CN107657412B - Unmanned aerial vehicle and automobile combined distribution system and distribution method for remote areas - Google Patents

Abstract

The invention discloses a combined distribution method of unmanned aerial vehicle and automobile for remote areas, which comprises the following steps: 1) the automobile carries the package and the unmanned aerial vehicle loaded with the package to the starting point of the delivery area, and the delivery area includes Multiple drone delivery points and car delivery points; 2) The drone takes off from the starting point or midway and delivers the package to the designated drone delivery point and then returns to the end of the delivery area, where each drone delivery point is at most Delivered by a UAV; 3) After all the UAVs return to the voyage and load them into the car, the distribution is completed. The present invention considers the geographic distribution of logistics distribution points in remote areas, classifies logistics distribution points, and performs logistics distribution by drones and cars respectively, optimizes the flight routes of drones and cars, and delivers packages simultaneously by drones and cars , can greatly reduce the risk of logistics transportation in remote areas, reduce logistics costs and transportation time.

Description

Unmanned aerial vehicle and automobile combined delivery system and delivery method for remote areas

technical field

The invention relates to the technical field of intelligent express delivery, in particular to a remote-area-oriented combined delivery system and delivery method for drones and cars.

Background technique

With the development of UAV technology, the reliability and carrying capacity of UAVs have been greatly improved, and the cost of purchasing and using UAVs has dropped rapidly. At present, UAVs have been widely used in the civil field of our country. In the express delivery industry, UAVs have been used for short-distance parcel delivery in cities, but there are some obvious constraints in this type of business: First, UAV parcel delivery in cities is faced with the responsibility of urban buildings, tree distribution , UAV collision avoidance and safety issues are a big challenge; in addition, the country has issued a series of regulations on the low-altitude flight of UAVs. high" question.

In the vast remote areas of my country, the level of economic development is relatively low, and the construction of road transportation infrastructure is lagging behind. The parcel express business in these areas is facing several problems: First, the parcel express business volume is not high, and the logistics cost is too high if the parcel express business is not high. Second, these areas are often blocked by mountains and rivers, and the roads are circuitous and in poor condition, so it is difficult to guarantee the safety of parcel delivery by car. In remote areas with a relatively small population, the low-altitude flight control of drones is relatively loose, and there are few problems with collision avoidance and flight safety. Therefore, this technical invention comprehensively uses drones, cars and parcel receiving boxes to carry out logistics distribution in remote areas.

Chinese patent document CN201620108719.1 proposes an unmanned aerial vehicle express system, which consists of unmanned aerial vehicles and express cabinets distributed in various outlets. There is a device for fixing the landing gear of the unmanned aerial vehicle on the top of the express cabinet. There is a directional mark, and an image recognition system is installed under the drone to recognize the directional mark and control it to land in a preset direction. Chinese patent document CN201510240751.5 provides a method and system for UAV cargo transportation. The method sets a landing recognition pattern on the ground, and the UAV flies to the sky to automatically perform image recognition search. If the search is successful, the UAV lands. Carry out the shipment of the package. Chinese patent document CN201520244456.2 provides an automatic receiving system for unmanned express delivery, which includes a receiving trolley, a lifting platform and a receiving window. The above-mentioned patent documents focus on the composition of the UAV express delivery system, without considering the distribution of logistics distribution points in remote areas. Chinese patent document CN201520606406.4 proposes an intelligent express delivery system. After the logistics vehicle arrives at the stop, the package is manually fixed on the drone. The drone flies according to the optimized flight route, puts the package into the intelligent cabinet, and notifies Customer picks up. The logistics vehicle in the above-mentioned patent document is only parked at the stop and is in a static state. It does not deliver packages in parallel with the drone, and does not involve how to specifically optimize the flight route of the drone. This is not conducive to reducing logistics delivery costs and shortening logistics delivery. In addition, this patent document does not consider the distribution of logistics distribution points in remote areas.

Contents of the invention

The purpose of the present invention is to aim at the technical defects existing in the prior art, and provide a kind of unmanned aerial vehicle and automobile combined delivery system and delivery method for remote areas.

