Unmanned aerial vehicle blood distribution path optimization method and system for mountainous area three-dimensional terrain

Abstract

The invention discloses an unmanned aerial vehicle blood distribution path optimization method and system for a mountainous area three-dimensional terrain, and the method comprises the steps: building an objective function through the sum of the minimum fixed cost, the transportation energy consumption cost, the refrigeration cost and the time window penalty cost, building a mountainous area complex constraint condition system, setting a decision variable definition domain, and carrying out the optimization of the unmanned aerial vehicle blood distribution path. A path optimization model adapting to the three-dimensional terrain of the mountainous area is constructed; and solving the model by using an improved elite ant colony algorithm, and performing iterative optimization through a dynamic volatilization factor and an elite reward and punishment mechanism to obtain an optimal unmanned aerial vehicle blood distribution path and a corresponding parameter scheme. According to the method, a corresponding algorithm is designed on the basis of the path optimization model to improve the global optimization capability and the convergence rate, the cost minimization, the time maximization and the reliability enhancement of the blood distribution path are realized, and the method is particularly suitable for emergency medical logistics scenes in mountainous areas.

Description

Technical Field

[0001] This invention relates to the field of robot autonomous navigation and path planning technology, specifically to a method and system for optimizing blood delivery routes for unmanned aerial vehicles (UAVs) in mountainous three-dimensional terrain. Background Technology

[0002] Blood, as a critical resource in clinical treatment, is characterized by a short preservation period and high requirements for timely delivery; its delivery efficiency is directly related to patient safety. Meanwhile, the increasing demand for medical services and the strengthening of regional medical collaboration are driving the transformation of blood delivery from a traditional single-point model to a regionalized and networked model.

[0003] In the existing blood distribution system, blood is typically delivered directly from central blood banks to medical institutions, or transferred through a limited number of transit stations. In mountainous and complex terrain areas like Sichuan and Chongqing, where the terrain is undulating and transportation conditions are difficult, traditional ground vehicle delivery methods suffer from problems such as circuitous routes, high costs, and poor timeliness. Especially during frequent geological disasters or public health emergencies, the emergency response capacity of blood distribution is insufficient, easily leading to blood waste or medical delays. In recent years, drone technology, due to its advantages of speed, flexibility, and contactless delivery, has been gradually applied to medical supply distribution, providing a new technological means to improve delivery efficiency.

[0004] However, existing drone delivery route optimization methods are mostly designed for plains or urban environments, and have the following limitations in mountainous terrain: on the one hand, the route planning does not fully consider three-dimensional constraints such as terrain elevation, obstacle distribution and weather conditions, resulting in low route feasibility; on the other hand, there is a lack of integrated optimization models for the special needs of blood delivery, such as time window constraints and refrigeration costs, making it difficult to ensure blood viability.

[0005] In summary, existing technologies still lack a method that can comprehensively consider three-dimensional constraints such as terrain elevation, obstacle distribution, and meteorological conditions, and on this basis, conduct overall planning for the specific needs of blood distribution, so as to meet the comprehensive requirements of regional blood distribution networks in terms of cost control and distribution efficiency. Summary of the Invention

[0006] This invention provides a method and system for optimizing blood delivery routes by drones in mountainous three-dimensional terrain, in order to solve the problems of insufficient rationality of dynamic scheduling of drones in mountainous terrain, low optimization efficiency, and difficulty in balancing delivery costs and service coverage requirements in the existing technology.

[0007] According to the first aspect, one embodiment provides a method for optimizing unmanned aerial vehicle (UAV) blood delivery routes for three-dimensional terrain in mountainous areas, the method comprising: Collect basic parameter data and define decision variables; Construct models for transportation energy consumption costs, cooling costs, and time window penalty costs; The objective function is established to minimize the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs. A complex constraint system for mountainous areas is established, and the domain of decision variables is defined. Thus, a path optimization model adapted to the three-dimensional terrain of mountainous areas is constructed. An improved elite ant colony algorithm was used to solve a path optimization model adapted to three-dimensional mountainous terrain. The objective function was transformed into a fitness function, and the constraints were transformed into path feasibility judgment rules. The optimal UAV blood delivery path and corresponding parameter scheme were obtained through iterative optimization using dynamic volatile factors and elite reward and punishment mechanisms.

[0008] Furthermore, basic parameter data is collected, and decision variables are defined, specifically including: Obtain the set of blood transfer stations determined by the site selection component. Gathering with blood receiving stations The geographic coordinates, including elevation, constitute a three-dimensional spatial coordinate dataset. Obtain the predicted blood demand at each receiving station. and service hours ,in For the earliest delivery time, The latest delivery time, exceeding This will result in the blood becoming unusable; Obtain drone performance parameters, including: drone set Each of the drones The parameters include maximum load capacity. cruising speed Maximum range Unloaded tare weight Windward area air drag coefficient Flight altitude ; Obtain environmental parameters for mountainous areas, including: average wind speed. Wind angle air density ; Obtain cost parameters, including: fixed cost per flight. Unit transportation cost coefficient during the flight phase Unit cooling cost coefficient during flight Refrigeration cost coefficient per unit time for loading and unloading blood ; Obtain flight status parameters: heading angle Ground speed Power consumption Service Hours Blood delivery volume Flight time Accelerated flight state ratio Proportion of uniform flight state Proportion of decelerated flight state takeoff power , decrease landing power Takeoff and ascent speed descent landing speed mileage by battery level at arrival station mileage after leaving the station ; Define decision variables: Unmanned aerial vehicle (UAV) service path variables Unmanned aerial vehicle (UAV) service site variables Site service relationship variables UAVs use state variables .

