Cooperative distribution method for unmanned aerial vehicle and unmanned vehicle

Abstract

The invention provides a cooperative distribution method for an unmanned aerial vehicle and an unmanned vehicle. According to the method, the order data is acquired, and the order data is clustered based on the clustering algorithm to obtain the clustering result; according to a clustering result, carrying out region division through a grid method and establishing a constraint condition; defining a fitness function according to the traffic flow and the weather data, and calculating a fitness value corresponding to each order according to the fitness function; performing global search based on a particle swarm optimization algorithm according to the fitness value corresponding to each order; performing local optimization based on an ant colony optimization algorithm according to the globally optimal solution; optimizing the optimization result based on the Pareto optimal solution to obtain a distribution scheme; and carrying out cooperative distribution on the unmanned aerial vehicle and the unmanned vehicle according to the distribution scheme. Compared with the prior art, the method has the advantages that by applying dynamic adjustment and adaptive optimization strategies, the task allocation efficiency and accuracy are remarkably improved, and efficient task allocation and path planning of cooperative distribution of the unmanned vehicle and the unmanned aerial vehicle are realized.

Description

Technical Field

[0001] The present invention is applicable to the field of path planning technology, and in particular relates to a collaborative delivery method for a drone and an unmanned vehicle. Background Art

[0002] With the rapid improvement and development of multi-agent technology, artificial intelligence technology and communication network technology, the application of drones and unmanned vehicles is becoming more and more extensive. For example, unmanned vehicles rely on advanced sensors (such as LIDAR, cameras) and GPS navigation systems for automatic navigation and path planning, and are widely used in logistics distribution. However, they have poor flexibility in complex urban environments, especially in the last mile of delivery, which is difficult to achieve efficient delivery due to the complexity of terrain and traffic conditions. In contrast, drones perform well in short-distance and light-weight cargo delivery. They can easily cross special terrain and reach the destination directly, greatly shortening the delivery time. However, drone delivery also has many problems, such as small load capacity and weak battery life, and can usually only deliver short distances. In addition, flights need to comply with strict aviation regulations and are easily affected by weather and environmental conditions, which limits the application of drones in long-distance delivery.

[0003] The two have broad research prospects due to their complementary advantages and have become a hot topic of research in recent years. As an emerging field, collaborative delivery technology can significantly improve delivery efficiency by combining the advantages of unmanned vehicles and drones. For example, Amazon and Google's projects are exploring the use of unmanned vehicles for long-distance transportation, and then drones to complete the last mile of fast delivery. This collaborative model can theoretically overcome their respective limitations, but in practical applications, the complexity of path planning and task scheduling is still a challenge, and the existing system still has a lot of room for improvement in this regard.

[0004] Existing drone and unmanned vehicle collaborative delivery solutions often have the following problems: 1. Inefficient task allocation. In logistics distribution and intelligent traffic management, traditional task allocation methods are often inefficient and difficult to handle large-scale data and complex task requirements. 2. Failure to consider the influence of multiple factors leads to insufficient accuracy of task allocation. In the task allocation process, it is difficult to comprehensively consider multiple actual factors such as traffic flow and weather conditions, resulting in inaccurate task allocation results, affecting the actual execution effect. 3. Insufficient global optimization and local optimization capabilities. When dealing with task allocation problems, many optimization algorithms are prone to fall into local optimal solutions and it is difficult to find global optimal solutions. 4. Insufficient dynamic adjustment and adaptive optimization. When dealing with task allocation problems, many optimization algorithms are prone to fall into local optimal solutions and it is difficult to find global optimal solutions. In the optimization process of traditional algorithms, the parameters are fixed and lack dynamic adjustment and adaptive capabilities, resulting in limited optimization effects. 5. It is difficult to handle multi-objective optimization problems. In actual task allocation, it is often necessary to optimize multiple objectives (such as task completion time, resource utilization, cost, etc.) at the same time. Traditional methods are difficult to effectively handle multi-objective optimization problems.

