicle; constructing a task matching graph based on material configuration information, task information, and task matching degree; obtaining the task nodes to be delivered and available vehicle nodes in the task matching graph; splitting the task nodes to be delivered into multiple sub-task nodes; constructing a task matching matrix based on the sub-task nodes and available vehicle nodes; determining an emergency task allocation strategy based on the task matching matrix; and sending the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can perform emergency material delivery based on the emergency task allocation strategy, thereby achieving accurate task allocation, improving the collaborative delivery capability of intelligent vehicles, and ensuring the timeliness of material delivery.

Description

Technical Field

[0001] This invention relates to the field of intelligent vehicle technology, and in particular to a method and system for collaborative delivery of goods by multiple intelligent vehicles. Background Technology

[0002] With the acceleration of globalization and urbanization, emergency logistics plays an increasingly important role in responding to emergencies such as natural disasters and public health events. Emergency logistics tasks typically require supplies to be delivered to designated destinations in the shortest possible time and along the optimal route. This places extremely high demands on logistics management and scheduling, requiring not only accurate decision-making in the shortest possible time but also consideration of potential traffic changes, unforeseen events, and various unpredictable factors. Therefore, how to complete emergency delivery tasks quickly and efficiently under limited time and resources has become an important topic in modern logistics research and application. Traditional emergency logistics systems usually rely on manual dispatching and single-vehicle delivery. This approach often proves inefficient and slow to respond when faced with large-scale, multi-point emergency needs. The limitations of traditional systems become even more apparent when dealing with a wide variety of supplies and tight delivery deadlines.

[0003] In recent years, the development of intelligent vehicle technology has brought new solutions to emergency logistics systems. Intelligent vehicles, by integrating advanced sensors, navigation systems, and communication technologies, can achieve autonomous navigation, real-time obstacle avoidance, and multi-vehicle collaborative operations. The application of these technologies not only improves delivery efficiency but also enhances the system's flexibility and reliability. However, existing multi-intelligent vehicle collaborative delivery systems suffer from problems such as a lack of dynamic adaptability in task allocation, route planning, and fault handling; imprecise task allocation; low route planning efficiency; insufficient obstacle avoidance capabilities; delayed fault response; and inflexible task transfer. Summary of the Invention

[0004] The main objective of this invention is to provide a method and system for collaborative delivery of goods using multiple intelligent vehicles, aiming to solve the technical problems of lack of dynamic adaptability, imprecise task allocation, low route planning efficiency, poor delivery timeliness, and delayed response in the existing technology of collaborative delivery using multiple intelligent vehicles.

[0005] To achieve the above objectives, the present invention provides a multi-intelligent vehicle collaborative material delivery method. This method is applied to a task management device, which is communicatively connected to a material management device and multiple intelligent vehicles. The multi-intelligent vehicle collaborative material delivery method includes:

[0006] In response to an emergency delivery request, a set of delivery tasks and task information are obtained based on the emergency delivery request, and the task information is sent to the material management equipment and each intelligent vehicle. The set of delivery tasks includes at least one emergency delivery task.

[0007] A task matching graph is constructed based on material configuration information, task information, and task matching degree. The material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between the task nodes and the vehicle nodes, edge weights between the material nodes and the vehicle nodes, and node states of the task nodes and the vehicle nodes.

[0008] A breadth-first search is performed on the task matching graph. Based on the node status, the task nodes to be delivered and the available vehicle nodes in the task matching graph are obtained, and the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node are obtained.

[0009] Based on the required load capacity and the remaining load capacity information, the task node to be delivered is split into multiple sub-task nodes;

[0010] Construct a task matching matrix based on the sub-task nodes and the available vehicle nodes;

[0011] An emergency task allocation strategy is determined based on the task matching matrix, and the emergency task allocation strategy is sent to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

[0012] Optionally, constructing the task matching matrix based on the sub-task nodes and the available vehicle nodes includes:

[0013] Based on the subtask node, a matching degree update request is sent to each available vehicle node, and the updated task matching degree of each available vehicle node based on the subtask node is obtained.

[0014] Construct an initial matching matrix based on the updated task matching degree:

[0015]

[0016] in, Represents the initial matching matrix. Representative task With vehicles The degree of task matching between them;

[0017] The initial matching matrix is ​​reduced to obtain the target matrix;

[0018] The target matrix is ​​adjusted based on a dynamic zero-element update strategy to obtain a task matching matrix. The dynamic zero-element update strategy includes:

[0019]

[0020] in, This is the adjusted task matching matrix. For the target matrix, To adjust the coefficient, Used to control the adjustment range The minimum non-zero matching degree value in the target matrix.

[0021] Optionally, reducing the initial matching matrix to obtain the target matrix includes:

[0022] Calculate adaptive weights based on the adaptive weight adjustment function:

[0023]

[0024] in, Representative vehicle With the task The degree of matching between them Represents the first element in the initial matching matrix. i The maximum matching degree in the row, Represents a constant. Represents adaptive weights;

[0025] The initial matching matrix is ​​reduced according to the adaptive weights to obtain a candidate matrix. The reduction is performed using the following formula:

[0026]

[0027] in, Represents the matching degree after reduction and adjustment. This represents the matching degree before adjustment in the initial matching matrix;

[0028] The candidate matrix is ​​adjusted based on the time window constraint to obtain the target matrix:

[0029]

[0030] in, This represents the degree of matching after adjusting the time window constraints. Representative vehicle Execute the task The estimated completion time, For the task The latest deadline stipulated The time window penalty coefficient, Used to control the severity of timeout penalties.

[0031] Optionally, the step of determining an emergency task allocation strategy based on the task matching matrix and sending the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can distribute emergency supplies based on the emergency task allocation strategy includes:

[0032] A task allocation model is constructed based on the task matching matrix, and the task allocation model includes:

[0033]

[0034]

[0035]

[0036]

[0037] in, Representative vehicle With the task The allocation status between them Representative vehicle With the task The degree of matching between them Represents the set of task nodes to be delivered. Represents the set of vehicle nodes. This represents the vehicle's maximum load capacity. This represents the required payload capacity for the task.

[0038] An emergency task allocation strategy is determined based on the task allocation model, and the emergency task allocation strategy is sent to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

[0039] Furthermore, to achieve the above objectives, this invention also proposes a multi-intelligent vehicle collaborative material delivery method. This method is applied to intelligent vehicles within an intelligent vehicle cluster, where the cluster comprises multiple intelligent vehicles. These intelligent vehicles are communicatively connected to a task management device and a material management device. The method includes:

[0040] In response to the task information sent by the task management device, the system determines the task matching degree between itself and each emergency delivery task based on the task information.

[0041] The task matching degree is sent to the task management device, and delivery messages are monitored. The delivery messages are sent by the task management device and include emergency task allocation strategies.

[0042] Upon receiving the delivery message, emergency supplies are delivered based on the emergency task allocation strategy in the delivery message.

[0043] Optionally, determining the task matching degree between oneself and each emergency delivery task based on the task information includes:

[0044] Based on the task information, determine the location of the material point, the location of the task point, and the required load capacity of the task.

[0045] The vehicle's task matching degree with each emergency delivery task is determined based on its maximum load capacity, the location of the material point, the location of the task point, and the required load capacity of the task.

[0046]

[0047] in, Indicates vehicle Distance to supply points This indicates the distance between the supply point and the mission point. This represents the vehicle's maximum load capacity. This represents the required payload capacity for the task. For adjustment coefficients, These are used to balance priority, operating cost, and load matching, respectively. As a task priority, Represents the vehicle's location. Represents the location of the supply point. This represents the location of the task point.

[0048] Optionally, after sending the task matching degree to the task management device, the method further includes:

[0049] Monitor matching score update requests;

[0050] Upon receiving a matching score update request, the task matching score is updated based on the sub-task nodes in the matching score update request to obtain the updated task matching score:

[0051]

[0052] in, Indicates vehicle sub-tasks The degree of matching between them For subtasks The task requires a certain load capacity;

[0053] The updated task matching score is sent to the task management device.