[0009] Furthermore, a model is constructed to represent transportation energy consumption costs, cooling costs, and time window penalty costs, specifically including: Transportation energy consumption cost modeling includes: Ground speed correction under wind field interference: Calculating the ground speed of drones in wind conditions with wind angle The actual ground speed affected is: ; Payload-power dynamic coupling: Flight power is: ,in, Represents the radius of the drone's rotor; Segmented Flight Energy Consumption: The flight process is divided into three stages: takeoff and climb, level cruise, and landing and descent. The energy consumption of each stage is calculated separately and then summed. The final transportation energy cost C5 is:

[0010] in, This refers to a collection of drone bases, distribution centers, or takeoff points.

[0011] Furthermore, a model is constructed to represent transportation energy consumption costs, cooling costs, and time window penalty costs, specifically including: Refrigeration cost modeling includes: The cooling cost for C6 is:

[0012] in, and They represent the first [number] starting from the base. Type of drone and the first from the hospital or blood collection site The flight time required for a type of drone to perform a delivery task from node i to node j; This refers to a collection of drone bases, distribution centers, or takeoff points.

[0013] Furthermore, a model is constructed to represent transportation energy consumption costs, cooling costs, and time window penalty costs, specifically including: Modeling the time window penalty cost includes: The time window segmentation penalty cost C7 is:

[0014]

[0015] in, The actual time when the k-th type of UAV arrives at node i. This represents a very large positive penalty value. and These are the penalty coefficients for arriving early and the penalty coefficients for arriving late. and These are the earliest and latest dynamically allowed service times, respectively. This refers to a collection of drone bases, distribution centers, or takeoff points.

[0016] Furthermore, an objective function is established to minimize the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs, specifically including: Objective function:

[0017] Where C represents the total cost, and C4, C5, C6, and C7 represent fixed costs, transportation energy costs, refrigeration costs, and time window penalty costs, respectively.

[0018] Furthermore, a complex constraint system for mountainous areas will be established, specifically including: Each blood collection center and blood receiving station can only accept the service of one drone, as shown below:

[0019] in, It is a binary decision variable. When the value is 1, it means that the service of node i is undertaken by the k-th type of drone. When the value is 0, it means that it is not undertaken by the drone. This refers to a collection of drone bases, distribution centers, or takeoff points. This refers to a collection of blood receiving stations; Each blood collection center and blood receiving station on a path can only be visited once, as shown below:

[0020]

[0021] The flow balance condition for each drone after visiting the blood collection center and blood receiving station requires that the number of paths entering and leaving the station be equal, expressed as:

[0022]

[0023] in, This refers to the path taken by drone k from node i to base m; This refers to the path taken by drone k from base m to node j; It is a 0-1 binary decision variable. When the value is 1, it means that the base m is provided with services by the k-th type of drone. When the value is 0, it means that it is not provided with services by the drone. A collection representing drone bases, distribution centers, or takeoff points; This refers to the path taken by drone k from node i to base h. This refers to the path taken by drone k from base h to node j; It is a 0-1 binary decision variable. When the value is 1, it means that the base h is provided by the k-th type of UAV. When the value is 0, it means that it is not provided by the UAV. If the total blood delivery volume at stations along a route is less than the maximum carrying capacity of a drone, it is represented as:

[0024] in, It is a 0-1 binary decision variable. When the value is 1, it means that node i is served by the k-th type of drone. When the value is 0, it means that it is not served by the drone. All drones departing from and returning to the blood transfer station are represented as follows:

[0025]

[0026] in, and Both are 0-1 binary decision variables. When the value is 1, it means: the k-th type of UAV flies from node i to node j, and the k-th type of UAV flies from node j to node i, respectively. The blood flow balance at the blood transfer station is represented as:

[0027] in, and Representing the first The amount of blood carried by the type of drone on the flight segment from node i to node j and the first The amount of blood carried by the type of drone on the flight segment from node j to node h; Once the drone departs from the transfer station and is fully charged, its maximum flight distance is represented as:

[0028] Ensuring the drone has enough power to return to the blood transfusion station is expressed as:

[0029] The drone's flight distance remains unchanged when entering and leaving the site, as shown below:

[0030] The change in flight distance of a drone passing through two adjacent stations is represented as follows:

[0031]

[0032] in, This represents the flight distance from node i to node j; It is a 0-1 binary decision variable. When the value is 1, it means that the drone has chosen the path from node i to fly directly to node j. This represents a very large positive penalty value; This refers to a collection of drone bases, distribution centers, or takeoff points. Ensuring the continuity of blood delivery time by drone is represented as follows:

[0033]

[0034] in, and These represent the start or departure times of the type k drone at nodes i and j, respectively. The total amount of blood awaiting delivery from all blood collection centers closest to the blood transfer station cannot exceed the storage limit of the corresponding blood transfer station, expressed as:

[0035] in, This refers to the maximum storage capacity of blood transfer station j.

