Intelligent unmanned vehicle for campus distribution

Abstract

The invention relates to the technical field of intelligent control, and particularly discloses a campus distribution intelligent unmanned vehicle, and the method comprises the steps: obtaining the spatial layout data and real-time traffic data of a campus, and obtaining the distribution task data of all intelligent unmanned vehicles; determining a distribution task graph and a distribution scheme set; evaluating the distribution scheme set based on the spatial layout data, the distribution scheme set and the real-time traffic data, and determining a distribution evaluation set of the distribution scheme set; determining an optimal delivery scheme of all the intelligent unmanned vehicles, and obtaining delivery states of all the intelligent unmanned vehicles in real time to determine real-time delivery data; and optimizing the optimal distribution scheme of the intelligent unmanned vehicle. According to the method, the distribution efficiency and accuracy can be improved, the method adapts to a dynamically changing complex campus environment, the intelligence and real-time response capability of intelligent unmanned vehicle distribution are improved, the distribution scheme is optimized in time, and the high efficiency and flexibility of unmanned vehicle distribution during task execution are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and particularly to an intelligent unmanned vehicle for campus distribution. Background Art

[0002] With the rapid development of artificial intelligence technology, intelligent unmanned distribution vehicles have gradually entered the fields of campus, commercial areas and urban distribution. Traditional campus distribution usually relies on manual distribution or traditional distribution vehicles, which is not only inefficient, but also often affected by factors such as traffic and congestion. In recent years, using intelligent unmanned vehicles for in-campus item distribution has become an innovative solution, but it still faces challenges such as path planning, traffic conditions, and complexity of spatial layout.

[0003] In this regard, the Particle Swarm Optimization (PSO) algorithm, as an optimization technology, has been widely applied to problems such as path planning and resource scheduling, and can perform efficient search and optimization in complex environments. In addition, with the development of Internet of Things (IoT) technology and big data, the acquisition of real-time traffic data and spatial layout data has become more efficient, providing more real-time decision-making basis for intelligent distribution. However, how to comprehensively consider real-time traffic data, spatial layout, and the distribution tasks of multiple intelligent unmanned vehicles is still a challenging research direction.

[0004] Therefore, the present invention proposes an intelligent unmanned vehicle for campus distribution. Summary of the Invention

[0005] The present invention provides an intelligent unmanned vehicle for campus distribution. By analyzing the acquired spatial layout data and distribution task data, a distribution task graph is determined. According to the distribution task graph and the Particle Swarm Optimization algorithm, a set of distribution plans is determined. The set of distribution plans is evaluated based on the spatial layout data, the set of distribution plans, and the real-time traffic data to determine a distribution evaluation set of the set of distribution plans. According to the distribution evaluation set, the optimal distribution plan for all intelligent unmanned vehicles is determined, and the optimal distribution plan of the intelligent unmanned vehicle is optimized by combining the real-time distribution data obtained in real time. It can improve the distribution efficiency and accuracy, adapt to the dynamically changing complex campus environment, enhance the intelligence and real-time response ability of intelligent unmanned vehicle distribution, optimize the distribution plan in a timely manner, and ensure the efficiency and flexibility of unmanned vehicle distribution when performing tasks.

[0006] The present invention provides an intelligent unmanned vehicle for campus distribution, including:

[0007] S1: Acquire the spatial layout data and real-time traffic data of the campus, and acquire the distribution task data of all intelligent unmanned vehicles;

[0008] S2: Analyze the spatial layout data of the campus and the delivery task data of all intelligent unmanned vehicles, determine the delivery task graph, and determine the set of delivery plans based on the delivery task graph and the particle swarm optimization algorithm;

[0009] S3: Evaluate the set of delivery plans based on the spatial layout data, the set of delivery plans, and the real-time traffic data, and determine the delivery evaluation set of the set of delivery plans;

[0010] S4: Determine the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, and obtain the real-time delivery data by acquiring the delivery status of all intelligent unmanned vehicles in real time;

[0011] S5: Optimize the optimal delivery plan of the intelligent unmanned vehicle based on the real-time delivery data and the set of delivery plans.

[0012] Preferably, based on an intelligent unmanned vehicle for campus delivery, the spatial layout data includes campus road data, campus building data, and campus function data. Among them, the campus building data includes at least the building names and building locations of all buildings on the campus;

[0013] The delivery task data of all intelligent unmanned vehicles includes at least the shipping addresses and receiving addresses of all delivery tasks;

[0014] The real-time traffic data includes at least road traffic flow data and road vehicle speed data.

[0015] Preferably, based on an intelligent unmanned vehicle for campus delivery, analyzing the spatial layout data of the campus and the delivery task data of all intelligent unmanned vehicles to determine the delivery task graph includes:

[0016] Preprocess the acquired spatial layout data and standardize the acquired delivery task data of all intelligent unmanned vehicles;

[0017] Based on the building names and building locations of all buildings in the campus building data in the spatial layout data, determine the node set of the delivery task data;

[0018] Based on the campus road data in the spatial layout data, determine the connection paths and the path lengths of the connection paths between every two nodes in the node set, and determine the edge set of the delivery task data based on the connection paths and the path lengths between all nodes in the node set;

[0019] Based on the node set and the edge set of the delivery task data, determine the delivery task graph of the delivery task data.