[0054] Optionally, after distributing emergency supplies based on the emergency task allocation strategy in the delivery message, the method further includes:

[0055] A raster map is constructed by raster mapping based on point cloud data collected by sensors during the emergency supplies distribution process;

[0056] Determine the occupancy probability of each grid cell in the grid map:

[0057]

[0058] in, Represents a grid The likelihood of being occupied. Represents a grid Prior probability of being occupied Represents the normalization factor. Represents a grid The probability of occupancy, This indicates the current sensor measurement result. This indicates historical sensor measurement results;

[0059] Based on the occupancy probability, a passage analysis is performed on the grid map to determine the currently passable area in the grid map;

[0060] Based on the currently passable area, topological nodes, topological edges between adjacent topological nodes, and edge weights of the topological edges are determined. The edge weights are calculated based on the following formula:

[0061]

[0062] in, Represents edge weight, Indicates adjacent nodes and The Euclidean distance between them For the driving speed of intelligent vehicles, Traffic correction factor The obstacle influence factor, Represents a node and The accident rate of the road sections between them within the same time period Indicates the number of accidents. Indicates the number of times a vehicle passes through. and Indicates the adjustment factor;

[0063] A topology map is constructed based on the topology nodes, the topology edges, and the edge weights;

[0064] Based on the topology map and the emergency task allocation strategy, path planning is performed, and emergency supplies are distributed based on the path planning results.

[0065] Optionally, the route planning based on the topology map and the emergency task allocation strategy includes:

[0066] The dynamic weight adjustment factor for the current node is determined based on the topology map and the emergency task allocation strategy:

[0067]

[0068] in, Indicates the dynamic weight adjustment factor. and These are preset fixed parameter values. Indicates the starting point and the current node The distance between them Indicates the current node Distance to the target node;

[0069] Cost evaluation is performed based on a dynamic weight adjustment factor to determine the total delivery cost for the current node:

[0070]

[0071] in, Indicates the current node n Total delivery cost This represents the actual cost from the starting point to the current node. Indicates the current node n The heuristic cost of reaching the target point;

[0072] Based on the total delivery cost, perform global path planning to generate an initial delivery path;

[0073] The initial delivery route is smoothed to obtain the target delivery route, and route planning is performed based on the target delivery route.

[0074] In addition, to achieve the above objectives, the present invention also proposes a multi-intelligent vehicle material collaborative delivery system, which includes a task management device, a material management device, and multiple intelligent vehicles, wherein the task management device communicates with the material management device and the multiple intelligent vehicles.

[0075] The task management device is used to respond to an emergency task delivery request, obtain a set of delivery tasks and task information based on the emergency task delivery request, and send the task information to the material management device and each intelligent vehicle. The set of delivery tasks includes at least one emergency delivery task.

[0076] The material management device is used to configure materials based on the task information, generate material configuration information, and send the material configuration information to the task management device.

[0077] The intelligent vehicle is used to respond to the task information sent by the task management device and determine the task matching degree between itself and each emergency delivery task based on the task information;

[0078] The intelligent vehicle is also used to send the task matching degree to the task management device and monitor delivery messages, which are sent by the task management device and include emergency task allocation strategies.

[0079] The task management device is also used to construct a task matching graph based on material configuration information, task information, and task matching degree. The material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between the task nodes and the vehicle nodes, edge weights between the material nodes and the vehicle nodes, and node states of the task nodes and the vehicle nodes.

[0080] The task management device is also used to perform a breadth-first search on the task matching graph, obtain the task nodes to be delivered and available vehicle nodes in the task matching graph based on the node status, and obtain the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node.

[0081] The task management device is also used to split the task node to be delivered based on the task requirement load capacity and the remaining load capacity information to obtain multiple sub-task nodes;

[0082] The task management device is also used to construct a task matching matrix based on the sub-task nodes and the available vehicle nodes;

[0083] The task management device is also used to determine an emergency task allocation strategy based on the task matching matrix and send the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

[0084] The intelligent vehicle is also used to deliver emergency supplies based on the emergency task allocation strategy in the delivery message when it receives the delivery message.

[0085] In addition, to achieve the above objectives, this application also proposes a task management device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the multi-intelligent vehicle collaborative material delivery method described above.

[0086] In addition, to achieve the above objectives, this application also proposes an intelligent vehicle, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the multi-intelligent-vehicle collaborative material delivery method described above.

[0087] In addition, to achieve the above objectives, this application also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the multi-intelligent vehicle collaborative material delivery method described above.

[0088] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the multi-intelligent vehicle material collaborative delivery method described above.

[0089] This invention responds to emergency delivery requests via a task management device. Based on the emergency delivery request, it acquires a set of delivery tasks and task information, and sends the task information to a materials management device and each intelligent vehicle. The delivery task set includes at least one emergency delivery task. The materials management device configures materials based on the task information, generates materials configuration information, and sends the materials configuration information to the task management device. Each intelligent vehicle responds to the task information sent by the task management device, determines its own task matching degree with each emergency delivery task based on the task information, sends the task matching degree to the task management device, and monitors delivery messages. The delivery messages are sent by the task management device and include an emergency task allocation strategy. The task management device constructs a task matching graph based on the materials configuration information, task information, and task matching degree. The materials configuration information is generated by the materials management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, materials nodes, edge weights between task nodes and vehicle nodes, and edge weights between materials nodes and vehicles. The edge weights between nodes and the node states of the task nodes and vehicle nodes are used to perform a breadth-first search on the task matching graph. Based on the node states, the task nodes to be delivered and available vehicle nodes in the task matching graph are obtained, along with the required load capacity of each task node to be delivered and the remaining load capacity of each available vehicle node. Based on the required load capacity and the remaining load capacity, the task nodes to be delivered are split into multiple sub-task nodes. A task matching matrix is ​​constructed based on the sub-task nodes and the available vehicle nodes. An emergency task allocation strategy is determined based on the task matching matrix and sent to each intelligent vehicle so that each intelligent vehicle can deliver emergency supplies based on the emergency task allocation strategy. When an intelligent vehicle receives a delivery message, it performs emergency supply delivery based on the emergency task allocation strategy in the delivery message. Because this invention splits the delivery task into multiple sub-tasks, constructs a matching matrix based on the sub-tasks, and allocates tasks based on the matching matrix, it achieves refined allocation of delivery tasks, ensuring the rationality of task allocation. Path planning is performed based on the emergency task allocation strategy, thereby improving the collaborative delivery capability of intelligent vehicles and ensuring the timeliness of emergency supply delivery. Attached Figure Description

[0090] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0091] Figure 1 This is a schematic diagram of the structure of a multi-intelligent vehicle collaborative material delivery device in the hardware operating environment of the embodiment of the present invention;

[0092] Figure 2 This is a flowchart illustrating an embodiment of the multi-intelligent vehicle collaborative material delivery method of the present invention;

[0093] Figure 3 This is a flowchart illustrating an embodiment of the multi-intelligent vehicle collaborative material delivery method of the present invention;

[0094] Figure 4 This is a flowchart illustrating the intelligent vehicle path planning process in an embodiment of the multi-intelligent vehicle collaborative material delivery method of the present invention.

[0095] Figure 5 This is a structural block diagram of an embodiment of the multi-intelligent vehicle material collaborative delivery system of the present invention.

[0096] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0097] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0098] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of a multi-intelligent vehicle collaborative material delivery equipment in the hardware operating environment of the embodiment of the present invention.

[0099] like Figure 1 As shown, the multi-vehicle collaborative material delivery device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk storage device. Optionally, the memory 1005 may also be a storage system independent of the aforementioned processor 1001.

[0100] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on multi-intelligent vehicle material delivery equipment, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0101] like Figure 1 As shown, the memory 1005, which is a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a multi-intelligent vehicle material collaborative delivery program.

[0102] exist Figure 1 In the multi-vehicle collaborative delivery device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the multi-vehicle collaborative delivery device of the present invention can be set in the multi-vehicle collaborative delivery device, and the multi-vehicle collaborative delivery device calls the multi-vehicle collaborative delivery program stored in the memory 1005 through the processor 1001 and executes the multi-vehicle collaborative delivery method provided in the embodiment of the present invention.