The technical scheme adopted for realizing the purpose of the present invention is:

A combined distribution method for unmanned aerial vehicles and vehicles for remote areas, comprising the following steps:

1) The car carries the package and the drone loaded with the package to the starting point of the delivery area, which includes multiple drone delivery points and car delivery points;

2) The car drives to the end of the delivery area according to the car delivery route, and at the same time traverses the said car delivery point and delivers. According to the drone delivery strategy, the drone takes off from the starting point or midway and delivers the package to the designated drone delivery point and return to the end of the delivery area, where each drone delivery point is delivered by at most one drone;

3) After all the drones return to the voyage and load them into the car, the distribution is completed.

The drone delivery point mentioned above is a delivery point that requires circuitous transportation or a delivery road with a grade of four or below.

The UAV is provided with a ejection device to assist take-off or direct vertical take-off and landing.

The parcel is provided with an RFID tag, and the drone delivery point and the car delivery point respectively include a parcel receiving box, an RFID reader and a wireless communication module. After the parcel is placed in the parcel receiving box, the RFID reader reads Package information, and notify the user of the arrival.

Send a package arrival notification to the user's mobile phone through 3G or 4G wireless communication technology.

The calculation method of the car delivery route is:

1) Combining the start point, end point and car delivery point into a set; the car needs to traverse each point in the set once;

2) The car delivery route problem is transformed into a traveling salesman problem, using a heuristic algorithm to find the delivery route with the shortest distance;

3) Delete the connection path between the specified start point and end point, and the remaining path is the car delivery path.

The calculation method of the UAV distribution strategy is:

Constrained by the number of drones and the maximum flight distance, each drone delivery point is delivered by only one drone, and the drone takes off from the starting point or the car delivery point to deliver to the corresponding drone delivery point , and then return to the end of the delivery area and land.

The UAV distribution strategy is solved based on the multi-objective optimization algorithm of decomposition and elite strategy. The characteristics of this algorithm are:

1) Decompose the optimization goal into two sub-problems, one is to use the least number of UAVs, and the other is the shortest delivery path of UAVs;

2) Generate a flight path population with more than 100 paths of the UAV, and the weight vector of each flight path is 2 real numbers greater than 0 and less than 1; and the corresponding neighborhood flight path, the number of neighborhood paths is ≥ 5; the flight path The population is set with an objective function reference point;

3) Based on the flight path of the UAV, with the maximum flight distance of the UAV as the constraint condition, the flight path is divided into sub-paths to ensure that the length of each sub-path does not exceed the maximum flight distance of the UAV. This can obtain the number of UAV distribution points contained in the sub-path and the number of sub-paths; generate the corresponding objective function value of each flight path; use the flight path with the smallest objective function value as the elite path, and update the objective function reference point;

4) Calculate the Chebyshev value of each flight path in the neighborhood where the elite path is located, each flight path corresponds to a real weight value and an objective function value, then its corresponding Chebyshev value=max{real weight value*(objective function value - the value of the reference point of the objective function)}; when the Chebyshev value of the elite path ≤ the Chebyshev value of the neighborhood flight path, the elite path is used to replace other flight paths in the neighborhood, and the UAV flight path population is updated to realize the elite Strategy;

5) Set the crossover and mutation probability, the crossover probability is set to 0.7-0.9, the mutation probability is set to 0.1-0.15, and the class crossover operation and reverse sequence mutation operation are performed on all flight paths to improve the diversity of flight paths;

6) Calculate the objective function value of each flight path of the flight path population after crossover and mutation, find out the flight path with the minimum objective function value as the elite path, and update the objective function reference point;

7) Calculate the Chebyshev value of each flight path in the neighborhood where the elite path is located in step 6), and when the Chebyshev value of the elite path ≤ the Chebyshev value of the neighborhood flight path, use the elite path instead of the neighborhood flight path, update the UAV flight path population, and realize the elite strategy;

8) Return to step 5), perform loop iterations, calculate the flight path of the best UAV, and obtain the number of UAVs and the delivery trajectory of each UAV.