[0020] Preferably, based on an intelligent unmanned vehicle for campus delivery, determining the set of delivery plans based on the delivery task graph and the particle swarm optimization algorithm includes:

[0021] Based on the shipping addresses and receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, determine the shipping nodes, receiving nodes, and multiple feasible delivery routes for each delivery task in the delivery task data;

[0022] Determine that all delivery tasks with the same shipping node in the delivery task data are a task cluster, and allocate one or more intelligent unmanned vehicles to each task cluster based on the number of delivery tasks in each task cluster;

[0023] Based on the receiving nodes of all delivery tasks in each task cluster, determine the vehicle delivery set of each intelligent unmanned vehicle in each task cluster, where the vehicle delivery set of the intelligent unmanned vehicle includes multiple delivery tasks with the same shipping node;

[0024] Based on the delivery task graph, the vehicle delivery set of each intelligent unmanned vehicle, and the multiple feasible delivery routes of each delivery task in the vehicle delivery set, randomly generate the initial particle swarm of each intelligent unmanned vehicle, and determine the subset of delivery plans for each intelligent unmanned vehicle based on the initial particle swarm;

[0025] Determine the delivery plan set based on the subset of delivery plans of all intelligent unmanned vehicles.

[0026] Preferably, based on a campus delivery intelligent unmanned vehicle, based on the shipping addresses and receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, determine the shipping nodes, receiving nodes, and multiple feasible delivery routes for each delivery task in the delivery task data, including:

[0027] For the shipping addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, split them based on the building names of all buildings in the campus building data in the spatial layout data, and determine the characters before and after the building name of any building in the shipping address of each delivery task in the delivery task data as the first delimiter character for each delivery task;

[0028] For the receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, split them based on the building names of all buildings in the campus building data in the spatial layout data, and determine the characters before and after the building name of any building in the receiving address of each delivery task in the delivery task data as the second delimiter character for each delivery task;

[0029] Based on the first delimiter character and the second delimiter character of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, determine the delimiter character set of the delivery task data;

[0030] Truncate the shipping address of each delivery task in the delivery task data based on the segmentation character set to determine the shipping node of each delivery task in the delivery task data. At the same time, truncate the receiving address of each delivery task in the delivery task data based on the segmentation character set to determine the receiving node of each delivery task in the delivery task data;

[0031] Match the shipping nodes and receiving nodes of each delivery task in the delivery task data with the building names of all buildings in the campus building data in the spatial layout data, and manually identify the shipping addresses and receiving addresses of each delivery task that do not fully match, and update the shipping nodes and receiving nodes of each delivery task that do not fully match;

[0032] Based on the shipping node, receiving node, and delivery task graph of each delivery task in the delivery task data, determine multiple feasible delivery paths for each delivery task in the delivery task data.

[0033] Preferably, based on a campus delivery intelligent unmanned vehicle, evaluate the delivery plan set based on the spatial layout data, delivery plan set, and real-time traffic data to determine the delivery evaluation set of the delivery path set, including:

[0034] Determine the restricted path set in the delivery task graph based on the campus function data in the spatial layout data;

[0035] Based on the restricted path set, real-time traffic data, and the delivery plan subset of each intelligent unmanned vehicle in the delivery plan set, determine the delivery plan evaluation value of the delivery plan corresponding to each initialization example in the delivery plan subset of each intelligent unmanned vehicle in the delivery plan set;

[0036]

[0037]

[0038] Among them, F ak represents the delivery plan evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, FR ak represents the delivery path evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, FL ak represents the delivery penalty evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, α1 represents the delivery path evaluation weight, α2 represents the delivery penalty evaluation weight, d j represents the path length of the connection path corresponding to the jth edge in the delivery task graph, Denote the average vehicle speed on the connection path corresponding to the \(t\)-th passing through the \(j\)-th edge in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle, based on the road vehicle speed data. Denote the flow impact factor based on the road traffic volume when the \(t\)-th passing through the connection path corresponding to the \(j\)-th edge in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(a_{kj}^N\) denotes the number of times the connection path corresponding to the \(j\)-th edge in the delivery task graph is passed through in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(a_{k}^N\) denotes the number of connection paths corresponding to the edges in the delivery task graph passed through in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(R\) j Denote the connection path corresponding to the \(j\)-th edge in the delivery task graph. \(LC\) denotes the set of restricted paths. \(u\) j Denote the path usage factor based on the campus function data of the connection path corresponding to the \(j\)-th edge in the delivery task graph. \(1(R\) j \(\in LC\) means taking the value of 1 when the connection path corresponding to the \(j\)-th edge in the delivery task graph belongs to the set of restricted paths. Denote the vehicle load when the \(t\)-th passing through the connection path corresponding to the \(j\)-th edge in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(\max(w\) ak ) denotes the maximum load of the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(\gamma_1\) denotes the load adjustment factor, and \(\gamma_2\) denotes the path repetition adjustment factor. Denote the minimum delivery plan evaluation value of all delivery plans in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. Denote the maximum delivery plan evaluation value of all delivery plans in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle;

[0039] Based on the delivery plan evaluation values of the delivery plans corresponding to all initialization examples in the subset of delivery plans of each intelligent unmanned vehicle in the delivery plan set, determine the delivery sub-evaluation set of each intelligent unmanned vehicle;

[0040] Based on the delivery sub-evaluation sets of all intelligent unmanned vehicles, determine the delivery evaluation set.