[0103] This invention provides a method for collaborative delivery of goods using multiple intelligent vehicles, referring to... Figure 2 , Figure 2 This is a flowchart illustrating an embodiment of the multi-intelligent vehicle collaborative material delivery method of the present invention.

[0104] In this embodiment, the multi-intelligent vehicle material collaborative delivery method is applied to a task management device. The task management device communicates with the material management device and multiple intelligent vehicles. The multi-intelligent vehicle material collaborative delivery method includes the following steps:

[0105] Step S10: In response to the emergency task delivery request, obtain the delivery task set and task information based on the emergency task delivery request, and send the task information to the material management equipment and each intelligent vehicle.

[0106] In some embodiments, the material management device is responsible for equipping the intelligent vehicle with materials and calculating the remaining delivery material quantity, ultimately feeding the results back to the task management device. Upon receiving a new task, the intelligent vehicle calculates the matching degree between the task and the vehicle based on its own status and feeds the result back to the task management device. The task management device constructs a task-vehicle-material point matching map and uses an improved Hungarian algorithm for efficient task allocation, ultimately broadcasting the task allocation results. During the delivery process, the intelligent vehicle combines the construction and improvement of a hybrid map... The algorithm performs path planning while incorporating virtual potential field and virtual impedance models to ensure local obstacle avoidance and dynamic path adjustment. When the intelligent vehicle malfunctions, the system triggers an early warning mechanism on the emergency management side. Based on the priority scores of task transfer and support vehicles, it automatically selects the best task transfer vehicle and support vehicle to ensure that the malfunctioning vehicle receives rapid support and continues to complete the delivery task.

[0107] It should be understood that the executing entity of this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a computer, or a terminal electronic device capable of performing the above functions. The following description uses a task management device as an example to illustrate this embodiment and the subsequent embodiments.

[0108] It should be noted that the set of delivery tasks includes at least one emergency delivery task, which can be an emergency logistics material delivery task. The task information may include the delivery destination, the type of materials required, the quantity of materials, the delivery time limit, and other specific requirements (such as storage temperature requirements, special loading and unloading requirements, etc.).

[0109] In some embodiments, upon receiving a task request (a request sent by the task management device to the material management device based on task information), the material management device begins preparing and scheduling materials. The material management device checks its existing inventory to ensure sufficient materials are available for delivery, based on task requirements. If inventory is sufficient, the device further confirms the type, specifications, and quantity of materials to ensure they meet delivery requirements. If inventory is insufficient, the device sends replenishment requests to suppliers or other warehouses to ensure task execution is not affected by material shortages. If sufficient materials are confirmed, the device pre-sorts and packages them. Once all materials are assembled, the material management device sends a material preparation completion message to the task management device, providing a detailed material list and assembly information, including material type, quantity, and weight, to prepare for subsequent task allocation.

[0110] Step S20: Construct a task matching graph based on material allocation information, task information, and task matching degree.

[0111] It should be noted that the material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between the task nodes and vehicle nodes, edge weights between the material nodes and vehicle nodes, and the node states of the task nodes and vehicle nodes. Here, the task nodes represent emergency delivery tasks, the vehicle nodes represent intelligent vehicles, and the material nodes represent material points.

[0112] It should be noted that the above node status includes the node status of task nodes and the node status of vehicle nodes. For example, two node statuses can be set: for task nodes, 1 indicates that the task has been assigned and is being delivered, and 0 indicates that the task is to be assigned; for vehicle nodes, 1 indicates that the vehicle has reached its maximum load, and 0 indicates that the vehicle has not reached its maximum load.

[0113] Understandably, after the task management device sends task information to the material management device, it monitors the material preparation completion message. Upon receiving the material preparation completion message, it obtains the material configuration information based on the message.

[0114] It should be understood that after the task management device sends task information to each intelligent vehicle, it monitors the task matching degree sent by the intelligent vehicle and constructs a task matching map based on the task matching degree, material configuration information and task information sent by each intelligent vehicle.

[0115] In some embodiments, the task management device performs task allocation. First, a task-vehicle-material point matching graph is constructed, and task nodes are defined. Vehicle node Material distribution points Each task corresponds to a resource point, and the edge weight between the task node and the resource point node represents the quantity of resources that the task needs to deliver. Each vehicle may have multiple delivery tasks, meaning each vehicle may correspond to multiple resource nodes and task nodes. Need to go to the supply point Once supplies are received, vehicles will be established. Supply Point The edge weight represents the vehicle. The transportation cost from the current location to the supply point; if the vehicle Delivery tasks required Then establish vehicle With the task The edges are defined, and their weights represent the matching degree between vehicles and tasks. Two node states are set: for task nodes, 1 indicates a task has been assigned and is being delivered, and 0 indicates a task is pending assignment; for vehicle nodes, 1 indicates a vehicle has reached its maximum load, and 0 indicates a vehicle has not reached its maximum load.

[0116] Step S30: Perform a breadth-first search on the task matching graph, obtain the task nodes to be delivered and available vehicle nodes in the task matching graph based on the node status, and obtain the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node.

[0117] It should be noted that the task node to be delivered can be a task node with a node status of 0 (i.e., a task that needs to be assigned a vehicle for delivery); the available vehicle node can be a vehicle node with a node status of 0 (i.e., a vehicle that has not reached its maximum load).

[0118] Understandably, the task management device uses breadth-first search to search for all task nodes and vehicle nodes with a state of 0 in the task-vehicle-resource point matching graph. Assume there are m task nodes and n vehicle nodes. If a certain task... Delivery can be completed using a single smart vehicle within the vehicle node; this vehicle should be prioritized for assignment. If the task Quantity of goods to be delivered The load exceeds the current remaining load of any vehicle in set V. If so, the task is split up and completed by multiple vehicles.

[0119] Step S40: Based on the required load capacity and the remaining load capacity information, the task node to be delivered is split into multiple sub-task nodes.

[0120] It should be noted that the task management device can classify each task node to be delivered and determine the first task node and the second task node. The first task node is the task node that does not need to be split (i.e., the task node that can be delivered by a single smart vehicle), and the second task node is the task node that needs to be split (i.e., the task node that requires multiple smart vehicles to work together for delivery).

[0121] In some embodiments, the task management device can directly match the first task node with an available vehicle node that meets the load capacity required for the task (or directly match based on the matching degree fed back by the intelligent vehicle), split the second task node into multiple sub-task nodes, obtain the updated matching degree of the intelligent vehicle for the sub-task nodes, and allocate tasks to the sub-task nodes based on the updated matching degree.

[0122] Understandably, the task management equipment can determine the delivery task nodes based on the task's required load capacity and the remaining load capacity information of each available vehicle node. The minimum number of vehicles required can be determined by referring to the following formula for the delivery task nodes. Split into Sub-tasks:

[0123]

[0124] The task Split into Each sub-task contains the following resources:

[0125]

[0126] Task-vehicle allocation forms a matching matrix, and the matching degree is adjusted according to the new sub-task delivery material requirements. Therefore, in some embodiments, the task management device sends the sub-task node to each available vehicle node so that the smart vehicle updates the matching degree and determines the matching degree between each available vehicle node and the sub-task node.

[0127] Step S50: Construct a task matching matrix based on the subtask nodes and the available vehicle nodes.

[0128] It should be noted that the matching degree between the intelligent vehicle computing of each available vehicle node and the subtask nodes... This is done to update the matching degree, and the updated matching degree results are fed back to each task management device. The task management device then constructs a matching matrix based on the updated matching degree.

[0129] Furthermore, in order to accurately construct the task matching matrix, in some embodiments, step S50 may include:

[0130] Step S501: Send a matching degree update request to each available vehicle node based on the sub-task node, and obtain the updated task matching degree of each available vehicle node based on the sub-task node;

[0131] Step S502: Construct an initial matching matrix based on the updated task matching degree;

[0132] Step S503: Reduce the initial matching matrix to obtain the target matrix;

[0133] Step S504: Adjust the target matrix based on the dynamic update zero-element strategy to obtain the task matching matrix.