In the optimization model of UAV distribution strategy, the number of logistics distribution points is n, the designated starting point and end point are recorded as 0 and n+1 respectively, the number of available UAVs is m, and D _ij is the distribution path (i,j ), F _u is the maximum flight distance of the u-th UAV; the optimization model is:

1)

2)

3)

4)

5)

6)

7)

8)

9) x _uij = {0,1}

Among them, the formula (1) is the objective function 1, the number of UAVs used is the least; the formula (2) is the objective function 2, the delivery path of the UAV is the shortest; take off; the meaning of formula (4) is that the UAV lands at the designated terminal; the meaning of formula (5) is that for any drone, the delivery distance does not exceed its maximum flight distance; the meaning of formula (6) The number of aircraft arriving at the destination does not exceed the number of existing drones; the meaning of formula (7) is that for any logistics distribution point, at most one drone arrives, and the meaning of formula (8) is that for any logistics distribution point Point, at most one UAV leaves; formula (9) is a decision variable, when the uth UAV passes through the delivery path (i, j), x _uij takes the value of 1.

A combined distribution system of unmanned aerial vehicles and automobiles for remote areas, including automobiles, unmanned aerial vehicles for carrying parcels, and a distribution area containing multiple distribution points. The distribution area includes automobile distribution points and unmanned vehicles. An RFID tag is set on the package. The delivery point includes a package receiving box, an RFID reader and a wireless communication module. After the package is placed in the package receiving box, the RFID reader reads the package information and passes the wireless communication module Arrival notification is issued to the user, characterized in that an apron is provided at the end of the delivery area.

Compared with prior art, the beneficial effect of the present invention is:

The present invention considers the geographic distribution of logistics distribution points in remote areas, classifies logistics distribution points, and performs logistics distribution by drones and cars respectively, optimizes the flight routes of drones and cars, and delivers packages simultaneously by drones and cars , can greatly reduce the risk of logistics transportation in remote areas, reduce logistics costs and transportation time.

Description of drawings

Figure 1 is a schematic diagram of the structure of the combined delivery system of drones and cars for remote areas.

Figure 2 is a schematic diagram of the distribution of logistics distribution points.

Figure 3 is a schematic diagram of the route of parallel delivery of parcels by car and UAV.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

The combined distribution system of unmanned aerial vehicle and automobile for remote areas of the present invention includes automobile 1, unmanned aerial vehicle 2 for carrying parcels, and a distribution area comprising a plurality of distribution points, and the distribution area includes automobile distribution points and At the drone delivery point, the parcel 3 is provided with an RFID tag 31. The delivery point includes a parcel receiving box 4, an RFID reader and a wireless communication module. After the parcel is placed in the parcel receiving box, the RFID reader reads the parcel information and notify the user of the arrival. Wherein, an apron is provided at the end of the delivery area. The wireless communication module is a 3G or 4G wireless communication module to send a package arrival notification to the mobile phone 5 of the user.

The lower part of the drone is equipped with a multi-position warehouse and a package ejection device. The distribution drone and the ejection unloading mechanism of the goods are similar to the prior art, and will not be described here.

The invention combines automobiles and unmanned aerial vehicles to realize classified delivery in a certain area, with high speed, and reduces transportation costs and transportation risks caused by problems such as poor road conditions or excessive detours caused by mountains and rivers in remote mountainous areas. The delivery area mentioned above has a diameter of 10-20km. By dividing the areas reasonably, an organic combination can be realized to ensure delivery efficiency and safety. Human-machine charging station. The use of sub-area and time-division enhances the total effective endurance of the UAV.

The remote area-oriented UAV and automobile combined distribution method of the present invention comprises the following steps:

1) The car carries the package and the drone loaded with the package to the starting point of the delivery area, which includes multiple drone delivery points and car delivery points;

2) The car drives to the end of the delivery area according to the delivery route of the car, and at the same time traverses the said car delivery point and delivers. The drone takes off from the starting point or midway and delivers the package to the designated drone delivery point and then returns to the end of the delivery area , where each drone delivery point has at most one drone delivery;

3) After all the drones return and load them into the car, the delivery is completed.

Remote areas are often blocked by rivers and mountains, and some roads have poor traffic conditions. If the traditional car is used to deliver parcels, there are certain hidden dangers in transportation safety. At the same time, circuitous transportation is required. Logistics costs and time have increased significantly. Therefore, remote areas with rivers, mountains, and poor road traffic conditions, such as logistics distribution points with a distribution road level of level 4 and below, are allocated to drones for distribution, and the remaining distribution points are assigned to The car makes the delivery. Set a designated starting point in the logistics distribution area. The setting conditions for the starting point include: 1. The starting point is in an open and flat area, which is conducive to the parking of vehicles and the take-off and landing of drones; 2. Within a radius of 10-20 kilometers from the starting point Within the scope, there are multiple, such as nearly ten or more than a dozen, more fixed logistics distribution points; 3. There is a horizontal field for drone landing at the starting point, and the area of the horizontal field is not less than 10 square meters. The starting point can be planned and constructed in advance by the logistics company in a certain distribution area according to the above principles.