[0036] Furthermore, an improved elite ant colony algorithm is used to solve a path optimization model adapted to three-dimensional mountainous terrain. The objective function is transformed into a fitness function, and the constraints are transformed into path feasibility judgment rules. Through iterative optimization using a dynamic volatile factor and an elite reward / penalty mechanism, the optimal UAV blood delivery path and corresponding parameter scheme are obtained, specifically including: a. Initialize a series of control parameters at the beginning of the algorithm and set the pheromone values ​​of all path nodes to the same constant to ensure that the initial exploration is random and does not favor any path. b. During the path exploration process, the ant combines the pheromone intensity on the current path with the heuristic expectation value to calculate the state transition probability to each feasible node; the above process is repeated until all nodes have been visited, thereby generating all possible path combinations. Calculating ants From a directed path Visibility above: ,in For path length, , This indicates the node number, and also the start and end points of the path. Calculate the state transition probability of the ant choosing a path:

[0037] in Indicates the pheromone concentration coefficient. Represents the heuristic function coefficients, This refers to the website. A station that meets all constraints but has not yet been reached; This indicates historical experience; if many ants have previously traversed the current path, the pheromone concentration will be high, thus increasing the attraction. This represents the quality of the current path itself; the shorter the distance, the greater the attraction. c. When all ants have reached the destination or there is no feasible path, the pheromone is updated based on the reward and punishment mechanism. That is, the worst path will be punished, and its pheromone concentration will decay rapidly, thereby reducing the probability of it being selected. A dynamic evaporation factor is introduced, and the improved formula correlates the value of the pheromone evaporation factor with the number of iterations. Calculation after adding a reward and punishment mechanism Pheromones concentration at specific times:

[0038]

[0039] in This indicates the remaining concentration after the previous stage of volatilization. Indicates the number of newly added pheromones. and These represent the reward and penalty factors for the optimal and worst paths, respectively. and These are the optimal and worst path lengths during the iteration process, respectively. As an enhancing factor, As a reward factor; Introducing dynamic volatile factors: When the initial number of iterations is lower than a preset threshold, the volatile factor takes the maximum value. This ensures that pheromones evaporate quickly in the early stages of the search process, thus avoiding premature entrapment in local optima; among which... Represents the number of iterations. Represents the maximum number of iterations. This represents the minimum value of the volatile factor; d. For path optimization problems with time windows, add time window influencing factors to improve the ant state transition rules; Time window deviation function:

[0040] in Represents the delivery arrival time; the time window is... Indicates; among which Indicates customer Earliest permitted arrival time Indicates customer Latest permitted arrival time; Time window correction function:

[0041]

[0042] The above formula represents the ant's position on the node. by state transition probability Select to move to node .

[0043] Furthermore, the method also includes: Post-validation optimization: Integrate a 3D flight simulation engine to perform collision detection, endurance verification, and time window compliance checks on the algorithm's output path, mark abnormal arc segments, and trigger local replanning.

[0044] According to a second aspect, one embodiment provides a drone blood delivery route optimization system for mountainous three-dimensional terrain, the system comprising: The basic data acquisition module is used to collect basic parameter data and define decision variables; The cost modeling module is used to build models for transportation energy consumption costs, cooling costs, and time window penalty costs. The path optimization module is used to build an objective function that minimizes the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs. It establishes a complex constraint system for mountainous areas and defines the domain of decision variables, thereby constructing a path optimization model adapted to the three-dimensional terrain of mountainous areas. The optimization solution module is used to solve the path optimization model adapted to the three-dimensional terrain of mountainous areas using an improved elite ant colony algorithm. It transforms the objective function into a fitness function and the constraints into path feasibility judgment rules. Through dynamic volatile factors and elite reward and punishment mechanisms, iterative optimization is performed to obtain the optimal UAV blood delivery path and corresponding parameter scheme.

[0045] This invention provides a method and system for optimizing blood delivery routes using unmanned aerial vehicles (UAVs) in mountainous three-dimensional terrain, which has the following beneficial effects: 1) The improved elite ant colony algorithm proposed in this invention demonstrates superior path planning capabilities in practical applications in complex regions. Compared to traditional methods, this algorithm systematically optimizes delivery order and routes, significantly reducing the total delivery time. In regional network examples involving multiple centers and multiple demand points, the algorithm effectively reduces vehicle empty driving distances and waiting times through intelligent iteration and pheromone mechanism adjustments, thereby compressing the total time required to complete all tasks. This means that blood products can be delivered from central blood banks to various medical terminals more quickly, greatly enhancing the timeliness and guarantee of blood supply, and saving valuable time for clinical blood use.