[0134] It should be noted that the matching degree between the intelligent vehicle computing of each available vehicle node and the subtask nodes... This is done to update the matching degree, and the updated matching degree results are fed back to each task management device. The task management device then constructs the following matching matrix based on the updated matching degree:

[0135]

[0136] in, Represents the initial matching matrix. Representative task With vehicles The degree of task matching between them.

[0137] It should be noted that after completing row and column reduction, it is checked whether there are enough zero elements to match the tasks. If the matching is complete, the process ends directly; if not all tasks are matched, a dynamic zero-element update strategy is executed to gradually reduce non-zero elements. The dynamic zero-element update strategy includes:

[0138]

[0139] in, This is the adjusted task matching matrix. For the target matrix, To adjust the coefficient, Used to control the adjustment range, for example It can take the value 0.5. The minimum non-zero matching degree value in the target matrix.

[0140] In some embodiments, the task management device adjusts the target matrix based on a dynamic zero-element update strategy to obtain a task matching matrix, which may be a binary matrix. X ,in, Subtasks Assigned to vehicles , Subtasks Not assigned to vehicles And calculate the actual amount of materials carried by each smart vehicle:

[0141]

[0142] If a vehicle's load exceeds its maximum load capacity The vehicle retains only its maximum load capacity. Overload Assign it to another vehicle. Iterate through all currently available vehicles; if a vehicle... There is still enough remaining load capacity Then it is directly allocated to Conversely, if the task assignment is not complete, a second split is performed. After the task assignment is completed, the split sub-task nodes are added to the task-vehicle-supply point matching graph, and the status of the task nodes and vehicle nodes is updated.

[0143] Furthermore, in order to effectively avoid matching imbalance and improve matching accuracy, in some embodiments, step S503 above may include:

[0144] Step S5031: Calculate the adaptive weights based on the adaptive weight adjustment function;

[0145] Step S5032: Reduce the initial matching matrix according to the adaptive weights to obtain the candidate matrix;

[0146] Step S5033: Adjust the candidate matrix based on the time window constraint to obtain the target matrix.

[0147] In some embodiments, the task management device may use an improved Hungarian algorithm on the initial matching matrix, as follows:

[0148] During row reduction, an adaptive weight adjustment function is introduced. Tasks with a higher matching degree are less affected by the adjustment, while tasks with a lower matching degree are adjusted more significantly to reduce potential mismatches. This is achieved through an adaptive weight adjustment function. Refer to the following formula:

[0149]

[0150] in, Representative vehicle With the task The degree of matching between them Represents the first element in the initial matching matrix. i The maximum matching degree in the row, Represents a constant. This represents adaptive weights.

[0151] To ensure that task allocation still favors tasks with higher matching scores during reduction, while also considering tasks with lower matching scores to avoid matching imbalance, the improved formula for row reduction is as follows:

[0152]

[0153] in, Represents the matching degree after reduction and adjustment. This represents the matching degree before adjustment in the initial matching matrix.

[0154] In some embodiments, a time window constraint is used to adjust the matching degree, prioritizing vehicles that can complete the task within the specified time window. For tasks that are completed after the time limit, their matching degree is reduced, decreasing the probability of them being selected. The time window constraint condition for adjusting the matching degree refers to the following formula:

[0155]

[0156] in, This represents the degree of matching after adjusting the time window constraints. Representative vehicle Execute the task The estimated completion time, For the task The latest deadline stipulated The time window penalty coefficient, Used to control the severity of timeout penalties.

[0157] Step S60: Determine an emergency task allocation strategy based on the task matching matrix, and send the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

[0158] In some embodiments, when a single new task occurs If there are still vehicles with remaining load, then calculate the remaining load vehicles and the new task. Matching degree between Vehicles are sorted in descending order:

[0159]

[0160]

[0161] in, W The amount of supplies required to be delivered for each vehicle. To update the remaining required delivery supplies. If all vehicles with current remaining load are used and the supplies are still not fully allocated, i.e. Alternatively, when a new task arises and there are no remaining vehicles with available load, the priority of the new task is compared with the priority of the task set in the vehicles. (Assuming the vehicle...) The current set of tasks carried is Then the vehicle priority is: .

[0162] If the priority of the new task is lower than the priority of the task set in the vehicle, then wait for an idle vehicle; if the priority of the new task is lower than the priority of the task set in the vehicle, then wait for an idle vehicle. Larger than some vehicles If the priority of the task set is determined, vehicle preemption will be performed. The matching degree between these available vehicles and new tasks will be calculated. The vehicles are sorted according to their matching degree, and each vehicle is assigned the maximum amount of resources required for the new mission until all the resources required for the new mission have been allocated.

[0163] vehicle The assigned vehicles immediately returned to the nearest supply point, unloaded all supplies, and began their new mission.

[0164] When one or more emergencies occur, multiple new tasks are added. These new tasks... Add the vehicle to the task set T and update the task-vehicle-supply point matching graph. Using breadth-first search, search all vehicle nodes with a status of 0 in the task-vehicle-supply point matching graph. Determine whether to use a single vehicle or multiple vehicles based on the required delivery quantity for each task and the vehicle status. If multiple vehicles are used, calculate the new task according to the task matching degree calculation formula from step 12. The matching degree between these vehicles is used to construct a matching matrix, and the optimal task-vehicle matching is calculated using an improved Hungarian algorithm.

[0165] After the task is successfully completed, the task management device deletes the task node from the task-vehicle-resource point matching graph, cancels all vehicles related to the completion of the task, and updates the vehicle status.

[0166] In some embodiments, to improve task allocation efficiency and the rationality of collaborative delivery, step S60 above may include:

[0167] Step S601: Construct a task allocation model based on the task matching matrix;

[0168] Step S602: Determine the emergency task allocation strategy according to the task allocation model, and send the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

[0169] It should be noted that in emergency logistics scenarios, an emergency typically requires multiple different supplies, meaning there will be multiple allocation tasks simultaneously. Assuming T is the set of tasks to be allocated, and V is the set of vehicles that have not yet reached their maximum capacity, the task allocation problem can be defined as a task allocation model, which includes:

[0170]

[0171]

[0172]

[0173]

[0174] in, Representative vehicle With the task The allocation status between them Representative vehicle With the task The degree of matching between them Represents the set of task nodes to be delivered. Represents the set of vehicle nodes. This represents the vehicle's maximum load capacity. This represents the required load capacity for the task.

[0175] This embodiment responds to emergency delivery requests by acquiring a set of delivery tasks and task information based on the requests, and sending the task information to the material management device and each intelligent vehicle. The delivery task set includes at least one emergency delivery task. A task matching graph is constructed based on material configuration information, task information, and task matching degree. The material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between task nodes and vehicle nodes, edge weights between material nodes and vehicle nodes, and node states of task nodes and vehicle nodes. A breadth-first search is performed on the task matching graph. Based on the node status, obtain the task nodes to be delivered and available vehicle nodes in the task matching graph, and obtain the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node; based on the task load requirement and the remaining load information, split the task nodes to be delivered to obtain multiple sub-task nodes; construct a task matching matrix based on the sub-task nodes and the available vehicle nodes; determine an emergency task allocation strategy based on the task matching matrix, and send the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can carry out emergency material delivery based on the emergency task allocation strategy, thereby realizing the fine allocation of delivery tasks, ensuring the rationality of task allocation, and performing path planning based on the emergency task allocation strategy, thereby improving the collaborative delivery capability of intelligent vehicles and ensuring the timeliness of emergency material delivery.

[0176] refer to Figure 3 , Figure 3 This is a flowchart illustrating an embodiment of the multi-intelligent vehicle collaborative material delivery method of the present invention.

[0177] In this embodiment, the multi-intelligent vehicle collaborative material delivery method is applied to intelligent vehicles in an intelligent vehicle cluster. The intelligent vehicle cluster includes multiple intelligent vehicles, and the multiple intelligent vehicles are communicatively connected with task management equipment and material management equipment. The method includes:

[0178] Step S1: In response to the task information sent by the task management device, determine the task matching degree between itself and each emergency delivery task based on the task information.

[0179] Understandably, the intelligent vehicle calculates the matching degree between itself and the new task based on its current situation and feeds back the matching degree result to the task management device.