[0005] Therefore, there is an urgent need for a new collaborative delivery method of drones and unmanned vehicles to solve the above technical problems. Summary of the invention

[0006] The present invention proposes a collaborative delivery method for drones and unmanned vehicles, aiming to solve technical problems existing in the prior art, such as low task allocation efficiency, insufficient task allocation accuracy under the influence of multiple factors, insufficient global and local optimization capabilities, and difficulty in handling multi-objective optimization problems.

[0007] The present invention proposes a collaborative delivery method of a drone and an unmanned vehicle, comprising the following steps:

[0008] S1. Obtain order data, and cluster the order data based on a clustering algorithm to obtain a clustering result;

[0009] S2. Performing regional division and establishing constraint conditions based on the clustering results by using a grid method to obtain multiple order information;

[0010] S3, defining a fitness function according to traffic flow and weather data, and calculating a fitness value corresponding to each order information according to the fitness function;

[0011] S4, performing a global search based on a particle swarm optimization algorithm according to the fitness value corresponding to each of the order information to obtain a global optimal solution;

[0012] S5, performing local optimization based on the ant colony optimization algorithm according to the global optimal solution to obtain an optimization result;

[0013] S6. Optimizing the optimization result based on the Pareto optimal solution to obtain an allocation plan;

[0014] S7. Coordinate delivery of drones and unmanned vehicles according to the distribution plan.

[0015] Preferably, the constraints include:

[0016] The flight altitude of the drone shall not be lower than the preset minimum flight altitude, nor higher than the preset limit altitude;

[0017] There must be no obstacles in the drone’s flight path;

[0018] The climbing angle of the drone during flight must be less than the preset maximum climbing angle;

[0019] The turning angle of the drone during flight cannot exceed the preset maximum turning angle;

[0020] The flight distance that the drone returns to the starting point after completing the search mission must be less than the drone's preset maximum flight distance.

[0021] Preferably, the fitness function satisfies the following relationship:

[0022] K = e1 × G + e2 × W + e3 × C;

[0023] Among them, K represents the fitness value, G represents the influencing factor of the traffic flow, W represents the influencing factor of the weather data, e1, e2 and e3 represent weight coefficients, and C represents the delivery cost.

[0024] Preferably, the influencing factor of the traffic flow satisfies the following relationship:

[0025]

[0026] Where V represents the current traffic speed of the road, V max represents the maximum allowed speed of the current road, YI represents the traffic congestion index, and Q1 and Q2 represent traffic weight coefficients.

[0027] Preferably, the influencing factor of the weather data satisfies the following relationship:

[0028] W=1+α(β1R+β2S+β3F s );

[0029] Among them, R represents rainfall, S represents snowfall, and F s represents wind speed, α represents weather influence coefficient, and β1β2β3 represents weather factor weight coefficient.

[0030] Preferably, the delivery cost satisfies the following relationship:

[0031]

[0032] Among them, C r represents the fuel cost per unit distance, C t represents the labor cost per unit time, L represents the delivery time, and T represents the total time required to complete the delivery.

[0033] Preferably, step S4 includes the following sub-steps:

[0034] S41, initializing the position and velocity of the particle swarm;

[0035] S42, adjusting the position and speed of the particle swarm according to the sinusoidal adaptive inertia weight formula and the exponential adaptive strategy;

[0036] S43, iteratively updating the speed and position of each particle in the particle swarm according to the adaptive function to obtain the global optimal solution.

[0037] Preferably, the sinusoidal adaptive inertia weight formula satisfies the following relationship:

[0038]

[0039] Among them, Z max and Z min They represent the maximum and minimum values ​​of the inertia weight respectively, t represents the current number of iterations, and T represents the maximum number of iterations.