Multiple drones and a car are at the same starting point, which is conducive to simultaneously loading packages for multiple drones inside the car, reducing the time for loading goods multiple times and in batches. At the same time, it is also conducive to determining the delivery of packages in advance The number and location information of points can be used to optimize the delivery route of drones in advance to reduce the flight time of drones. In layman's terms, advance planning, unified loading, unified distribution, cost savings, and UAVs fly according to the established trajectory, reducing intermediate communication links, avoiding loss of communication signals, and reducing communication links.

The drone is provided with a ejection device or the vehicle is provided with a ejection device to assist take-off. Of course, it can also be lifted vertically. The package is provided with an RFID tag. The distribution point of the drone includes a package Receiving box, RFID reader and wireless communication module. After the drone or car driver puts the package into the package receiving box, the RFID reader reads the package information and sends an arrival notification to the user, such as through 3G or 4G wireless communication technology Send a package arrival notification to the user's mobile phone.

Wherein, the calculation method of the car delivery route is:

1) Combining the start point, end point and car delivery point into a set; the car needs to traverse each point in the set once;

2) The car delivery route problem is transformed into a Traveling Salesman Problem, using a heuristic algorithm to find the delivery route with the shortest distance;

3) Delete the connection path between the specified start point and end point, and the remaining path is the car delivery path.

Among them, the calculation method of the UAV delivery strategy is:

1) Decompose the optimization goal into two sub-problems, one is to use the least number of UAVs, and the other is the shortest delivery path of UAVs;

2) Generate a flight path population with more than 100 paths of the UAV, and the weight vector of each flight path is 2 real numbers greater than 0 and less than 1; and the corresponding neighborhood flight path, the number of neighborhood paths is ≥ 5; the flight path The population is set with an objective function reference point, that is, the initial state of the objective function reference point of each flight path is the same;

For example, the objective function reference point is set to (4,60), where 4 represents 4 unmanned aircraft, and 60 represents a delivery route of 60km. The function of this value is to provide a reference for subsequent objective function value updates. The two real numbers The sum is 1, corresponding to the two goals respectively, that is, the minimum number of drones and the shortest delivery path.

3) Based on the flight path of the UAV, with the maximum flight distance of the UAV as the constraint condition, the flight path is divided into sub-paths to ensure that the length of each sub-path does not exceed the maximum flight distance of the UAV. This can obtain the number of UAV distribution points contained in the sub-path and the number of sub-paths; generate the corresponding objective function value of each flight path; select the flight path with the smallest objective function value as the elite path, and update the objective function reference point;

After one of the flight paths is divided, the number of its sub-paths can determine the number of unmanned aircraft used, and then, according to the length of each sub-path, the cumulative can determine the cruising distance of the unmanned aircraft. The number of UAVs used and the cruising distance of UAVs constitute the value of the objective function. For example, assuming that the starting point and the ending point are A and B respectively, the UAV delivery point is regarded as a set, and the UAV departs from the starting point and returns to the end point. If the number of UAV delivery points is 7, an integer of 1-7 will be generated Random number sequence, generating flight path populations that traverse the 7 points, such as 1-2-3-4-5-6-7, 4-3-2-1-5-6-7, 6-3-2-1 -5-4-7, etc., where the number represents the number of the delivery point of the drone, and the integer random number sequence is randomly generated 100 times, and 100 flight paths are obtained, and these paths form a population. A real number is a number greater than 0 and less than 1, and the sum of two real numbers is 1, such as 0.35 and 0.65, 0.55 and 0.45; the neighborhood refers to a flight path (such as the aforementioned 6-3-2-1- 5-4-7) Other adjacent flight paths, such as the 23rd, 47th, and 78th flight paths in the population, the neighborhood paths are used for crossover and mutation operations in subsequent steps to improve the optimization ability of the algorithm.