[0180] In some embodiments, the intelligent vehicle calculates the vehicle-task matching degree based on task priority (which varies over time), vehicle operating costs, and the vehicle's maximum load capacity (or remaining load capacity).

[0181] Furthermore, in order to accurately calculate the task matching degree, in some embodiments, step S1 above may include:

[0182] Step S11: Determine the location of the material point, the location of the task point, and the required load capacity based on the task information;

[0183] Step S12: Determine the task matching degree between the vehicle and each emergency delivery task based on the vehicle's maximum load capacity, the location of the material point, the location of the task point, and the required load capacity of the task.

[0184] It should be noted that the intelligent vehicle can be based on the location of the material point, the location of the task point, and the task priority. (Varies over time) and vehicle operating costs Calculate the vehicle-task matching degree. The calculation formula is as follows:

[0185]

[0186] in, Indicates vehicle Distance to supply points This indicates the distance between the supply point and the mission point. This represents the vehicle's maximum load capacity. This represents the required payload capacity for the task. For adjustment coefficients, These are used to balance priority, operating cost, and load matching, respectively. As a task priority, Represents the vehicle's location. Represents the location of the supply point. This represents the location of the task point.

[0187] Step S2: Send the task matching degree to the task management device and monitor the delivery message.

[0188] It should be noted that the delivery message is sent by the task management device, and the delivery message includes an emergency task allocation strategy. When the intelligent vehicle receives the delivery message, it determines that the task management device has completed task allocation, identifies the target task based on the emergency task allocation strategy, and then proceeds to the material management device to load materials before delivering them.

[0189] Furthermore, in order to accurately calculate the matching degree between the task and the subtask, and thus update the task matching degree, in some embodiments, after sending the task matching degree to the task management device, the method further includes:

[0190] Step S21: Monitor matching score update requests;

[0191] Step S22: Upon receiving a matching degree update request, update the task matching degree based on the sub-task nodes in the matching degree update request to obtain the updated task matching degree;

[0192] Step S23: Send the updated task matching score to the task management device.

[0193] In some embodiments, the task management device will assign tasks Split into Each sub-task contains the following resources: The task-vehicle allocation will form a matching matrix. The matching degree will be adjusted according to the new sub-task's material delivery requirements. The calculation formula is as follows:

[0194]

[0195] in, Indicates vehicle sub-tasks The degree of matching between them For subtasks The task requires a certain load capacity.

[0196] Step S3: Upon receiving the delivery message, emergency supplies are delivered based on the emergency task allocation strategy in the delivery message.

[0197] It should be noted that the matching degree update request is sent by the task management device. After the task management device breaks down the second task node (i.e. the task that requires multiple smart vehicles to coordinate delivery) into multiple sub-tasks, the smart vehicles need to recalculate the task matching degree between themselves and the sub-tasks in order to reasonably formulate a collaborative delivery plan for the sub-tasks.

[0198] In some embodiments, the intelligent vehicle assigned to a task first assembles supplies at a material depot and then proceeds to the task location to deliver the supplies. The intelligent vehicle collects environmental data using LiDAR, GPS+IMU, and cameras, and converts the point cloud data into a raster map using an occupancy grid mapping method to detect obstacles and passable areas. Simultaneously, a topology map is constructed based on road connectivity analysis, and segment weights are defined for global path search. Global planning employs an improved heuristic search algorithm, combined with dynamic weight adjustment factors and B-spline curves based on curvature constraints to optimize and smooth the path. Local planning guides the intelligent vehicle along the global path using a virtual potential field model; a virtual attraction field guides the vehicle towards the target, a virtual repulsion field avoids obstacles, and a virtual navigation field adjusts the path based on traffic information. Furthermore, a virtual impedance model further smooths the vehicle's movement and performs dynamic obstacle avoidance adjustments. The entire process combines global path planning and local dynamic obstacle avoidance strategies to ensure the intelligent vehicle can complete its tasks efficiently and smoothly in complex environments.

[0199] Furthermore, in order to improve the transportation efficiency of intelligent vehicles, enhance the accuracy of transportation routes, and improve obstacle avoidance capabilities, refer to Figure 4 , Figure 4 This is a flowchart illustrating the intelligent vehicle path planning process in one embodiment. In some embodiments, after step S3, the process further includes:

[0200] Step S4: Based on the point cloud data collected by sensors during the emergency supplies delivery process, perform raster mapping to construct a raster map.

[0201] In some embodiments, the intelligent vehicle obtains point cloud data via LiDAR and acquires obstacle information based on the point cloud data; it also obtains vehicle location information via GPS+IMU; and cameras are used to assist in detecting passable areas. Occupied grid mapping is used to convert the point cloud data into a grid map for fine-grained local obstacle avoidance.

[0202] Step S5: Determine the occupancy probability of each grid cell in the grid map.

[0203] It should be noted that, in order to accurately analyze the passable areas in the grid map, the intelligent vehicle can calculate the occupancy probability of each grid cell in the grid map, referring to the following formula:

[0204]

[0205] in, Represents a grid The likelihood of being occupied. Represents a grid Prior probability of being occupied Represents the normalization factor. Represents a grid The probability of occupancy, This indicates the current sensor measurement result. This indicates historical sensor measurement results.

[0206] In some embodiments, if a certain grid If the probability of an object being occupied is greater than 0.7, it is set as an obstacle M(x,y)=1; otherwise, it is set as a passable area M(x,y)=0.

[0207] Step S6: Perform a passage analysis on the grid map based on the occupancy probability to determine the currently passable area in the grid map.

[0208] In some embodiments, the intelligent vehicle can perform neighborhood traversal on the grid map based on the occupancy probability, analyze the passable direction of the current node, detect key intersections (T-shaped intersections, cross intersections, etc.), and determine the current passable area in the grid map based on key intersections, material points, and task points.

[0209] Step S7: Determine the topology nodes, the topology edges between adjacent topology nodes, and the edge weights of the topology edges based on the currently passable area.

[0210] It should be noted that the edge weights are calculated based on the following formula:

[0211]

[0212] in, Represents edge weight, Indicates adjacent nodes and The Euclidean distance between them For the driving speed of intelligent vehicles, Traffic correction factor The obstacle influence factor, Represents a node and The accident rate of the road sections between them within the same time period Indicates the number of accidents. Indicates the number of times a vehicle passes through. and This represents the adjustment factor.

[0213] In some embodiments, the traffic correction factor is determined based on the road congestion status, taking a value of 1.0 for no congestion, 0.7 for light congestion, and 0.5 for heavy congestion. The obstacle factor takes a value of 0 for no obstacles, 1 for light obstacles, and 5 for complete closure.

[0214] Step S8: Construct a topology map based on the topology nodes, the topology edges, and the edge weights.

[0215] It should be noted that the intelligent vehicle uses key intersections, supply points, and task points as nodes V in the topology map. Based on road connectivity analysis, topological edges E are established between adjacent nodes, with edge weights representing the cost of moving from adjacent node i to node j. The topology map constructed in this way... Used for global planning, efficiently searching global paths.

[0216] Step S9: Perform route planning based on the topology map and the emergency task allocation strategy, and continue emergency material distribution based on the route planning results.

[0217] In some embodiments, the intelligent vehicle may employ an improved topology map. The algorithm calculates the globally optimal path from the starting point to the ending point.

[0218] Furthermore, in order to perform accurate path planning, in some embodiments, step S9 above may include:

[0219] Step S91: Determine the dynamic weight adjustment factor of the current node based on the topology map and the emergency task allocation strategy;

[0220] Step S92: Evaluate the cost based on the dynamic weight adjustment factor to determine the total delivery cost of the current node;

[0221] Step S93: Perform global path planning based on the total delivery cost to generate an initial delivery path;

[0222] Step S94: Smooth the initial delivery route to obtain the target delivery route, and perform route planning based on the target delivery route.

[0223] It should be noted that intelligent vehicles can utilize improved [technology / methods] on topology maps. The algorithm calculates the globally optimal path from the starting point to the destination. The algorithm is a heuristic search algorithm commonly used in path planning. To enhance the algorithm's adaptability, this embodiment introduces a dynamic weight adjustment factor. The algorithm adjusts its approach based on the current stage of the path. Near the starting point, the search should be accelerated; near the target point, the overestimation of path cost should be reduced. The core of this algorithm lies in its evaluation function. It consists of two parts: actual cost and heuristic cost.