[0040] Preferably, step S5 includes the following sub-steps:

[0041] S51, initializing the ant colony and pheromone according to the global optimal solution;

[0042] S52, establishing a heuristic function, wherein the heuristic function is used to calculate the distance weight between the optional grid and the end grid;

[0043] S53, initializing the pheromone concentration according to the directional information between the starting point and the end point;

[0044] S54, ants in the ant colony select a path according to the pheromone concentration and the heuristic function to obtain an ant path;

[0045] S55, updating the pheromone concentration, iteratively optimizing the ant path, and obtaining the optimization result.

[0046] Compared with the prior art, the present invention obtains order data, clusters the order data based on a clustering algorithm, and obtains clustering results; divides the area according to the clustering results by a grid method and establishes constraints to obtain multiple orders; defines a fitness function according to traffic flow and weather data, and calculates the fitness value corresponding to each order according to the fitness function; performs a global search based on a particle swarm optimization algorithm according to the fitness value corresponding to each order to obtain a global optimal solution; performs local optimization based on an ant colony optimization algorithm according to the global optimal solution to obtain an optimization result; optimizes the optimization result based on a Pareto optimal solution to obtain an allocation scheme; and performs collaborative delivery of drones and unmanned vehicles according to the allocation scheme. The present invention significantly improves the efficiency and accuracy of task allocation by applying dynamic adjustment and adaptive optimization strategies, and realizes efficient task allocation and path planning for collaborative delivery of unmanned vehicles and drones. The present invention can realize efficient task allocation in a complex urban environment by combining K-means clustering, a grid method, an improved particle swarm optimization (PSO) algorithm, and an ant colony optimization (ACO) algorithm. The efficiency and accuracy of distribution are improved, and the problems of low efficiency and insufficient accuracy of task allocation in existing methods are effectively addressed. At the same time, the present invention pays attention to environmental protection. Through relatively accurate flight path planning and task scheduling, the power consumption of drones during transportation is minimized. This helps to reduce the negative impact on the environment and improve the sustainability of logistics distribution. The present invention introduces dynamic adjustment and adaptive optimization strategies, which enable the system to flexibly adjust task allocation and path planning according to real-time traffic, weather and other environmental changes. This flexibility enhances the system's adaptability to complex environments and effectively improves the operating efficiency and response speed of the overall system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The present invention will be described in detail below in conjunction with the accompanying drawings. The above and other aspects of the present invention will become clearer and easier to understand through the detailed description made in conjunction with the following drawings. In the accompanying drawings:

[0048] Figure 1 It is a flowchart of the collaborative delivery method of a drone and an unmanned vehicle provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] Please refer to Figure 1 The present invention proposes a collaborative delivery method of a drone and an unmanned vehicle, comprising the following steps:

[0051] S1. Obtain order data, and cluster the order data based on a clustering algorithm to obtain a clustering result;

[0052] In the embodiment of the present invention, the clustering algorithm is a K-means clustering algorithm.

[0053] S2. Performing regional division and establishing constraint conditions based on the clustering results by using a grid method to obtain multiple order information;

[0054] In the embodiment of the present invention, the constraint conditions include:

[0055] The flight altitude of the drone shall not be lower than the preset minimum flight altitude, nor higher than the preset limit altitude;

[0056] There must be no obstacles in the drone’s flight path;

[0057] The climbing angle of the drone during flight must be less than the preset maximum climbing angle;

[0058] The turning angle of the drone during flight cannot exceed the preset maximum turning angle;

[0059] The flight distance that the drone returns to the starting point after completing the search mission must be less than the drone's preset maximum flight distance.

[0060] S3, defining a fitness function according to traffic flow and weather data, and calculating a fitness value corresponding to each order information according to the fitness function;

[0061] In this embodiment of the present invention, the fitness function satisfies the following relationship:

[0062] K = e1 × G + e2 × W + e3 × C;

[0063] Among them, K represents the fitness value, G represents the influencing factor of the traffic flow, W represents the influencing factor of the weather data, e1, e2 and e3 represent weight coefficients for reflecting the relative influence of traffic flow and weather on fitness, and C represents the delivery cost.