[0041] Preferably, based on a campus delivery intelligent unmanned vehicle, determine the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, including:

[0042] Select the delivery plan corresponding to the maximum delivery plan evaluation value in the delivery sub-evaluation set of each intelligent unmanned vehicle in the delivery evaluation set as the sub-optimal delivery plan of each intelligent unmanned vehicle;

[0043] Based on the sub-optimal delivery plans of all intelligent unmanned vehicles, determine the optimal delivery plan.

[0044] Preferably, based on an intelligent unmanned vehicle for campus delivery, real-time delivery data is obtained by acquiring the delivery status of all intelligent unmanned vehicles in real time, including:

[0045] Acquire the delivery status of each intelligent unmanned vehicle based on the corresponding optimal delivery plan in real time, and determine real-time sub-delivery data;

[0046] Determine the real-time delivery data based on the real-time sub-delivery data of all intelligent unmanned vehicles.

[0047] The beneficial effects of the present invention compared with the prior art are as follows: By analyzing the acquired spatial layout data and delivery task data, a delivery task graph is determined. According to the delivery task graph and the particle swarm algorithm, a set of delivery plans is determined. The set of delivery plans is evaluated based on the spatial layout data, the set of delivery plans, and the real-time traffic data to determine the delivery evaluation set of the set of delivery plans. The optimal delivery plan for all intelligent unmanned vehicles is determined according to the delivery evaluation set, and the optimal delivery plan of the intelligent unmanned vehicle is optimized by combining the real-time delivery data acquired in real time. It can improve the delivery efficiency and accuracy, adapt to the dynamic and complex campus environment, enhance the intelligence and real-time response ability of intelligent unmanned vehicle delivery, optimize the delivery plan in time, and ensure the efficiency and flexibility of unmanned vehicle delivery when performing tasks.

[0048] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in this application document.

[0049] The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments. Description of the Drawings

[0050] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0051] Figure 1 It is a flowchart of a method for an intelligent unmanned vehicle for campus delivery in an embodiment of the present invention. Detailed Embodiments

[0052] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0053] Embodiment 1:

[0054] The present invention provides an intelligent unmanned vehicle for campus delivery, refer toFigure 1 , including:

[0055] S1: Obtain the spatial layout data and real-time traffic data of the campus, and obtain the delivery task data of all intelligent unmanned vehicles;

[0056] S2: Analyze the spatial layout data of the campus and the delivery task data of all intelligent unmanned vehicles, determine the delivery task graph, and determine the set of delivery plans based on the delivery task graph and the particle swarm optimization algorithm;

[0057] S3: Evaluate the set of delivery plans based on the spatial layout data, the set of delivery plans, and the real-time traffic data, and determine the delivery evaluation set of the set of delivery plans;

[0058] S4: Determine the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, and obtain the real-time delivery data by obtaining the delivery status of all intelligent unmanned vehicles in real time;

[0059] S5: Optimize the optimal delivery plan of the intelligent unmanned vehicle based on the real-time delivery data and the set of delivery plans.

[0060] In this embodiment, by analyzing the campus spatial layout data and the delivery task data, a delivery task graph is constructed. The task graph includes various delivery nodes (buildings) on the campus and the paths between them. Then, the particle swarm optimization algorithm is used to optimize these delivery paths to generate a preliminary delivery plan. The particle swarm optimization algorithm simulates the process of multiple path selections.

[0061] In this embodiment, in combination with the spatial layout data, the delivery plan, and the real-time traffic data, the generated delivery plan is evaluated. The applicability of each delivery path is judged.

[0062] In this embodiment, based on the evaluation results of the delivery plan, the sub-optimal delivery plan for each intelligent unmanned vehicle is selected. Through the real-time monitoring system, the delivery status of each vehicle is obtained, including the vehicle location, the estimated arrival time, etc., to ensure that the progress of each delivery task is under control.

[0063] In this embodiment, based on the obtained real-time delivery data, the optimal delivery path is dynamically optimized. Ensure that the intelligent unmanned vehicle can complete the task quickly and efficiently.

[0064] The beneficial effects of the above technology are as follows: By analyzing the obtained spatial layout data and distribution task data, a distribution task map is determined. According to the distribution task map and the particle swarm algorithm, a set of distribution plans is determined. The set of distribution plans is evaluated based on the spatial layout data, the set of distribution plans, and the real-time traffic data to determine the distribution evaluation set of the set of distribution plans. The optimal distribution plan for all intelligent unmanned vehicles is determined according to the distribution evaluation set, and the optimal distribution plan of the intelligent unmanned vehicles is optimized by combining the real-time distribution data obtained in real time. It can improve the distribution efficiency and accuracy, adapt to the dynamic and complex campus environment, enhance the intelligence and real-time response ability of intelligent unmanned vehicle distribution, optimize the distribution plan in a timely manner, and ensure the high efficiency and flexibility of unmanned vehicle distribution when performing tasks.

[0065] Embodiment 2:

[0066] Based on Embodiment 1, for a smart unmanned vehicle for campus distribution, the spatial layout data includes campus road data, campus building data, and campus function data. Among them, the campus building data at least includes the building names and building locations of all buildings on campus;

[0067] The distribution task data of all intelligent unmanned vehicles at least includes the shipping address and the receiving address of all distribution tasks;

[0068] The real-time traffic data at least includes road traffic flow data and road vehicle speed data.