[0224] The formula for calculating the dynamic weight adjustment factor is as follows:

[0225]

[0226] in, Indicates the dynamic weight adjustment factor. and These are preset fixed parameter values, which depend on factors such as environmental complexity, the dynamic characteristics of the intelligent vehicle, and path smoothness, and need to be determined through actual application results. Indicates the starting point and the current node The distance between them Indicates the current node Distance to the target node.

[0227] Evaluation function The calculation formula is as follows:

[0228]

[0229] in, Indicates the current node n Total delivery cost This represents the actual cost from the starting point to the current node. Indicates the current node n Heuristic cost to reach the target point.

[0230] In some embodiments, the intelligent vehicle may introduce a B-spline curve based on curvature constraints to smooth the path. The B-spline path is defined by the following formula:

[0231]

[0232] Curvature calculation is based on the following formula:

[0233]

[0234] in, Represents path control points. Describes the B-spline basis functions. Indicates curvature, if Exceeding the threshold Then, the path control points will be automatically adjusted to limit the steering angle of the intelligent vehicle and improve driving stability.

[0235] In some embodiments, the intelligent vehicle can be designed using local planning algorithms. A virtual potential field is a guidance method based on attractive and repulsive forces, including a virtual attraction field and a virtual repulsion field, further combined with a virtual navigation field. The virtual attraction field is used to guide the intelligent vehicle along a global path. This embodiment introduces dynamic attractive forces, which are adjusted according to the accuracy of path tracking. Assuming the target position is , the formulas for calculating the attractive potential energy and the gravitational force are as follows:

[0236]

[0237]

[0238]

[0239] in, As the baseline attraction constant, To adjust the parameters, For the maximum target distance, This represents the distance from the current point to the target.

[0240] A virtual repulsion field is used to avoid obstacles. The repulsion force constant is dynamically adjusted based on the occupancy status of each grid cell in the grid map and the distance between the vehicle and the obstacle. The repulsion force increases as the vehicle approaches an obstacle. The formula for calculating the obstacle's repulsion force is:

[0241]

[0242]

[0243]

[0244] in, refers to obstacles. The distance from the current point to the obstacle. As the reference repulsive force constant, To affect the distance threshold, As a regulating factor, To control the parameters of the repulsive force deceleration rate, The local grid is the rectangular area within the perception range of the intelligent vehicle, and the local grid is the influencing factor of the raster map. The weight of each grid point, This indicates the degree of influence of the raster map influence factor on the repulsive potential energy. This indicates the degree of influence of the raster map influence factor on the repulsion force.

[0245] A virtual navigation field is used to provide navigation guidance based on traffic rules and road data. This field can influence route selection through external inputs, such as traffic flow data. Simultaneously, if a smart vehicle detects traffic congestion on a certain road segment, it transmits this message to other smart vehicles via V2X communication. Upon receiving the notification, other vehicles can recalculate their routes to avoid the congested area.

[0246]

[0247] in, For position Road i The guiding force above, Indicates the vehicle's location. Indicates position The navigational power at a given location is the sum of the navigational power of all roads. Indicates the number of roads.

[0248] The final total virtual force is the sum of attractive force, repulsive force, and guiding force:

[0249]

[0250] in, For position The total virtual force at a given location is the sum of attractive force, repulsive force, and guiding force. Indicates position The attraction of the place Indicates the number of repulsive force sources. Indicates position First The repulsive force of a repulsive force source.

[0251] By simulating a virtual spring-damped-inertial system, this invention proposes a virtual impedance model that enables intelligent vehicles to smoothly follow the global path and make dynamic obstacle avoidance adjustments in complex environments.

[0252] In virtual impedance control, the motion of the intelligent vehicle is constrained by the following dynamic equations:

[0253]

[0254] in, This is the current actual location of the intelligent vehicle. and For the current speed and acceleration of the intelligent vehicle, The desired trajectory is typically provided by a globally planned path. and Given the velocity and acceleration of the desired trajectory, For the calculated virtual total force, the virtual inertia matrix Virtual damping matrix Virtual elasticity matrix All are diagonal matrices, with the top left corner representing the value along the x-axis and the bottom right corner representing the value along the y-axis.

[0255] By transforming the constraint formulas of the dynamic equations, the formula for calculating the actual acceleration of the intelligent vehicle is as follows:

[0256]

[0257] The intelligent vehicle's speed and final trajectory are updated using points.

[0258]

[0259]

[0260] In some embodiments, each smart vehicle is equipped with multiple sensors (such as temperature sensors, oil pressure sensors, battery power sensors, speed sensors, etc.) to collect various operational status data of the vehicle in real time, including the vehicle's real-time speed. Battery power Hydraulic Engine temperature And so on. This data is analyzed in real time, and anomalies are detected using a threshold method. For each sensor parameter... Let its normal range be... .

[0261]

[0262] Perform the above detection on each sensor data point and define a fault detection signal. The calculation formula is:

[0263]

[0264] Where n is the number of sensors. This indicates whether each sensor detected an anomaly. If... If a certain threshold k is reached, the vehicle is considered to have malfunctioned, triggering the fault response mechanism. After the fault response mechanism is triggered, on the one hand, the original vehicle stops operating due to the malfunction, and the original workload needs to be transferred to other vehicles, i.e., task transfer; on the other hand, an idle vehicle needs to be selected to assist the malfunctioning vehicle, assuming there are hooks between the intelligent vehicles, to help tow it back to the base or repair point.

[0265] For task transfer, if there are currently vehicles with remaining load, assuming the maximum load of the remaining vehicles is... The original load of the faulty vehicle was L. If the remaining load exceeds L, then all vehicles are considered as candidate vehicles for task transfer. A priority score is calculated using the priority scoring formula, and the vehicle with the highest score is ultimately selected as the task transfer vehicle. If Priority scores are calculated based on the priority scoring formula. The vehicles are then sorted in descending order of score and allocated the remaining maximum load of supplies to each vehicle until all supplies from the faulty vehicle have been transferred. The priority scoring formula is as follows:

[0266]

[0267] in, It is a vehicle With the disabled vehicle The distance between them It is a vehicle The remaining load capacity, It is a vehicle The current load, It is a vehicle Available time, All are weighting coefficients.

[0268] If there are no remaining loaded vehicles, compare the priority of the disabled vehicle with the task sets of other vehicles. If the priority of the disabled vehicle is lower than that of other vehicles, wait. If the priority of the disabled vehicle exceeds that of some vehicles, take over the vehicle with the lowest priority among the vehicles that exceeded the priority. That vehicle immediately returns to the nearest supply point to unload the supplies and then goes to the disabled vehicle to transfer the task.

[0269] For assisted towing, calculate the priority score between other vehicles and the disabled vehicle, and select the vehicle with the highest score as the assisted towing vehicle, referring to the following formula:

[0270]

[0271] in, It is a vehicle energy status, All are weighting coefficients.

[0272] This embodiment responds to task information sent by the task management device, determines the task matching degree between itself and each emergency delivery task based on the task information, sends the task matching degree to the task management device, and monitors delivery messages. The delivery messages are sent by the task management device and include emergency task allocation strategies. Upon receiving the delivery message, emergency supplies are delivered based on the emergency task allocation strategy in the delivery message, thereby ensuring that the intelligent vehicle reasonably delivers material transportation tasks that match itself, improving the collaborative delivery efficiency of the intelligent vehicle, ensuring that emergency supplies can be quickly transported to the target location, and effectively improving the response capability of emergency material transportation.

[0273] In addition, to achieve the above objectives, this application also proposes a task management device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the multi-intelligent vehicle collaborative material delivery method described above.

[0274] In addition, to achieve the above objectives, this application also proposes an intelligent vehicle, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the multi-intelligent-vehicle collaborative material delivery method described above.

[0275] Furthermore, embodiments of the present invention also propose a computer-readable storage medium storing a multi-vehicle material collaborative delivery program, wherein when the multi-vehicle material collaborative delivery program is executed by a processor, it implements the steps of the multi-vehicle material collaborative delivery method described above.