For example, for the flight path 6-3-2-1-5-4-7, first analyze whether the distance of the sub-path A-6-B is greater than the maximum flight distance of the unmanned aircraft, if not, then analyze the sub-path Is A-6-3-B greater than the maximum flight distance of the unmanned aircraft, if not greater, then analyze whether the sub-path A-6-3-2-B is greater than the maximum flight distance of the unmanned aircraft, if greater, then The sub-path A-6-3-B is the first path; then, based on the remaining path 2-1-5-4-7, continue to divide the sub-paths according to the above method until all origins and destinations or delivery The points are divided. For example, the final division results are A-6-3-B, A-2-1-5-4-B and A-7-B, then there are 3 unmanned sub-paths, and the corresponding objective function value of each path is calculated; According to the geographical coordinates of each drone delivery point and the starting and ending points, the Euclidean distance calculation between points can be performed to determine the length of each sub-path. Assuming that the lengths of the three sub-paths are 10km, 12km, and 15km respectively, the delivery distance of the UAV corresponding to the flight path is 10+12+15=37km. Then the objective function value of the flight path is (3,37).

4) Calculate the Chebyshev value of each flight path in the neighborhood where the elite path is located, each flight path corresponds to a real weight value and an objective function value, then its corresponding Chebyshev value=max{real weight value*(objective function value - the value of the reference point of the objective function)}; when the Chebyshev value of the elite path ≤ the Chebyshev value of the neighborhood flight path, the elite path is used to replace other flight paths in the neighborhood, and the UAV flight path population is updated to realize the elite Strategy;

5) Set the crossover and mutation probability, the crossover probability is set to 0.7-0.9, the mutation probability is set to 0.1-0.15, and the class crossover operation and reverse sequence mutation operation are performed on all flight paths to improve the diversity of flight paths;

The purpose of crossover mutation is to disturb the flight path, improve the diversity of flight paths, so as to find the optimal flight path; if the flight path changes, the Chebyshev value will change accordingly.

6) Calculate the objective function value of each flight path of the flight path population after crossover and mutation, find out the flight path with the minimum objective function value as the elite path, and update the objective function reference point;

7) Calculate the Chebyshev value of each flight path in the neighborhood where the elite path is located in step 6), and when the Chebyshev value of the elite path ≤ the Chebyshev value of the neighborhood flight path, use the elite path instead of the neighborhood flight path, update the UAV flight path population, and realize the elite strategy;

8) Return to step 5), perform loop iterations, calculate the flight path of the best UAV, and obtain the number of UAVs and the delivery trajectory of each UAV. The termination condition is the maximum number of iterations, generally set to 300 -500 times.

At the same time, for different UAVs and different trajectories, in order to achieve the final best execution, UAVs include multiple UAVs with different carrying capacities for package distribution.

Specifically, in the UAV distribution strategy, the number of logistics distribution points is n, the designated starting point and end point are recorded as 0 and n+1 respectively, the number of available UAVs is m, and D _ij is the distribution path (i, j ), F _u is the maximum flight distance of the u-th UAV; the optimization model is:

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9) x _uij = {0,1}

Among them, the formula (1) is the objective function 1, the number of UAVs used is the least; the formula (2) is the objective function 2, the delivery path of the UAV is the shortest; take off; the meaning of formula (4) is that the UAV lands at the designated terminal; the meaning of formula (5) is that for any drone, the delivery distance does not exceed its maximum flight distance; the meaning of formula (6) The number of aircraft arriving at the destination does not exceed the number of existing drones; the meaning of formula (7) is that for any logistics distribution point, at most one drone arrives, and the meaning of formula (8) is that for any logistics distribution point Point, at most one UAV leaves; formula (9) is a decision variable, when the uth UAV passes through the delivery path (i, j), x _uij takes the value of 1.

With reference to the accompanying drawings, an exemplary description is given in conjunction with specific embodiments:

1) Classification of logistics distribution points. The logistics distribution points in remote areas that are blocked by rivers and mountains and have poor road traffic conditions are allocated to drones for distribution, and the remaining distribution points are allocated to cars for distribution. As shown in Figure 2, logistics distribution points 6, 7, 10, and 11 are assigned to cars, and logistics distribution points 1, 2, 3, 4, 5, 8, and 9 are assigned to drones.