[0064] The influencing factor of the traffic flow satisfies the following relationship:

[0065]

[0066] Where V represents the current traffic speed of the road, V max It represents the maximum allowable speed of the current road. YI represents the traffic congestion index, which can usually be obtained through the traffic monitoring system. The larger the value, the higher the congestion level. Q1 and Q2 represent traffic weight coefficients, which are used to balance the impact of speed and congestion index.

[0067] The influencing factors of the weather data satisfy the following relationship:

[0068] W=1+α(β1R+β2S+β3F s );

[0069] Among them, R represents rainfall (unit: mm), S represents snowfall (unit: mm), F s represents wind speed (unit: m / s), α represents weather influence coefficient, which is used to control the influence of overall weather factors on fitness, and β1β2β3 represents weather factor weight coefficient.

[0070] The delivery cost satisfies the following relationship:

[0071]

[0072] Among them, C r represents the fuel cost per unit distance, C t represents the labor cost per unit time, L represents the delivery time, and T represents the total time required to complete the delivery.

[0073] S4, performing a global search based on a particle swarm optimization algorithm according to the fitness value corresponding to each of the order information to obtain a global optimal solution;

[0074] In this embodiment of the present invention, step S4 includes the following sub-steps:

[0075] S41, initializing the position and velocity of the particle swarm;

[0076] S42, adjusting the position and speed of the particle swarm according to the sinusoidal adaptive inertia weight formula and the exponential adaptive strategy;

[0077] S43, iteratively updating the speed and position of each particle in the particle swarm according to the adaptive function to obtain the global optimal solution.

[0078] The sinusoidal adaptive inertia weight formula satisfies the following relationship:

[0079]

[0080] Among them, Z max and Z min They represent the maximum and minimum values ​​of the inertia weight respectively, t represents the current number of iterations, and T represents the maximum number of iterations.

[0081] Specifically, the particle swarm position and velocity are initialized, and the fitness function is used. Among them, position: randomly distributed in the solution space, the solution of each particle is a vector, which represents its position in the solution space. Speed: The velocity vector is also randomly distributed within a reasonable range and does not exceed the speed limit.

[0082] Dynamically adjust particle speed and position according to the sinusoidal adaptive inertia weight formula and EATS (Exponential Adaptive Term Strategy).

[0083] The speed adjustment method is as follows:

[0084] V i (t+1)=W(t)×V i (t)+c1×r1(p i -x i (t))+c2×r2(gx i (t));

[0085] c1 and c2 are individual and global learning factors; r1 and r2 are random numbers; p i is the best historical position of the particle; g is the global best position; x i (t) and V i (t) are the position and velocity of the particle respectively.

[0086] Update particle speed and position, calculate fitness value, and find the global optimal solution.

[0087] Update position: x i (t+1)=x i (t)+V i (t+1);

[0088] If the current fitness value of a particle is better than its historical best value, the individual best position is updated.

[0089] If the current fitness value of the particle is better than the global optimal value, the global optimal position is updated.

[0090] S5, performing local optimization based on the ant colony optimization algorithm according to the global optimal solution to obtain an optimization result;

[0091] In this embodiment of the present invention, step S5 includes the following sub-steps:

[0092] S51. Initialize the ant colony and pheromone according to the global optimal solution; specifically, set the pheromone concentration of the initial paths to a higher value to emphasize the superiority of these paths, and set a lower pheromone concentration for other unexplored paths.

[0093] S52, establishing a heuristic function, wherein the heuristic function is used to calculate the distance weight between the optional grid and the end grid; specifically, the distance weight between the optional grid and the end grid is defined as:

[0094]

[0095] σij is the heuristic factor from grid i to grid j.

[0096] d ij is the distance from grid i to grid j.

[0097] S53. Initialize the pheromone concentration according to the directional information between the starting point and the end point; specifically, the pheromone concentration is slightly higher in the grid near the straight path; and the pheromone concentration is slightly lower in the grid far from the straight path.