[0276] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0277] The aforementioned computer-readable storage medium may be included in the multi-intelligent vehicle material collaborative delivery equipment; or it may exist independently and not be installed in the multi-intelligent vehicle material collaborative delivery equipment.

[0278] Furthermore, this invention also proposes a computer program product, including a multi-intelligent vehicle material collaborative delivery program, which, when executed by a processor, implements the steps of the multi-intelligent vehicle material collaborative delivery method described above.

[0279] The specific implementation of the computer program product of the present invention is basically the same as the embodiments of the above-mentioned multi-intelligent vehicle material collaborative delivery method, and will not be repeated here.

[0280] Reference Figure 5 , Figure 5 This is a structural block diagram of an embodiment of the multi-intelligent vehicle material collaborative delivery system of the present invention.

[0281] like Figure 5As shown, the multi-intelligent vehicle material collaborative delivery system proposed in this embodiment of the invention includes a task management device 10, a material management device 20, and multiple intelligent vehicles 30. The task management device 10 is communicatively connected to the material management device 20 and the multiple intelligent vehicles 30.

[0282] The task management device 10 is used to respond to an emergency task delivery request, obtain a set of delivery tasks and task information based on the emergency task delivery request, and send the task information to the material management device and each intelligent vehicle. The set of delivery tasks includes at least one emergency delivery task.

[0283] The material management device 20 is used to configure materials based on the task information, generate material configuration information, and send the material configuration information to the task management device.

[0284] The intelligent vehicle 30 is used to respond to the task information sent by the task management device and determine the task matching degree between itself and each emergency delivery task based on the task information.

[0285] The intelligent vehicle 30 is also used to send the task matching degree to the task management device and monitor delivery messages, which are sent by the task management device and include emergency task allocation strategies.

[0286] The task management device 10 is also used to construct a task matching graph based on material configuration information, task information, and task matching degree. The material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between the task nodes and the vehicle nodes, edge weights between the material nodes and the vehicle nodes, and node states of the task nodes and the vehicle nodes.

[0287] The task management device 10 is also used to perform a breadth-first search on the task matching graph, obtain the task nodes to be delivered and available vehicle nodes in the task matching graph based on the node status, and obtain the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node.

[0288] The task management device 10 is also used to split the task node to be delivered based on the task requirement load capacity and the remaining load capacity information to obtain multiple sub-task nodes.

[0289] The task management device 10 is also used to construct a task matching matrix based on the sub-task nodes and the available vehicle nodes;

[0290] The task management device 10 is also used to determine an emergency task allocation strategy based on the task matching matrix and send the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

[0291] The intelligent vehicle 30 is also used to deliver emergency supplies based on the emergency task allocation strategy in the delivery message when it receives the delivery message.

[0292] In this embodiment, a task management device responds to an emergency delivery request, obtains a set of delivery tasks and task information based on the request, and sends the task information to a materials management device and each intelligent vehicle. The delivery task set includes at least one emergency delivery task. The materials management device configures materials based on the task information, generates materials configuration information, and sends this information to the task management device. Each intelligent vehicle responds to the task information sent by the task management device, determines its own task matching degree with each emergency delivery task based on the task information, sends the matching degree to the task management device, and monitors delivery messages. These delivery messages, sent by the task management device, include an emergency task allocation strategy. The task management device constructs a task matching graph based on the materials configuration information, task information, and task matching degree. The materials configuration information is generated by the materials management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, materials nodes, edge weights between task nodes and vehicle nodes, and edge weights between materials nodes and vehicles. The edge weights between nodes and the node states of the task nodes and vehicle nodes are used to perform a breadth-first search on the task matching graph. Based on the node states, the task nodes to be delivered and available vehicle nodes in the task matching graph are obtained. The task load requirement of each task node to be delivered and the remaining load information of each available vehicle node are also obtained. Based on the task load requirement and the remaining load information, the task nodes to be delivered are split into multiple sub-task nodes. A task matching matrix is ​​constructed based on the sub-task nodes and the available vehicle nodes. An emergency task allocation strategy is determined based on the task matching matrix and sent to each intelligent vehicle so that each intelligent vehicle can carry out emergency material delivery based on the emergency task allocation strategy. When an intelligent vehicle receives a delivery message, it carries out emergency material delivery based on the emergency task allocation strategy in the delivery message. Since this embodiment splits the delivery task into multiple sub-tasks, constructs a matching matrix based on the sub-tasks, and allocates tasks based on the matching matrix, it achieves fine-grained allocation of delivery tasks, ensures the rationality of task allocation, and performs path planning based on the emergency task allocation strategy, thereby improving the collaborative delivery capability of intelligent vehicles and ensuring the timeliness of emergency material delivery.

[0293] The multi-vehicle collaborative material delivery system provided in this application, employing the multi-vehicle collaborative material delivery method described in the above embodiments, can solve the technical problems of multi-vehicle collaborative material delivery. Compared with the prior art, the beneficial effects of the multi-vehicle collaborative material delivery system provided in this application are the same as those of the multi-vehicle collaborative material delivery method described in the above embodiments, and other technical features of the multi-vehicle collaborative material delivery system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0294] It should be understood that the above are merely illustrative examples and do not constitute any limitation on the technical solutions of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any restrictions on this.

[0295] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0296] In addition, for technical details not described in detail in this embodiment, please refer to the multi-intelligent vehicle material collaborative delivery method provided in any embodiment of the present invention, which will not be repeated here.

[0297] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0298] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0299] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0300] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for collaborative delivery of goods using multiple intelligent vehicles, characterized in that, The multi-intelligent vehicle collaborative material delivery method is applied to a task management device, wherein the task management device communicates with the material management device and multiple intelligent vehicles, and the multi-intelligent vehicle collaborative material delivery method includes: In response to an emergency delivery request, a set of delivery tasks and task information are obtained based on the emergency delivery request, and the task information is sent to the material management equipment and each intelligent vehicle. The set of delivery tasks includes at least one emergency delivery task. A task matching graph is constructed based on material configuration information, task information, and task matching degree. The material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between the task nodes and the vehicle nodes, edge weights between the material nodes and the vehicle nodes, and node states of the task nodes and the vehicle nodes. A breadth-first search is performed on the task matching graph. Based on the node status, the task nodes to be delivered and the available vehicle nodes in the task matching graph are obtained, and the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node are obtained. Based on the required load capacity and the remaining load capacity information, the task node to be delivered is split into multiple sub-task nodes; Construct a task matching matrix based on the sub-task nodes and the available vehicle nodes; An emergency task allocation strategy is determined based on the task matching matrix and sent to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy. The construction of the task matching matrix based on the sub-task nodes and the available vehicle nodes includes: Based on the subtask node, a matching degree update request is sent to each available vehicle node, and the updated task matching degree of each available vehicle node based on the subtask node is obtained. Construct an initial matching matrix based on the updated task matching degree: in, Represents the initial matching matrix. Representative task With vehicles The degree of task matching between them; The initial matching matrix is ​​reduced to obtain the target matrix; The target matrix is ​​adjusted based on a dynamic zero-element update strategy to obtain a task matching matrix. The dynamic zero-element update strategy includes: in, This is the adjusted task matching matrix. For the target matrix, To adjust the coefficient, Used to control the adjustment range The minimum non-zero matching degree value in the target matrix.

2. The multi-intelligent vehicle collaborative material delivery method as described in claim 1, characterized in that, The step of reducing the initial matching matrix to obtain the target matrix includes: Calculate adaptive weights based on the adaptive weight adjustment function: in, Representative vehicle With the task The degree of matching between them Represents the first element in the initial matching matrix. i The maximum matching degree in the row, Represents a constant. Represents adaptive weights; The initial matching matrix is ​​reduced according to the adaptive weights to obtain a candidate matrix. The reduction is performed using the following formula: in, Represents the matching degree after reduction and adjustment. This represents the matching degree before adjustment in the initial matching matrix; The candidate matrix is ​​adjusted based on the time window constraint to obtain the target matrix: in, This represents the degree of matching after adjusting the time window constraints. Representative vehicle Execute the task The estimated completion time, For the task The latest deadline stipulated The time window penalty coefficient, Used to control the severity of timeout penalties.