[0046] 2) Furthermore, when faced with accessibility constraints imposed by special terrains such as mountainous and hilly areas, the algorithm can automatically adjust its path evaluation system to find feasible and efficient solutions. Simultaneously, in simulated scenarios involving sudden fluctuations in demand, temporary changes in traffic conditions, or abrupt weather changes, the scheduling scheme generated by the algorithm demonstrates good stability and anti-interference capabilities. This ensures that the blood delivery plan is not only efficient under ideal conditions but also reliable under complex real-world conditions, significantly improving the resilience and robustness of the entire delivery system in the face of uncertainty.

[0047] 3) By introducing a dynamic scheduling strategy and a refined cost control model, this invention achieves a reduction in overall costs. The algorithm not only considers the shortest path but also comprehensively weighs multiple cost factors such as transportation, manpower, and time window defaults. Particularly in handling time window constraints, through optimized penalty mechanism settings, the algorithm prioritizes delivery tasks with strict timeliness requirements, thereby significantly reducing various derivative costs caused by delayed delivery. Sensitivity analysis shows that the model has good economic adjustment capabilities, guiding overall operating costs towards a more optimal level while meeting high service level requirements, thus improving resource utilization efficiency.

[0048] 4) Beyond improving conventional delivery efficiency, the technical solution of this invention offers significant value in medical emergency response. Its optimized contactless delivery and collaborative network model is particularly suitable for emergency scenarios such as large-scale public health events or natural disasters. In such scenarios, the algorithm can quickly plan safe and efficient delivery routes, minimizing personnel contact and the risk of cross-infection, while ensuring the accurate and rapid supply of life-saving blood. By reducing the waste of blood products caused by untimely delivery and ensuring unimpeded access for critically ill patients, this technology provides key technical support for building a more resilient and responsive regional medical emergency logistics system, yielding significant social benefits. Attached Figure Description

[0049] Figure 1 A flowchart illustrating a method for optimizing unmanned aerial vehicle (UAV) blood delivery routes for three-dimensional mountainous terrain, as provided in one embodiment of the present invention; Figure 2 This is a flowchart illustrating a specific implementation of a method for optimizing blood delivery routes for unmanned aerial vehicles (UAVs) in mountainous three-dimensional terrain, as provided in one embodiment of the present invention. Detailed Implementation

[0050] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0051] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0052] The first embodiment of this invention provides a method for optimizing the blood delivery route of unmanned aerial vehicles (UAVs) in mountainous three-dimensional terrain. The following is a combination of... Figure 1 and Figure 2 Please provide a detailed explanation.

[0053] like Figure 1 As shown, in step S100, basic parameter data is collected and decision variables are defined.

[0054] In this embodiment, the problem feature analysis and basic parameter initialization are as follows: Obtain the set of blood transfer stations determined by the site selection component. Gathering with blood receiving stations The geographic coordinates (including elevation) constitute a three-dimensional spatial coordinate dataset.

[0055] Obtain the predicted blood demand at each receiving station. and service hours ,in For the earliest delivery time, The latest delivery time, exceeding This will cause the blood to become unusable.

[0056] Obtaining Drone Performance Parameters: Drone Collection Each of the drones Maximum load cruising speed Maximum range Unloaded tare weight Windward area air drag coefficient Flight altitude .

[0057] Obtaining environmental parameters in mountainous areas: average wind speed Wind angle air density Obtain cost parameters: Fixed cost per flight Unit transportation cost coefficient during the flight phase Unit cooling cost coefficient during flight Refrigeration cost coefficient per unit time for loading and unloading blood .

[0058] Obtain flight status parameters: heading angle Ground speed Power consumption Service Hours Blood delivery volume Flight time Accelerated flight state ratio Proportion of uniform flight state Proportion of decelerated flight state takeoff power , decrease landing power Takeoff and ascent speed descent landing speed mileage by battery level at arrival station mileage after leaving the station .

[0059] Define decision variables: Unmanned aerial vehicle (UAV) service path variables Unmanned aerial vehicle (UAV) service site variables Site service relationship variables UAVs use state variables .

[0060] like Figure 1 As shown, in step S200, a model for transportation energy consumption cost, cooling cost, and time window penalty cost is constructed.

[0061] The above steps specifically include: S210, Three-dimensional terrain-coupled transportation energy cost modeling Constructing an energy consumption calculation system that differs from traditional two-dimensional models to accurately characterize flight characteristics in mountainous areas: Ground speed correction under wind field interference: Calculating the ground speed of drones in wind conditions With wind direction The actual ground speed affected is: (1) Payload-power dynamic coupling: Flight power is: (2) Segmented flight energy consumption: The flight process is divided into three stages: takeoff and climb, level cruise, and landing and descent. The energy consumption is calculated separately for each stage and then summed up.