[0098] S54: The ants in the ant colony select a path according to the pheromone concentration and the heuristic function to obtain an ant path.

[0099] S55, updating the pheromone concentration, iteratively optimizing the ant path, and obtaining the optimization result. Specifically, the ant path satisfies the following relationship:

[0100]

[0101] P ij is the probability of an ant choosing from grid i to grid j, is the pheromone concentration from grid i to grid j, σ ij is the heuristic factor from grid i to grid j, α,ρ are weight coefficients used to balance the influence of pheromone concentration and heuristic factor,

[0102] Which updates the pheromone concentration:

[0103]

[0104] θ is the pheromone volatilization coefficient, which controls the volatilization speed of pheromone. is the pheromone increase, defined as:

[0105]

[0106] Q is a constant representing the pheromone intensity, L k is the length of the path found by the kth ant.

[0107] In each iteration, the ants select a path based on the current pheromone concentration and heuristic factor, and gradually optimize the path through repeated iterations until the predetermined number of iterations or convergence criteria is reached.

[0108] S6. Optimize the optimization result based on the Pareto optimal solution to obtain an allocation plan.

[0109] Specifically, define a multi-objective optimization problem and form a solution set. Use improved PSO (particle swarm optimization) and ACO (ant colony optimization) to generate a set of candidate solutions. These solutions represent different task allocation schemes. For each candidate solution, calculate all objective function values. That is, for minimizing the delivery time T, minimizing the delivery cost C, and maximizing the fitness function F (comprehensively considering the influence of traffic flow and weather conditions), each solution will have a corresponding objective function value vector:

[0110] F(x)=[T(x i ),C(x i ),F(x i )];

[0111] Find the Pareto frontier through the Pareto optimal solution method.

[0112] Set the weight W i For each target f i (delivery time, delivery cost, fitness function), calculate the comprehensive objective function value F(x) = ∑ i W i ×f i (x).

[0113] Select the solution with the smallest comprehensive objective function value, and choose the final task allocation plan on the Pareto frontier to achieve balance and optimization among multiple objectives.

[0114] S7. Coordinate delivery of drones and unmanned vehicles according to the distribution plan.

[0115] Compared with the prior art, the present invention obtains order data, clusters the order data based on a clustering algorithm, and obtains clustering results; divides the area according to the clustering results by a grid method and establishes constraints to obtain multiple orders; defines a fitness function according to traffic flow and weather data, and calculates the fitness value corresponding to each order according to the fitness function; performs a global search based on a particle swarm optimization algorithm according to the fitness value corresponding to each order to obtain a global optimal solution; performs local optimization based on an ant colony optimization algorithm according to the global optimal solution to obtain an optimization result; optimizes the optimization result based on a Pareto optimal solution to obtain an allocation scheme; and performs collaborative delivery of drones and unmanned vehicles according to the allocation scheme. The present invention significantly improves the efficiency and accuracy of task allocation by applying dynamic adjustment and adaptive optimization strategies, and realizes efficient task allocation and path planning for collaborative delivery of unmanned vehicles and drones. The present invention can realize efficient task allocation in a complex urban environment by combining K-means clustering, a grid method, an improved particle swarm optimization (PSO) algorithm, and an ant colony optimization (ACO) algorithm. The efficiency and accuracy of distribution are improved, and the problems of low efficiency and insufficient accuracy of task allocation in existing methods are effectively addressed. At the same time, the present invention pays attention to environmental protection. Through relatively accurate flight path planning and task scheduling, the power consumption of drones during transportation is minimized. This helps to reduce the negative impact on the environment and improve the sustainability of logistics distribution. The present invention introduces dynamic adjustment and adaptive optimization strategies, which enable the system to flexibly adjust task allocation and path planning according to real-time traffic, weather and other environmental changes. This flexibility enhances the system's adaptability to complex environments and effectively improves the operating efficiency and response speed of the overall system.