3. The multi-intelligent vehicle collaborative material delivery method as described in any one of claims 1 or 2, characterized in that, The step of determining an emergency task allocation strategy based on the task matching matrix and sending the emergency task allocation strategy to each intelligent vehicle, so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy, includes: A task allocation model is constructed based on the task matching matrix, the task allocation model including: in, Representative vehicle With the task The allocation status between them Representative vehicle With the task The degree of matching between them Represents the set of task nodes to be delivered. Represents the set of vehicle nodes. This represents the vehicle's maximum load capacity. This represents the required payload capacity for the task. An emergency task allocation strategy is determined based on the task allocation model, and the emergency task allocation strategy is sent to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy.

4. A method for collaborative delivery of goods using multiple intelligent vehicles, characterized in that, The multi-intelligent vehicle collaborative material delivery method is applied to intelligent vehicles in an intelligent vehicle cluster, wherein the intelligent vehicle cluster includes multiple intelligent vehicles, and the multiple intelligent vehicles are communicatively connected with task management equipment and material management equipment. The method includes: In response to the task information sent by the task management device, the system determines the task matching degree between itself and each emergency delivery task based on the task information. The task matching degree is sent to the task management device, and delivery messages are monitored. The delivery messages are sent by the task management device and include emergency task allocation strategies. Upon receiving the delivery message, emergency supplies are delivered based on the emergency task allocation strategy in the delivery message. The process of determining the task matching degree between itself and each emergency delivery task based on the task information includes: Based on the task information, determine the location of the material point, the location of the task point, and the required load capacity of the task. The vehicle's task matching degree with each emergency delivery task is determined based on its maximum load capacity, the location of the material point, the location of the task point, and the required load capacity of the task. in, Indicates vehicle Distance to supply points Indicates the distance between the supply point and the mission point. This represents the vehicle's maximum load capacity. This represents the required payload capacity for the mission. For adjustment coefficients, These are used to balance priority, operating cost, and load matching, respectively. As a task priority, Represents the vehicle's location. Represents the location of the supply point. This represents the location of the task point.

5. The multi-intelligent vehicle collaborative material delivery method as described in claim 4, characterized in that, After sending the task matching score to the task management device, the method further includes: Monitor matching score update requests; Upon receiving a matching score update request, the task matching score is updated based on the sub-task nodes in the matching score update request to obtain the updated task matching score: in, Indicates vehicle sub-tasks The degree of matching between them For subtasks The task requires a certain load capacity; The updated task matching score is sent to the task management device.

6. The multi-intelligent vehicle collaborative material delivery method as described in any one of claims 4 or 5, characterized in that, After distributing emergency supplies based on the emergency task allocation strategy in the delivery message, the process also includes: A raster map is constructed by raster mapping based on point cloud data collected by sensors during the emergency supplies distribution process; Determine the occupancy probability of each grid cell in the grid map: in, Represents grid The likelihood of being occupied. Represents grid Prior probability of being occupied, Represents the normalization factor. Represents grid The probability of occupancy, This indicates the current sensor measurement result. This indicates historical sensor measurement results; Based on the occupancy probability, a passage analysis is performed on the grid map to determine the currently passable area in the grid map; Based on the currently passable area, topological nodes, topological edges between adjacent topological nodes, and edge weights of the topological edges are determined. The edge weights are calculated based on the following formula: in, Represents edge weight, Indicates adjacent nodes and The Euclidean distance between them For the driving speed of intelligent vehicles, Traffic correction factor The obstacle influence factor, Represents a node and The accident rate of the road sections between them within the same time period Indicates the number of accidents. Indicates the number of times a vehicle passes through. and Indicates the adjustment factor; A topology map is constructed based on the topology nodes, the topology edges, and the edge weights; Based on the topology map and the emergency task allocation strategy, path planning is performed, and emergency supplies are distributed based on the path planning results.

7. The multi-intelligent vehicle collaborative material delivery method as described in claim 6, characterized in that, The route planning based on the topology map and the emergency task allocation strategy includes: The dynamic weight adjustment factor for the current node is determined based on the topology map and the emergency task allocation strategy: in, Indicates the dynamic weight adjustment factor. and These are preset fixed parameter values. Indicates the starting point and the current node The distance between them Indicates the current node Distance to the target node; Cost evaluation is performed based on a dynamic weight adjustment factor to determine the total delivery cost for the current node: in, Indicates the current node n Total delivery cost This represents the actual cost from the starting point to the current node. Indicates the current node n Heuristic cost to reach the target point; Based on the total delivery cost, perform global path planning to generate an initial delivery path; The initial delivery route is smoothed to obtain the target delivery route, and route planning is performed based on the target delivery route.

8. A multi-intelligent vehicle collaborative material delivery system, characterized in that, The multi-intelligent vehicle material collaborative delivery system includes a task management device, a material management device, and multiple intelligent vehicles. The task management device is communicatively connected to the material management device and the multiple intelligent vehicles. The task management device is used to respond to an emergency task delivery request, obtain a set of delivery tasks and task information based on the emergency task delivery request, and send the task information to the material management device and each intelligent vehicle. The set of delivery tasks includes at least one emergency delivery task. The material management device is used to configure materials based on the task information, generate material configuration information, and send the material configuration information to the task management device. The intelligent vehicle is used to respond to the task information sent by the task management device and determine the task matching degree between itself and each emergency delivery task based on the task information; The intelligent vehicle is also used to send the task matching degree to the task management device and monitor delivery messages, which are sent by the task management device and include emergency task allocation strategies. The task management device is also used to construct a task matching graph based on material configuration information, task information, and task matching degree. The material configuration information is generated by the material management device after configuring materials based on the task information. The task matching degree is the degree of matching between each intelligent vehicle and each emergency delivery task. The task matching graph includes task nodes, vehicle nodes, material nodes, edge weights between the task nodes and the vehicle nodes, edge weights between the material nodes and the vehicle nodes, and node states of the task nodes and the vehicle nodes. The task management device is also used to perform a breadth-first search on the task matching graph, obtain the task nodes to be delivered and available vehicle nodes in the task matching graph based on the node status, and obtain the task load requirement of each task node to be delivered and the remaining load information of each available vehicle node. The task management device is also used to split the task node to be delivered based on the task requirement load capacity and the remaining load capacity information to obtain multiple sub-task nodes; The task management device is also used to construct a task matching matrix based on the sub-task nodes and the available vehicle nodes; The task management device is also used to determine an emergency task allocation strategy based on the task matching matrix and send the emergency task allocation strategy to each intelligent vehicle so that each intelligent vehicle can carry out emergency material distribution based on the emergency task allocation strategy. The intelligent vehicle is also used to deliver emergency supplies based on the emergency task allocation strategy in the delivery message when it receives the delivery message. The task management device is also used to send a matching degree update request to each available vehicle node based on the sub-task node, and to obtain the updated task matching degree sent by each available vehicle node based on the sub-task node. Construct an initial matching matrix based on the updated task matching degree: in, Represents the initial matching matrix. Representative task With vehicles The degree of task matching between them; The initial matching matrix is ​​reduced to obtain the target matrix; the target matrix is ​​then adjusted based on a dynamic zero-element update strategy to obtain the task matching matrix, wherein the dynamic zero-element update strategy includes: in, This is the adjusted task matching matrix. For the target matrix, To adjust the coefficient, Used to control the adjustment range The minimum non-zero matching degree value in the target matrix; The intelligent vehicle is also used to determine the location of the material point, the location of the task point, and the required load capacity based on the task information; and to determine the task matching degree between itself and each emergency delivery task based on the vehicle's maximum load capacity, the location of the material point, the location of the task point, and the required load capacity. in, Indicates vehicle Distance to supply points Indicates the distance between the supply point and the mission point. This represents the vehicle's maximum load capacity. This represents the required payload capacity for the mission. For adjustment coefficients, These are used to balance priority, operating cost, and load matching, respectively. As a task priority, Represents the vehicle's location. Represents the location of the supply point. This represents the location of the task point.

Images