[0069] In this embodiment, the campus road data describes the road network on campus, including information such as main roads, paths, lanes, and sidewalks, and is used to plan the driving routes of unmanned delivery vehicles.

[0070] In this embodiment, the campus building data covers information such as the names and locations of all campus buildings. This helps to locate the shipping and receiving addresses of distribution tasks and ensures that unmanned vehicles can accurately reach specific buildings.

[0071] In this embodiment, the campus function data includes the functions of different areas on campus, such as teaching buildings, dormitory buildings, canteens, etc., which helps to optimize the distribution route and avoid unmanned vehicles passing through unnecessary areas.

[0072] In this embodiment, in addition to the shipping address and the receiving address, the distribution task data of all intelligent unmanned vehicles may also include the weight of the goods, the volume of the goods, and the latest delivery time, etc.

[0073] In this embodiment, the shipping address and the receiving address represent the shipping point and the receiving point of each distribution task in the distribution task data, which helps unmanned vehicles determine the distribution route.

[0074] In this embodiment, the road traffic flow data reflects the number of vehicles traveling on the road to evaluate whether the road is congested, which affects route selection.

[0075] In this embodiment, the road vehicle speed data represents the real-time driving speed on different roads to determine the estimated driving time for each route.

[0076] The beneficial effects of the above technologies are as follows: Determining the spatial layout data, distribution task data, and real-time traffic data can provide data support for determining the distribution task map, distribution plan set, and distribution evaluation set.

[0077] Embodiment 3:

[0078] Based on Embodiment 2, a smart unmanned vehicle for campus distribution analyzes the spatial layout data of the campus and the distribution task data of all smart unmanned vehicles to determine the distribution task map, including:

[0079] Preprocess the obtained spatial layout data and standardize the obtained distribution task data of all smart unmanned vehicles;

[0080] Based on the building names and locations of all buildings in the campus building data in the spatial layout data, determine the node set of the distribution task data;

[0081] Based on the campus road data in the spatial layout data, determine the connection paths between every two nodes in the node set and the path lengths of the connection paths, and determine the edge set of the distribution task data based on the connection paths between all nodes in the node set and the path lengths of the connection paths;

[0082] Based on the node set and edge set of the distribution task data, determine the distribution task map of the distribution task data.

[0083] In this embodiment, the preprocessing includes cleaning the spatial layout data of the campus to remove duplicate, incorrect, or incomplete information, such as correcting road coordinate deviations and supplementing missing attributes of buildings.

[0084] In this embodiment, the standardization of the distribution task data at least includes unifying the formats of the shipping address and the receiving address.

[0085] In this embodiment, each building is regarded as a node through the building name and location in the campus building data.

[0086] In this embodiment, according to the campus road data, determine the connection paths between every two nodes (buildings). The path length refers to the distance from one node to another node, and the actual length of the road and possible turns are considered during the calculation.

[0087] In this embodiment, the delivery task graph is composed of nodes and edges. The nodes represent each target (building) in the delivery task, and the edges represent the reachable paths between the nodes and their path lengths. By constructing the delivery task graph, effective path planning can be carried out for all delivery tasks.

[0088] The beneficial effects of the above technology are as follows: By analyzing the spatial layout data of the campus and the delivery task data of all intelligent unmanned vehicles, the delivery task graph can be determined, and the nodes and connection paths of all delivery tasks in the delivery task data can be accurately defined to adapt to the dynamically changing campus environment and improve the intelligence and real-time response ability of intelligent unmanned vehicle delivery.

[0089] Embodiment 4:

[0090] Based on the delivery task graph and the particle swarm optimization algorithm on the basis of Embodiment 2, a set of delivery plans is determined, including:

[0091] Based on the shipping addresses and receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, the shipping nodes, receiving nodes, and multiple feasible delivery paths of each delivery task in the delivery task data are determined;

[0092] All delivery tasks with the same shipping node in the delivery task data are determined as a task cluster, and one or more intelligent unmanned vehicles are allocated to each task cluster based on the number of delivery tasks in each task cluster;

[0093] Based on the receiving nodes of all delivery tasks in each task cluster, the vehicle delivery set of each intelligent unmanned vehicle in each task cluster is determined. Among them, the vehicle delivery set of an intelligent unmanned vehicle includes multiple delivery tasks with the same shipping node;

[0094] Based on the delivery task graph, the vehicle delivery set of each intelligent unmanned vehicle, and the multiple feasible delivery paths of each delivery task in the vehicle delivery set, the initial particle swarm of each intelligent unmanned vehicle is randomly generated, and the delivery plan subset of each intelligent unmanned vehicle is determined based on the initial particle swarm;

[0095] Based on the delivery plan subsets of all intelligent unmanned vehicles, a set of delivery plans is determined.

[0096] In this embodiment, by analyzing the shipping address and receiving address in the delivery task data, the shipping node and the receiving node of the termination node of each delivery task are determined. At the same time, based on the campus road data, multiple feasible delivery paths for each task are found.