[0116] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0117] The embodiments of the present invention are described above in conjunction with the accompanying drawings. What is disclosed is only the preferred embodiment of the present invention. However, the present invention is not limited to the above-mentioned specific implementation manner. The above-mentioned specific implementation manner is only illustrative rather than restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms and equivalent changes without departing from the scope of protection of the purpose of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A collaborative delivery method for unmanned aerial vehicles and unmanned vehicles, characterized in that: The following steps are involved: S1. Obtain order data, and cluster the order data based on a clustering algorithm to obtain a clustering result; S2. Performing regional division and establishing constraint conditions based on the clustering results by using a grid method to obtain multiple order information; S3, defining a fitness function according to traffic flow and weather data, and calculating a fitness value corresponding to each order information according to the fitness function; S4, performing a global search based on a particle swarm optimization algorithm according to the fitness value corresponding to each of the order information to obtain a global optimal solution; S5, performing local optimization based on the ant colony optimization algorithm according to the global optimal solution to obtain an optimization result; S6. Optimizing the optimization result based on the Pareto optimal solution to obtain an allocation plan; S7. Coordinate delivery of drones and unmanned vehicles according to the distribution plan.

2. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 1, characterized in that: The constraints include: The flight altitude of the drone shall not be lower than the preset minimum flight altitude, nor higher than the preset limit altitude; There must be no obstacles in the drone’s flight path; The climbing angle of the drone during flight must be less than the preset maximum climbing angle; The turning angle of the drone during flight cannot exceed the preset maximum turning angle; The flight distance that the drone returns to the starting point after completing the search mission must be less than the drone's preset maximum flight distance.

3. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 1, characterized in that: The fitness function satisfies the following relationship: K = e1 × G + e2 × W + e3 × C; Among them, K represents the fitness value, G represents the influencing factor of the traffic flow, W represents the influencing factor of the weather data, e1, e2 and e3 represent weight coefficients, and C represents the delivery cost.

4. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 3, characterized in that: The influencing factor of the traffic flow satisfies the following relationship: Where V represents the current traffic speed of the road, V max represents the maximum allowed speed of the current road, YI represents the traffic congestion index, and Q1 and Q2 represent traffic weight coefficients.

5. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 3, characterized in that: The influencing factors of the weather data satisfy the following relationship: W=1+α(β1R+β2S+β3F s ); Among them, R represents rainfall, S represents snowfall, and F s represents wind speed, α represents weather influence coefficient, and β1β2β3 represents weather factor weight coefficient.

6. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 3, characterized in that: The delivery cost satisfies the following relationship: Among them, C r represents the fuel cost per unit distance, C t represents the labor cost per unit time, L represents the delivery time, and T represents the total time required to complete the delivery.

7. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 1, characterized in that: Step S4 includes the following sub-steps: S41, initializing the position and velocity of the particle swarm; S42, adjusting the position and speed of the particle swarm according to the sinusoidal adaptive inertia weight formula and the exponential adaptive strategy; S43, iteratively updating the speed and position of each particle in the particle swarm according to the adaptive function to obtain the global optimal solution.

8. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 7, characterized in that: The sinusoidal adaptive inertia weight formula satisfies the following relationship: Among them, Z max and Z min They represent the maximum and minimum values ​​of the inertia weight respectively, t represents the current number of iterations, and T represents the maximum number of iterations.

9. The collaborative delivery method of a drone and an unmanned vehicle as claimed in claim 1, characterized in that: Step S5 includes the following sub-steps: S51, initializing the ant colony and pheromone according to the global optimal solution; S52, establishing a heuristic function, wherein the heuristic function is used to calculate the distance weight between the optional grid and the end grid; S53, initializing the pheromone concentration according to the directional information between the starting point and the end point; S54, ants in the ant colony select a path according to the pheromone concentration and the heuristic function to obtain an ant path; S55, updating the pheromone concentration, iteratively optimizing the ant path, and obtaining the optimization result.

Images