2) Optimization of vehicle distribution routes.

①Set the specified starting point (A in Figure 2) and end point (B in Figure 2) of the car, and form a set of delivery points, starting point, and end point, and the car needs to traverse each point in the set once;

②The car delivery route problem is transformed into the Traveling Salesman Problem (Traveling Salesman Problem), and the heuristic algorithm is used to solve the delivery route with the shortest distance, as shown in A-11-7-10-6-B-A in Figure 3;

③ On the basis of the above-mentioned optimized path, delete the connection path of the specified starting point and end point (as shown in B-A in Figure 2), and delete the remaining path (as shown in Figure 3 A-11-7-10-6-B) As an optimized vehicle distribution route.

3) UAV distribution path optimization.

① Use the designated starting point (A in Figure 2) and end point (B in Figure 2) of the car as the take-off point and landing point of the drone, respectively;

②Constrained by the number of UAVs and the maximum flight distance, each logistics distribution point is delivered by only one UAV. Delivery, and then landing at the landing point (B in Figure 2), with the least number of drones used and the shortest delivery distance of drones as the optimization goal, an optimization model for the delivery path of drones is established, and a multi-objective heuristic optimization algorithm is used , to determine the number of drones needed and the delivery route of the drones. As shown in Figure 3, there are 2 drones delivering parcels, the delivery route of the first drone is A-9-3-2-B, and the delivery route of the second drone is A-8 -5-4-1-B.

Car-drone parallel delivery of packages. At the designated starting point (A in Figure 3), the car and the UAV start parcel delivery at the same time, and the car delivers the package according to the optimized delivery route, as shown in A-11-7-10-6-B in Figure 3; 2 The UAV delivers packages according to the optimized delivery route, as shown in Figure 3 A-9-3-2-B and A-8-5-4-1-B. After the delivery task is completed, the car and the UAV respectively Return to the specified end point (B in Figure 3).

4) Package information notification. The UAV delivers parcels to logistics points 1, 2, 3, 4, 5, 8, and 9. When a new parcel is delivered into the parcel receiving box of the logistics point, its RFID reader reads the electronic label information of the parcel and passes the wireless 3G, 4G communication technology, send package arrival notification to user's mobile phone.

The above is only a preferred embodiment of the present invention, it should be pointed out that, for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, these improvements and Retouching should also be regarded as the protection scope of the present invention.

Claims (7)