[0097] In this embodiment, task aggregation refers to aggregating all distribution tasks with the same shipping node together and regarding them as an independent distribution point. By analyzing the number of distribution tasks in each task aggregation, the corresponding number of intelligent unmanned vehicles is allocated to ensure that each aggregation can be processed in a timely manner.

[0098] In this embodiment, the distribution tasks within each task aggregation will be optimized and arranged according to their receiving nodes to ensure that the same intelligent unmanned vehicle can efficiently complete the distribution of multiple tasks. Multiple feasible paths provide flexible choices for each vehicle to optimize the distribution efficiency.

[0099] In this embodiment, for each intelligent unmanned vehicle, an initial particle swarm is generated through the particle swarm algorithm. The particle swarm algorithm simulates all feasible distribution paths of all distribution tasks to determine a subset of distribution plans for each intelligent unmanned vehicle. The subset of distribution plans includes multiple distribution plans corresponding to each intelligent unmanned vehicle;

[0100] Integrate the subsets of distribution plans for all intelligent unmanned vehicles to determine the set of distribution plans.

[0101] The beneficial effects of the above technology are as follows: Based on the distribution task graph and the particle swarm algorithm, the set of distribution plans is determined, which can optimize the distribution task arrangement of intelligent unmanned vehicles in a complex campus environment and optimize the distribution efficiency.

[0102] Embodiment 5:

[0103] On the basis of Embodiment 4, a campus distribution intelligent unmanned vehicle determines the shipping node, receiving node, and multiple feasible distribution paths of each distribution task in the distribution task data based on the shipping addresses and receiving addresses of all distribution tasks in the distribution task data of all intelligent unmanned vehicles, including:

[0104] For the shipping addresses of all distribution tasks in the distribution task data of all intelligent unmanned vehicles, split them based on the building names of all buildings in the campus building data in the spatial layout data, and determine the characters before and after the building name of any building in the shipping address of each distribution task in the distribution task data as the first delimiter character for each distribution task;

[0105] For the receiving addresses of all distribution tasks in the distribution task data of all intelligent unmanned vehicles, split them based on the building names of all buildings in the campus building data in the spatial layout data, and determine the characters before and after the building name of any building in the receiving address of each distribution task in the distribution task data as the second delimiter character for each distribution task;

[0106] Based on the first delimiter characters and the second delimiter characters of all distribution tasks in the distribution task data of all intelligent unmanned vehicles, determine the delimiter character set of the distribution task data;

[0107] Truncate the shipping address of each delivery task in the delivery task data based on the split character set to determine the shipping node of each delivery task in the delivery task data. At the same time, truncate the receiving address of each delivery task in the delivery task data based on the split character set to determine the receiving node of each delivery task in the delivery task data;

[0108] Match the shipping node and receiving node of each delivery task in the delivery task data with the building names of all buildings in the campus building data in the spatial layout data, and manually identify the shipping address and receiving address of each delivery task that do not fully match, and update the shipping node and receiving node of each delivery task that do not fully match;

[0109] Based on the shipping node, receiving node and delivery task graph of each delivery task in the delivery task data, determine multiple feasible delivery paths for each delivery task in the delivery task data.

[0110] In this embodiment, based on the building names in the campus building data, the shipping address of the delivery task is split into separator characters. The separator characters are the characters before and after the building name, and are used as the first separator of the task. This process helps to associate complex address information with specific building locations and simplifies path planning.

[0111] In this embodiment, for example: determine the characters before and after the building name of any building in the shipping address of each delivery task in the delivery task data as the first separator character of each delivery task. For example: the shipping address of a certain delivery task is Room 302, 3rd Floor, East Area Technology Building. Split the shipping address of this delivery task into East Area, Technology Building and Room 302, 3rd Floor, and determine the separator characters of this delivery task as Area and 3.

[0112] In this embodiment, similar to the splitting of the shipping address, based on the campus building name, split the receiving address of each delivery task to determine the second separator character. This step helps to mark the specific building location of the receiving task and facilitates the determination of the precise delivery route.

[0113] In this embodiment, perform statistical analysis on the first separator character and the second separator character of all delivery tasks in the delivery task data of all intelligent unmanned vehicles to determine the split character set of the delivery task data. For example, the split character set includes numbers, areas, -, spaces, etc.

[0114] In this embodiment, based on the split character set, truncate the shipping address and receiving address of each delivery task respectively. The address information obtained after truncation will clearly mark the shipping node and receiving node of the delivery task, and these nodes are the core of the delivery path planning.

[0115] In this embodiment, the shipping and receiving nodes of each delivery task are matched with the building names in the campus building data to ensure the accuracy of the task address. For task addresses that do not fully match, they are manually identified and their nodes are updated to ensure the accuracy of the data.

[0116] In this embodiment, based on the determined shipping node, receiving node, and delivery task graph, multiple feasible delivery paths for each delivery task are determined, providing multiple alternative solutions for path planning to ensure that the optimal path can be flexibly selected under different conditions.

[0117] The beneficial effects of the above technology are as follows: Based on the shipping addresses and receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, the shipping nodes, receiving nodes, and multiple feasible delivery paths of each delivery task in the delivery task data are determined, which can improve the address accuracy of the delivery tasks, provide high-quality data support for determining the delivery plan set, and improve the intelligence and efficiency of intelligent unmanned vehicle delivery.