[0062] The final transportation energy cost is: (3) S220, integrating cold chain temperature control and time window penalty cost Construct the cost function under the special requirements of blood products: S221, refrigeration cost: (4) S222, Time Window Segmentation Penalty Cost: (5) (6) like Figure 1 As shown, in step S300, an objective function is established to minimize the sum of fixed costs, transportation energy consumption costs, cooling costs, and time window penalty costs. A complex constraint system for mountainous areas is established, and the domain of decision variables is defined, thereby constructing a path optimization model adapted to the three-dimensional terrain of mountainous areas.

[0063] S310, Construction of Complex Constraint System in Mountainous Areas: Establish a set of constraints to ensure flight safety and mission reachability: (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (twenty one) (twenty two) (twenty three) Equation (7) indicates that each blood collection center and blood receiving station can only accept the service of one drone.

[0064] Equations (8) and (9) indicate that all blood collection centers and blood receiving stations on a path can only be visited once.

[0065] Equations (10) and (11) represent the flow balance conditions for each drone after visiting the blood collection center station and the blood receiving station, requiring that the number of paths entering and leaving the station be equal.

[0066] Equation (12) indicates that the total blood delivery volume of stations along a route is less than the maximum carrying capacity of the drone.

[0067] Equations (13) and (14) represent all drones departing from and returning to the blood transfer station.

[0068] Equation (15) represents the blood flow balance at the blood transfer station.

[0069] Equation (16) indicates that the drone can fly a maximum distance after being fully charged from the transfer station.

[0070] Equation (17) ensures that the drone has enough power to return to the blood transfer station.

[0071] Equation (18) indicates that the flight distance of the UAV remains unchanged when entering and leaving the station.

[0072] Equations (19) and (20) represent the changes in the flight distance of the UAV when passing through two adjacent stations.

[0073] Equations (21) and (22) represent ensuring the continuity of blood delivery time by drone.

[0074] Equation (23) indicates that the total amount of blood to be delivered from all blood collection centers closest to the blood transfer station cannot exceed the storage limit of that blood transfer station.

[0075] S320, Considering the above costs and constraints, the drone blood delivery route optimization model constructed in this embodiment is as follows: Objective function: (twenty four) That is, to minimize the sum of fixed costs, energy consumption costs of three-dimensional terrain coupling, cold chain refrigeration costs, and time window penalty costs; Constraint system: Includes the above-mentioned constraints such as flow balancing, service uniqueness, and load capacity; Variable domain: .

[0076] like Figure 1 As shown, in step S400, the improved elite ant colony algorithm is used to solve the path optimization model adapted to the three-dimensional terrain of the mountainous area. The objective function is transformed into a fitness function, the constraint conditions are transformed into path feasibility judgment rules, and the optimal UAV blood delivery path and corresponding parameter scheme are obtained through iterative optimization by dynamic volatile factor and elite reward and punishment mechanism.

[0077] The above steps specifically include: S410, Model Solving Strategy and Algorithm Interface Design Given the NP-hard nature of the model and the complexity of mountain constraints, this embodiment adopts a three-stage solution strategy: 1) Preprocessing stage: Based on the site selection results, fixed service allocation is performed, and infeasible arc segments (excessive endurance, flying over no-fly zones) are eliminated. An energy consumption matrix and a terrain complexity matrix are constructed. The terrain complexity matrix quantifies the terrain challenge of flight segments between each node. This matrix will serve as a key input for calculating the flight energy consumption matrix, screening feasible arc segments, and providing the terrain environment data foundation for the final 3D simulation verification. 2) Algorithm solution stage: The objective function is transformed into the fitness function of the improved elite ant colony algorithm, and the constraints are transformed into path feasibility judgment rules. The algorithm is then iteratively optimized through dynamic volatile factors and elite reward and punishment mechanisms. 3) Post-optimization verification: Perform three-dimensional simulation verification on the output path to check that the arrival time falls within the hard time window and the remaining battery level is always higher than the safety threshold. Only after passing the post-verification can the actual delivery task be put into use.

[0078] S420 is an improved and optimized version of the elite ant colony algorithm, used for solving path optimization models. The algorithm specifically includes the following: S421 initializes a series of control parameters at the beginning of the algorithm and sets the pheromone values ​​on all path nodes to the same constant to ensure that the initial exploration is random and does not favor any path.

[0079] In step S422, during path exploration, the ant combines the pheromone intensity on the current path with the heuristic expectation value to calculate the state transition probability to each feasible node. This process is repeated until all nodes have been visited, thus generating all possible path combinations.

[0080] Calculating ants From a directed path Visibility above: (25) in This represents the path length.

[0081] Calculate the state transition probability of the ant choosing a path: (26) in Indicates the pheromone concentration coefficient. Represents the heuristic function coefficients, This refers to the website. A station that meets all constraints but has not yet been reached.