1. A combined distribution method for unmanned aerial vehicles and automobiles in remote areas, characterized in that it comprises the following steps: 1) The car carries the package and the drone loaded with the package to the starting point of the delivery area, which includes multiple drone delivery points and car delivery points; 2) The car drives to the end of the delivery area according to the car delivery route, and at the same time traverses the said car delivery point and delivers. According to the drone delivery strategy, the drone takes off from the starting point or midway and delivers the package to the designated drone delivery point and return to the end of the delivery area, where each drone delivery point is delivered by at most one drone; 3) After all the unmanned aerial vehicles return to the voyage and load them into the car mentioned above, the distribution is completed; Wherein, the calculation method of the car delivery route is: 1) Combining the start point, end point and car delivery point into a set; the car needs to traverse each point in the set once; 2) The car delivery route problem is transformed into a traveling salesman problem, and the heuristic algorithm is used to find the delivery route with the shortest distance; 3) Delete the connection path between the specified start point and end point, and the remaining path is the car delivery path; Constrained by the number of drones and the maximum flight distance, each drone delivery point is delivered by only one drone, and the drone takes off from the starting point or the car delivery point to deliver to the corresponding drone delivery point , and then return to the destination of the distribution area and land. The distribution strategy of the UAV is solved based on the multi-objective optimization algorithm of decomposition and elite strategy. The characteristics of this algorithm are: 1) Decompose the optimization goal into two sub-problems, one is to use the least number of UAVs, and the other is the shortest delivery path of UAVs; 2) Generate a flight path population with more than 100 paths of the UAV, and the weight vector of each flight path is 2 real numbers greater than 0 and less than 1; and the corresponding neighborhood flight path, the number of neighborhood paths is ≥ 5; the flight path The population is set with an objective function reference point; 3) Based on the flight path of the UAV, with the maximum flight distance of the UAV as the constraint condition, the flight path is divided into sub-paths to ensure that the length of each sub-path does not exceed the maximum flight distance of the UAV. This can obtain the number of UAV distribution points contained in the sub-path and the number of sub-paths; generate the corresponding objective function value of each flight path; use the flight path with the smallest objective function value as the elite path, and update the objective function reference point; 4) Calculate the Chebyshev value of each flight path in the neighborhood where the elite path is located, each flight path corresponds to a real weight value and an objective function value, then its corresponding Chebyshev value=max{real weight value*(objective function value - the value of the reference point of the objective function)}; when the Chebyshev value of the elite path ≤ the Chebyshev value of the neighborhood flight path, the elite path is used to replace other flight paths in the neighborhood, and the UAV flight path population is updated to realize the elite Strategy; 5) Set the crossover and mutation probability, the crossover probability is set to 0.7-0.9, and the mutation probability is set to 0.1-0.15, and the class crossover operation and reverse sequence mutation operation are performed on all flight paths to improve the diversity of flight paths; 6) Calculate the objective function value of each flight path of the flight path population after crossover and mutation, find out the flight path with the minimum objective function value as the elite path, and update the objective function reference point; 7) Calculate the Chebyshev value of each flight path in the neighborhood where the elite path is located in step 6), and when the Chebyshev value of the elite path ≤ the Chebyshev value of the neighborhood flight path, use the elite path instead of the neighborhood flight path, update the UAV flight path population, and realize the elite strategy; 8) Return to step 5), perform loop iterations, calculate the flight path of the best UAV, and obtain the number of UAVs and the delivery trajectory of each UAV. 2. The combined delivery method of drones and cars for remote areas as claimed in claim 1, characterized in that, the delivery point of the drone is a delivery that requires circuitous transportation or a delivery road with a grade of four or below point. 3. The remote-area-oriented combined distribution method of drones and cars as claimed in claim 1, wherein the drone is provided with a ejection device to assist take-off or direct vertical take-off and landing. 4. The combined distribution method of unmanned aerial vehicles and automobiles for remote areas as claimed in claim 3, wherein the parcels are provided with RFID tags, and the distribution points of the unmanned aerial vehicles and the automobile distribution points are respectively It includes a package receiving box, an RFID reader and a wireless communication module. After the package is placed in the package receiving box, the RFID reader reads the package information and sends an arrival notification to the user. 5. The combined drone and car delivery method for remote areas as claimed in claim 4, wherein a package arrival notification is sent to the user's mobile phone through 3G or 4G wireless communication technology. 6. The combined distribution method for unmanned aerial vehicles and automobiles facing remote areas as claimed in claim 1, wherein in the optimization model of the unmanned aerial vehicle distribution strategy, the number of logistics distribution points is n, and the designated starting point and end point are respectively Denoted as 0 and n+1, the number of available UAVs is m, D _ij is the distance of the delivery path (i, j), F _u is the maximum flight distance of the u-th UAV; the optimization model is: 1)

2)

3)

4)

5)

6)

7)

8)

9) x _uij = {0, 1} Among them, the formula (1) is the objective function 1, the number of UAVs used is the least; the formula (2) is the objective function 2, the delivery path of the UAV is the shortest; take off; the meaning of formula (4) is that the UAV lands at the designated terminal; the meaning of formula (5) is that for any drone, the delivery distance does not exceed its maximum flight distance; the meaning of formula (6) The number of aircraft arriving at the destination does not exceed the number of existing drones; the meaning of formula (7) is that for any logistics distribution point, at most one drone arrives, and the meaning of formula (8) is that for any logistics distribution point At most one UAV leaves; formula (9) is a decision variable, when the uth UAV passes through the delivery path (i, j), x _uij takes the value of 1. 7. A delivery system for realizing the combined delivery method of unmanned aerial vehicles and cars for remote areas as described in any one of claims 1-6, characterized in that it includes cars, unmanned vehicles for carrying packages machine, and a delivery area containing multiple delivery points, the delivery area includes car delivery points and drone delivery points, and RFID tags are set on the parcels, and the delivery points include parcel receiving boxes, RFID readers and wireless The communication module, after the package is put into the package receiving box, the RFID reader reads the package information and sends an arrival notification to the user through the wireless communication module, which is characterized in that an apron is set at the end of the distribution area.

Images