[0118] Embodiment 6:

[0119] Based on the embodiment 4, a campus delivery intelligent unmanned vehicle evaluates the delivery plan set based on the space layout data, delivery plan set, and real-time traffic data to determine the delivery evaluation set of the delivery path set, including:

[0120] Determine the restricted path set in the delivery task graph based on the campus function data in the space layout data;

[0121] Based on the restricted path set, real-time traffic data, and the delivery plan subset of each intelligent unmanned vehicle in the delivery plan set, determine the delivery plan evaluation value of the delivery plan corresponding to each initialization example in the delivery plan subset of each intelligent unmanned vehicle in the delivery plan set;

[0122]

[0123] Among them, F ak represents the delivery plan evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, FR ak represents the delivery path evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, FL ak represents the delivery penalty evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, α1 represents the delivery path evaluation weight, α2 represents the delivery penalty evaluation weight, d j represents the path length of the connection path corresponding to the jth edge in the delivery task graph, Denote the average vehicle speed on the connection path corresponding to the \(t\)-th passing through the \(j\)-th edge in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle, based on the road vehicle speed data. Denote the traffic flow influence factor based on the road traffic flow when the \(t\)-th passing through the connection path corresponding to the \(j\)-th edge in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(a_{kj}^N\) denotes the number of times the connection path corresponding to the \(j\)-th edge in the delivery task graph is passed through in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(a_{k}^N\) denotes the number of connection paths corresponding to the edges in the delivery task graph passed through in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(R\) j Denote the connection path corresponding to the \(j\)-th edge in the delivery task graph. \(LC\) denotes the set of restricted paths. \(u\) j Denote the path usage factor based on the campus function data of the connection path corresponding to the \(j\)-th edge in the delivery task graph. \(1(R\) j \(\in LC\) means taking the value of 1 when the connection path corresponding to the \(j\)-th edge in the delivery task graph belongs to the set of restricted paths. Denote the vehicle load when the \(t\)-th passing through the connection path corresponding to the \(j\)-th edge in the \(k\)-th delivery plan in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(\max(w\) ak ) denotes the maximum load of the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. \(\gamma_1\) denotes the load adjustment factor, and \(\gamma_2\) denotes the path repetition adjustment factor. Denote the minimum delivery plan evaluation value of all delivery plans in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle. Denote the maximum delivery plan evaluation value of all delivery plans in the subset of delivery plans of the \(a\)-th intelligent unmanned vehicle;

[0124] Based on the delivery plan evaluation values of the delivery plans corresponding to all initialization examples in the subset of delivery plans of each intelligent unmanned vehicle in the delivery plan set, determine the delivery sub-evaluation set of each intelligent unmanned vehicle.

[0125] Based on the delivery sub-evaluation sets of all intelligent unmanned vehicles, determine the delivery evaluation set.

[0126] In this embodiment, the restricted paths refer to those paths that may not be suitable or allowed for unmanned delivery vehicles due to factors such as the functional zoning of the campus or traffic control. For example, certain areas may be teaching building areas with dense crowds or restricted areas, and the delivery paths will be restricted by these. Through the campus function data (such as teaching areas, dormitory areas, restricted areas, etc.) in the spatial layout, these restricted paths are identified and determined; for another example: when the function of a certain building in the campus building data is relatively important, the connection path corresponding to the edge near the building in the delivery task map is determined as the restricted path. The more important the function of a building is, the smaller the path usage factor of the connection path corresponding to the edge near the building is, and the value ranges from 0 to 1. The restricted path set is determined based on all the restricted paths in the delivery task map.

[0127] In this embodiment, each initialization example represents a feasible delivery plan, and its impact on the overall delivery task is evaluated, and its delivery plan evaluation value is calculated.

[0128] In this embodiment, all the delivery plan subsets of each intelligent unmanned vehicle are evaluated to determine the delivery sub-evaluation set of each vehicle.

[0129] The beneficial effects of the above technology are as follows: Based on the spatial layout data, the delivery plan set, and the real-time traffic data, the delivery evaluation set of the delivery path set is evaluated, and the delivery plan selection of each intelligent unmanned vehicle can be dynamically optimized, avoiding congestion, improving the delivery efficiency and accuracy, and intelligently adapting to a more complex and changeable campus environment.

[0130] Embodiment 7:

[0131] Based on Embodiment 6, a campus delivery intelligent unmanned vehicle determines the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, including:

[0132] Select the delivery plan corresponding to the maximum delivery plan evaluation value in the delivery sub-evaluation set of each intelligent unmanned vehicle in the delivery evaluation set as the sub-optimal delivery plan of each intelligent unmanned vehicle;

[0133] Based on the sub-optimal delivery plans of all intelligent unmanned vehicles, the optimal delivery plan is determined.

[0134] In this embodiment, the delivery sub-evaluation set represents the set of delivery plan evaluation values of all delivery plans in the delivery plan set of the corresponding intelligent unmanned vehicle.

[0135] In this embodiment, among the delivery plan evaluation values of all delivery plans in the delivery sub-evaluation set, the maximum delivery plan evaluation value represents the evaluation result of the optimal path after considering all relevant restrictions and parameters. By selecting the delivery plan with the largest evaluation value, it can be ensured that each vehicle selects the most efficient and adaptable path.