[0082] S423: When all ants have reached the destination or there is no feasible path, the pheromone is updated based on a reward and punishment mechanism. That is, the worst path will be punished, and its pheromone concentration will rapidly decrease, thereby reducing its probability of being selected. A dynamic evaporation factor is introduced, and the improved formula correlates the value of the pheromone evaporation factor with the number of iterations.

[0083] Calculation after adding a reward and punishment mechanism ( Pheromones concentration at specific time points: (27) (28) (29) in and These represent the reward and penalty factors for the optimal and worst paths, respectively. and These are the optimal and worst path lengths during the iteration process, respectively. As an enhancing factor, This is a reward factor.

[0084] Introducing dynamic volatile factors: (30) When the initial number of iterations is small, the volatile factor reaches its maximum value. This ensures that pheromones evaporate quickly in the early stages of the search process, thus avoiding premature entrapment in local optima.

[0085] S424 addresses the path optimization problem with time windows by adding time window influencing factors to improve the ant state transition rules.

[0086] Time window deviation function: (31) in Represents the delivery arrival time; the time window is... express.

[0087] Time window correction function: (32) (33) (34) Equations (33) and (34) represent the ant's position at the node. by state transition probability Select to move to node .

[0088] This embodiment proposes a decision support system for optimizing blood delivery routes using drones in mountainous areas. The system is a hardware and software integrated platform for implementing the above method, and mainly includes the following interconnected modules: Data preprocessing module: responsible for receiving the transfer station-hospital service allocation relationship from the site selection system, obtaining real-time wind speed and direction data from the meteorological department API, extracting three-dimensional coordinates from the digital elevation model (DEM), and establishing a unified spatiotemporal database and energy consumption pre-calculation matrix.

[0089] Cost modeling module: Built-in 3D terrain coupled energy consumption calculation engine, configurable aerodynamic parameters, supports integrated modeling of segmented flight energy consumption, load-power coupling, and cold chain refrigeration costs, and dynamically generates objective function coefficients.

[0090] Constraint Management Module: Provides a graphical interface for configuring drone performance parameters, setting hospital time windows, and automatically constructing matrix representations of seven types of hard constraints.

[0091] Intelligent solution module: Encapsulates and improves the elite ant colony algorithm kernel, supports dynamic volatile factor and mechanism parameter tuning, and provides real-time monitoring of convergence curves and multi-scheme comparison function.

[0092] Path verification module: Integrates a 3D flight simulation engine to perform collision detection, endurance verification, and time window compliance checks on the algorithm output path, mark abnormal arc segments, and trigger local replanning.

[0093] Solution output module: Generates a standardized delivery task list (including drone number, path node sequence, take-off time, and estimated arrival time), a cost analysis report (cost percentage of each item), and a risk assessment report (terrain complexity and weather interference level). It also supports exporting KML track files for direct loading by the drone flight control system.

[0094] Corresponding to the aforementioned method for optimizing unmanned aerial vehicle (UAV) blood delivery routes in mountainous three-dimensional terrain, this invention also discloses a system for optimizing UAV blood delivery routes in mountainous three-dimensional terrain, which specifically includes: The basic data acquisition module is used to collect basic parameter data and define decision variables; The cost modeling module is used to build models for transportation energy consumption costs, cooling costs, and time window penalty costs. The path optimization module is used to build an objective function that minimizes the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs. It establishes a complex constraint system for mountainous areas and defines the domain of decision variables, thereby constructing a path optimization model adapted to the three-dimensional terrain of mountainous areas. The optimization solution module is used to solve the path optimization model adapted to the three-dimensional terrain of mountainous areas using an improved elite ant colony algorithm. It transforms the objective function into a fitness function and the constraints into path feasibility judgment rules. Through dynamic volatile factors and elite reward and punishment mechanisms, iterative optimization is performed to obtain the optimal UAV blood delivery path and corresponding parameter scheme.

[0095] It should be noted that for a detailed description of the UAV blood delivery route optimization system for three-dimensional mountainous terrain provided in the embodiments of the present invention, please refer to the relevant description of the UAV blood delivery route optimization method for three-dimensional mountainous terrain provided in the embodiments of the present invention, which will not be repeated here.

[0096] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A method for optimizing unmanned aerial vehicle (UAV) blood delivery routes in mountainous three-dimensional terrain, characterized in that, The method includes: Collect basic parameter data and define decision variables; Construct models for transportation energy consumption costs, cooling costs, and time window penalty costs; The objective function is established to minimize the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs. A complex constraint system for mountainous areas is established, and the domain of decision variables is defined. Thus, a path optimization model adapted to the three-dimensional terrain of mountainous areas is constructed. An improved elite ant colony algorithm was used to solve a path optimization model adapted to three-dimensional mountainous terrain. The objective function was transformed into a fitness function, and the constraints were transformed into path feasibility judgment rules. The optimal UAV blood delivery path and corresponding parameter scheme were obtained through iterative optimization using dynamic volatile factors and elite reward and punishment mechanisms.

2. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 1, characterized in that, Collect basic parameter data and define decision variables, specifically including: Obtain the set of blood transfer stations determined by the site selection component. Gathering with blood receiving stations The geographic coordinates, including elevation, constitute a three-dimensional spatial coordinate dataset. Obtain the predicted blood demand at each receiving station. and service hours ,in For the earliest delivery time, The latest delivery time, exceeding This will result in the blood becoming unusable; Obtain drone performance parameters, including: drone set Each of the drones The parameters include maximum load capacity. cruising speed Maximum range Unloaded tare weight Windward area air drag coefficient Flight altitude ; Obtain environmental parameters for mountainous areas, including: average wind speed. Wind angle air density ; Obtain cost parameters, including: fixed cost per flight. Unit transportation cost coefficient during the flight phase Unit cooling cost coefficient during flight Refrigeration cost coefficient per unit time for loading and unloading blood ; Obtain flight status parameters: heading angle Ground speed Power consumption Service Hours Blood delivery volume Flight time Accelerated flight state ratio Proportion of uniform flight state Proportion of decelerated flight state takeoff power , decrease landing power Takeoff and ascent speed descent landing speed mileage by battery level at arrival station mileage after leaving the station ; Define decision variables: Unmanned aerial vehicle (UAV) service path variables Unmanned aerial vehicle (UAV) service site variables Site service relationship variables UAVs use state variables .

3. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 2, characterized in that, Construct models for transportation energy consumption costs, cooling costs, and time window penalty costs, specifically including: Transportation energy consumption cost modeling includes: Ground speed correction under wind field interference: Calculating the ground speed of drones in wind conditions with wind angle The actual ground speed affected is: ; Payload-power dynamic coupling: Flight power is: ,in, Represents the radius of the drone's rotor; Segmented Flight Energy Consumption: The flight process is divided into three stages: takeoff and climb, level cruise, and landing and descent. The energy consumption of each stage is calculated separately and then summed. The final transportation energy cost C5 is: in, This refers to a collection of drone bases, distribution centers, or takeoff points.

4. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 2, characterized in that, Construct models for transportation energy consumption costs, cooling costs, and time window penalty costs, specifically including: Refrigeration cost modeling includes: The cooling cost for C6 is: in, and They represent the first [number] starting from the base. Type of drone and the first from the hospital or blood collection site The flight time required for a type of drone to perform a delivery task from node i to node j; This refers to a collection of drone bases, distribution centers, or takeoff points.

5. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 2, characterized in that, Construct models for transportation energy consumption costs, cooling costs, and time window penalty costs, specifically including: Modeling the time window penalty cost includes: The time window segmentation penalty cost C7 is: in, The actual time when the k-th type of UAV arrives at node i. This represents a very large positive penalty value. and These are the penalty coefficients for arriving early and the penalty coefficients for arriving late. and These are the earliest and latest dynamically allowed service times, respectively. This refers to a collection of drone bases, distribution centers, or takeoff points.

6. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 1, characterized in that, The objective function is established to minimize the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs, specifically including: Objective function: Where C represents the total cost, and C4, C5, C6, and C7 represent fixed costs, transportation energy costs, refrigeration costs, and time window penalty costs, respectively.

7. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 2, characterized in that, Establish a complex constraint system for mountainous areas, specifically including: Each blood collection center and blood receiving station can only accept the service of one drone, as shown below: in, It is a binary decision variable. When the value is 1, it means that the service of node i is undertaken by the k-th type of drone. When the value is 0, it means that it is not undertaken by the drone. This refers to a collection of drone bases, distribution centers, or takeoff points. This refers to a collection of blood receiving stations; Each blood collection center and blood receiving station on a path can only be visited once, as shown below: The flow balance condition for each drone after visiting the blood collection center and blood receiving station requires that the number of paths entering and leaving the station be equal, expressed as: in, This refers to the path taken by drone k from node i to base m; This refers to the path taken by drone k from base m to node j; It is a 0-1 binary decision variable. When the value is 1, it means that the base m is provided with services by the k-th type of drone. When the value is 0, it means that it is not provided with services by the drone. A collection representing drone bases, distribution centers, or takeoff points; This refers to the path taken by drone k from node i to base h. This refers to the path taken by drone k from base h to node j; It is a 0-1 binary decision variable. When the value is 1, it means that the base h is provided by the k-th type of UAV. When the value is 0, it means that it is not provided by the UAV. If the total blood delivery volume at stations along a route is less than the maximum carrying capacity of a drone, it is represented as: in, It is a 0-1 binary decision variable. When the value is 1, it means that node i is served by the k-th type of drone. When the value is 0, it means that it is not served by the drone. All drones departing from and returning to the blood transfer station are represented as follows: in, and Both are 0-1 binary decision variables. When the value is 1, it means: the k-th type of UAV flies from node i to node j, and the k-th type of UAV flies from node j to node i, respectively. The blood flow balance at the blood transfer station is represented as: in, and Representing the first The amount of blood carried by the type of UAV on the flight segment from node i to node j and the first The amount of blood carried by the type of drone on the flight segment from node j to node h; Once the drone departs from the transfer station and is fully charged, its maximum flight distance is represented as: Ensuring the drone has enough power to return to the blood transfusion station is expressed as: The drone's flight distance remains unchanged when entering and leaving the site, as shown below: The change in flight distance of a drone passing through two adjacent stations is represented as follows: in, This represents the flight distance from node i to node j; It is a 0-1 binary decision variable. When the value is 1, it means that the drone has chosen the path to fly directly from node i to node j. This represents a very large positive penalty value; This refers to a collection of drone bases, distribution centers, or takeoff points. Ensuring the continuity of blood delivery time by drone is represented as follows: in, and These represent the start or departure times of the type k drone at nodes i and j, respectively. The total amount of blood awaiting delivery from all blood collection centers closest to the blood transfer station cannot exceed the storage limit of the corresponding blood transfer station, expressed as: in, This refers to the maximum storage capacity of blood transfer station j.

8. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 1, characterized in that, An improved elite ant colony algorithm is used to solve a path optimization model adapted to three-dimensional mountainous terrain. The objective function is transformed into a fitness function, and the constraints are transformed into path feasibility judgment rules. Iterative optimization is then performed using a dynamic volatile factor and an elite reward / penalty mechanism to obtain the optimal UAV blood delivery path and corresponding parameter scheme, specifically including: a. Initialize a series of control parameters at the beginning of the algorithm and set the pheromone values ​​of all path nodes to the same constant to ensure that the initial exploration is random and does not favor any path. b. During the path exploration process, the ant combines the pheromone intensity on the current path with the heuristic expectation value to calculate the state transition probability to each feasible node; the above process is repeated until all nodes have been visited, thereby generating all possible path combinations. Calculating ants From a directed path Visibility above: ,in For path length, , This indicates the node number, and also the start and end points of the path. Calculate the state transition probability of the ant choosing a path: in Indicates the pheromone concentration coefficient. Represents the heuristic function coefficients, This refers to the website. A station that meets all constraints but has not yet been reached; This indicates historical experience; if many ants have previously traversed the current path, the pheromone concentration will be high, thus increasing the attraction. This represents the quality of the current path itself; the shorter the distance, the greater the attraction. c. When all ants have reached the destination or there is no feasible path, the pheromone is updated based on the reward and punishment mechanism. That is, the worst path will be punished, and its pheromone concentration will decay rapidly, thereby reducing the probability of it being selected. A dynamic evaporation factor is introduced, and the improved formula correlates the value of the pheromone evaporation factor with the number of iterations. Calculation after adding a reward and punishment mechanism Pheromones concentration at specific times: in This indicates the remaining concentration after the previous stage of volatilization. Indicates the number of newly added pheromones. and These represent the reward and penalty factors for the optimal and worst paths, respectively. and These are the optimal and worst path lengths during the iteration process, respectively. As an enhancing factor, As a reward factor; Introducing dynamic volatile factors: When the initial number of iterations is lower than a preset threshold, the volatile factor takes the maximum value. This ensures that pheromones evaporate quickly in the early stages of the search process, thus avoiding premature entrapment in local optima; among which... Represents the number of iterations. Represents the maximum number of iterations. This represents the minimum value of the volatile factor; d. For path optimization problems with time windows, add time window influencing factors to improve the ant state transition rules; Time window deviation function: in Represents the delivery arrival time; the time window is... Indicates; among which Indicates customer Earliest permitted arrival time Indicates customer Latest permitted arrival time; Time window correction function: The above formula represents the ant's position at the node. by state transition probability Select Move to Node .

9. The method for optimizing UAV blood delivery routes for three-dimensional mountainous terrain as described in claim 1, characterized in that, The method further includes: Post-validation optimization: Integrate a 3D flight simulation engine to perform collision detection, endurance verification, and time window compliance checks on the algorithm's output path, mark abnormal arc segments, and trigger local replanning.

10. A drone-based blood delivery route optimization system for three-dimensional mountainous terrain, characterized in that, The system includes: The basic data acquisition module is used to collect basic parameter data and define decision variables; The cost modeling module is used to build models for transportation energy consumption costs, cooling costs, and time window penalty costs. The path optimization module is used to build an objective function that minimizes the sum of fixed costs, transportation energy costs, cooling costs, and time window penalty costs. It establishes a complex constraint system for mountainous areas and defines the domain of decision variables, thereby constructing a path optimization model adapted to the three-dimensional terrain of mountainous areas. The optimization solution module is used to solve the path optimization model adapted to the three-dimensional terrain of mountainous areas using an improved elite ant colony algorithm. It transforms the objective function into a fitness function and the constraints into path feasibility judgment rules. Through dynamic volatile factors and elite reward and punishment mechanisms, iterative optimization is performed to obtain the optimal UAV blood delivery path and corresponding parameter scheme.

Images