[0136] In this embodiment, by selecting the delivery route corresponding to the maximum delivery plan evaluation value, the sub-optimal delivery plan for each intelligent unmanned vehicle is determined. This plan can maximize the delivery efficiency under the current traffic and road conditions, and reduce the situations of time delay or unreasonable routes.

[0137] The beneficial effects of the above technology are as follows: By determining the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, it can ensure that all intelligent unmanned vehicles select the corresponding optimal plan in a complex environment, providing stronger adaptability and flexibility, and improving the delivery efficiency and accuracy.

[0138] Embodiment 8:

[0139] Based on Embodiment 1, a campus delivery intelligent unmanned vehicle obtains the real-time delivery status of all intelligent unmanned vehicles to determine real-time delivery data, including:

[0140] Obtain the delivery status of each intelligent unmanned vehicle based on its corresponding optimal delivery plan in real time to determine real-time sub-delivery data;

[0141] Determine the real-time delivery data based on the real-time sub-delivery data of all intelligent unmanned vehicles.

[0142] In this embodiment, the delivery status of each intelligent unmanned vehicle includes the current location information, the delivery task being executed, the estimated arrival time, etc. Through real-time monitoring and data acquisition technologies, these information can be continuously obtained to ensure that the delivery status of each vehicle is updated at all times. According to the sub-optimal delivery plan, the status data of the vehicle will be updated in real time to reflect the actual delivery situation.

[0143] In this embodiment, the real-time sub-delivery data refers to the delivery progress, current location, remaining route, estimated arrival time, etc. data obtained in real time by each intelligent unmanned vehicle during the delivery process based on its sub-optimal delivery plan. Through these information, the execution status of each delivery task can be tracked and adjusted in real time.

[0144] In this embodiment, the real-time delivery data is the comprehensive information obtained by summarizing the real-time sub-delivery data of all vehicles. This data reflects the current delivery status in the entire unmanned vehicle delivery, helps to optimize the route, adjust the vehicle tasks, and respond to possible delivery delays.

[0145] The beneficial effects of the above technology are as follows: Obtaining the real-time delivery status of all intelligent unmanned vehicles to determine the real-time delivery data can ensure the efficiency and flexibility of the unmanned vehicle delivery during the task execution, optimize the delivery tasks in a timely manner, avoid delays and improve the overall delivery efficiency, and improve the adaptability and response ability of the intelligent unmanned vehicle delivery.

[0146] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent unmanned vehicle for campus delivery, characterized in that: include: S1: Obtain the spatial layout data and real-time traffic data of the campus, and the delivery task data of all intelligent unmanned vehicles; S2: Analyze the spatial layout data of the campus and the delivery task data of all intelligent unmanned vehicles, determine the delivery task graph, and determine the delivery plan set based on the delivery task graph and particle swarm algorithm; S3: Evaluate the distribution plan set based on the spatial layout data, the distribution plan set and the real-time traffic data to determine the distribution evaluation set of the distribution plan set; S4: Determine the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, and obtain the delivery status of all intelligent unmanned vehicles in real time to determine the real-time delivery data; S5: Optimize the optimal delivery plan for intelligent unmanned vehicles based on real-time delivery data and a set of delivery plans.

2. According to claim 1, the intelligent unmanned vehicle for campus delivery is characterized in that: The spatial layout data includes campus road data, campus building data and campus function data, wherein the campus building data at least includes the building names and building locations of all buildings on campus; The delivery task data of all intelligent unmanned vehicles at least includes the shipping address and receiving address of all delivery tasks; The real-time traffic data includes at least road traffic volume data and road vehicle speed data.

3. The campus delivery intelligent unmanned vehicle according to claim 2 is characterized in that: Analyze the spatial layout data of the campus and the delivery task data of all intelligent unmanned vehicles to determine the delivery task map, including: Preprocess the acquired spatial layout data and standardize the acquired delivery task data of all intelligent unmanned vehicles; Determine a node set of the delivery task data based on the building names and building locations of all buildings in the campus building data in the spatial layout data; Determine the connection path between every two nodes in the node set and the path length of the connection path based on the campus road data in the spatial layout data, and determine the edge set of the delivery task data based on the connection path between all nodes in the node set and the path length of the connection path; A delivery task graph of the delivery task data is determined based on a node set and an edge set of the delivery task data.

4. The campus delivery intelligent unmanned vehicle according to claim 2 is characterized in that: Determine the delivery solution set based on the delivery task graph and particle swarm algorithm, including: Based on the shipping addresses and receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, determine the shipping node, receiving node and multiple feasible delivery paths for each delivery task in the delivery task data; Determine all delivery tasks with the same delivery node in the delivery task data as a task cluster, and assign one or more intelligent unmanned vehicles to each task cluster based on the number of delivery tasks in each task cluster; Based on the receiving nodes of all delivery tasks of each task cluster, determining a vehicle delivery set of each intelligent unmanned vehicle of each task cluster, wherein the vehicle delivery set of the intelligent unmanned vehicle includes multiple delivery tasks with the same delivery node; Based on the delivery task graph, the vehicle delivery set of each intelligent unmanned vehicle, and multiple feasible delivery paths of each delivery task in the vehicle delivery set, an initialization particle swarm of each intelligent unmanned vehicle is randomly generated, and a delivery solution subset of each intelligent unmanned vehicle is determined based on the initialization particle swarm; A set of delivery plans is determined based on a subset of delivery plans of all intelligent unmanned vehicles.

5. The campus delivery intelligent unmanned vehicle according to claim 4 is characterized in that: Based on the shipping addresses and receiving addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles, the shipping node, receiving node and multiple feasible delivery paths of each delivery task in the delivery task data are determined, including: The delivery addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles are split based on the building names of all buildings in the campus building data in the spatial layout data, and the characters before and after the building name of any building in the delivery address of each delivery task in the delivery task data are determined as the first separator characters for each delivery task; The delivery addresses of all delivery tasks in the delivery task data of all intelligent unmanned vehicles are split based on the building names of all buildings in the campus building data in the spatial layout data, and the characters before and after the building name of any building in the delivery address of each delivery task in the delivery task data are determined as the second separator characters for each delivery task; Determine a delimiter character set for the delivery task data based on first delimiter characters and second delimiter characters for all delivery tasks in the delivery task data for all intelligent unmanned vehicles; The shipping address of each delivery task in the delivery task data is truncated based on the segmentation character set to determine the shipping node of each delivery task in the delivery task data. At the same time, the receiving address of each delivery task in the delivery task data is truncated based on the segmentation character set to determine the receiving node of each delivery task in the delivery task data. Match the shipping node and receiving node of each delivery task in the delivery task data with the building names of all buildings in the campus building data in the spatial layout data, manually identify the shipping address and receiving address of each delivery task that does not fully match, and update the shipping node and receiving node of each delivery task that does not fully match; Based on the shipping node, the receiving node and the delivery task graph of each delivery task in the delivery task data, multiple feasible delivery paths for each delivery task in the delivery task data are determined.

6. The campus delivery intelligent unmanned vehicle according to claim 4, characterized in that: Based on the spatial layout data, the distribution plan set and the real-time traffic data, the distribution plan set is evaluated to determine the distribution evaluation set of the distribution path set, including: Determine the restricted path set in the delivery task graph based on the campus function data in the spatial layout data; Determine a delivery plan evaluation value of a delivery plan corresponding to each initialization example in the delivery plan subset of each intelligent unmanned vehicle in the delivery plan set based on the restricted path set, the real-time traffic data, and the delivery plan subset of each intelligent unmanned vehicle in the delivery plan set; Among them, F ak represents the delivery plan evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, FR ak represents the delivery path evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, FL ak represents the delivery penalty evaluation value of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, α1 represents the delivery path evaluation weight, α2 represents the delivery penalty evaluation weight, d j represents the path length of the connection path corresponding to the jth edge in the delivery task graph, represents the average vehicle speed based on the road speed data on the connection path when the a-th intelligent unmanned vehicle passes the connection path corresponding to the j-th edge for the t-th time in the k-th delivery plan in the delivery plan subset, represents the traffic flow impact factor based on road traffic flow when the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle passes through the connection path corresponding to the jth edge for the tth time, akjN represents the number of times the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle passes through the connection path corresponding to the jth edge in the delivery task graph, akN represents the number of connection paths corresponding to the edge in the delivery task graph passed by the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle, R j represents the connection path corresponding to the jth edge in the delivery task graph, LC represents the restricted path set, and u j represents the path usage factor of the connection path corresponding to the jth edge in the distribution task graph based on campus function data, 1(R j ∈LC means that the corresponding connection path of the jth edge in the distribution task graph belongs to the restricted path set and takes the value of 1. represents the vehicle load of the kth delivery plan in the delivery plan subset of the ath intelligent unmanned vehicle when it passes the connection path corresponding to the jth edge for the tth time, max(w ak ) represents the maximum load of the a-th intelligent unmanned vehicle’s distribution plan subset, γ1 represents the load adjustment factor, γ2 represents the path repetition adjustment factor, represents the minimum delivery plan evaluation value of all delivery plans in the delivery plan subset of the a-th intelligent unmanned vehicle, represents the maximum delivery plan evaluation value of all delivery plans in the delivery plan subset of the a-th intelligent unmanned vehicle; Determine a delivery sub-evaluation set for each intelligent unmanned vehicle based on the delivery plan evaluation values ​​of the delivery plans corresponding to all the initialized examples in the delivery plan sub-set of each intelligent unmanned vehicle in the delivery plan set; A delivery evaluation set is determined based on the delivery sub-evaluation sets of all intelligent unmanned vehicles.

7. The campus delivery intelligent unmanned vehicle according to claim 6, characterized in that: Determine the optimal delivery plan for all intelligent unmanned vehicles based on the delivery evaluation set, including: Select the delivery plan corresponding to the maximum delivery plan evaluation value in the delivery sub-evaluation set of each intelligent unmanned vehicle in the delivery evaluation set as the sub-optimal delivery plan for each intelligent unmanned vehicle; Determine the optimal delivery plan based on the sub-optimal delivery plans of all intelligent unmanned vehicles.

8. The campus delivery intelligent unmanned vehicle according to claim 1, characterized in that: Obtain the delivery status of all intelligent unmanned vehicles in real time to determine real-time delivery data, including: Obtain the delivery status of each intelligent unmanned vehicle based on the corresponding optimal delivery plan in real time, and determine the real-time sub-delivery data; Determine real-time delivery data based on the real-time sub-delivery data of all intelligent unmanned vehicles.

Images