Material delivery task planning method based on multi-objective optimization
Through the genetic algorithm based on multi-objective optimization, the number and sequence of the drone cluster is planned, and the problem of range and load capacity limitation in the drone material delivery mission is solved, the material delivery efficiency and resource scheduling efficiency are improved, and the efficiency of resource scheduling is realized. Reasonable task planning and efficient material delivery.
Patent Information
- Application Number
- CN202411527450.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-30
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-10-30
AI Technical Summary
Existing drones are limited by their range and load capacity in material delivery missions, which cannot meet the needs of large-scale material delivery, and the task planning efficiency is inefficient.
Using a multi-objective optimization method, the number and sequence of the drone group is planned through genetic algorithm, and the load capacity, resource consumption and resource scheduling are comprehensively considered, and the drone is represented by 4-bit binary encoding. Number of delivery times, build the initial population and iteratively evolve, and optimize the delivery order and number of times.
The material delivery efficiency and resource utilization rate of the drone group have been improved, reasonable task planning has been realized, operating procedures have been simplified, and delivery efficiency and reliability have been improved.
Smart Images

Figure CN119067403A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of unmanned aerial vehicle mission planning, and in particular to a material delivery mission planning method based on multi-objective optimization. Background Art
[0002] In recent years, drones have played an increasingly important role in the field of material delivery. Their main advantages are:
[0003] (1) Rapid response: Drones can be quickly launched and deployed, reducing the time required for material delivery. In an emergency, drones can quickly reach their destination and provide much-needed supplies.
[0004] (2) Flexibility: Drones can fly in a variety of terrains and environments, including mountains, jungles, and waters. They can bypass obstacles and choose the best path to ensure that supplies can be accurately delivered to their destinations.
[0005] (3) Lower cost: Compared with traditional material delivery methods, drones have lower operating costs. Drones do not require human piloting, nor do they require large amounts of fuel and maintenance costs. This makes drones an economical and efficient way to deliver materials.
[0006] (4) Efficiency: Drones can complete multiple delivery missions in a short period of time, improving the efficiency of material delivery. They can make multiple deliveries at different locations, reducing the need for human resources and increasing the speed of material distribution.
[0007] (5) Traceability: Drones are equipped with advanced navigation and positioning systems that can track the location and status of materials in real time.
[0008] However, the following factors need to be considered when carrying out material delivery missions based on drones:
[0009] Range and load capacity limitations: The range and load capacity of drones are limiting factors for material delivery. Currently, most commercial drones have a short range and limited load capacity, which cannot meet the needs of large-scale material delivery. Therefore, while drones bring convenience to material delivery tasks, it is also necessary to solve the task planning problem of drone swarms during material delivery tasks to improve the efficiency of drone material delivery. Summary of the invention
[0010] In view of this, the present invention proposes a material delivery task planning method based on multi-objective optimization. The method comprehensively considers the load capacity, resource consumption and overall resource scheduling of the drone group, and the method is simple, effective and reasonable.
[0011] In order to achieve the above object, the technical solution adopted by the present invention is as follows:
[0012] A material delivery task planning method based on multi-objective optimization is applied to a drone swarm for delivering materials, and plans the delivery times and delivery sequence of each drone in the drone swarm; the method comprises the following steps:
[0013] Step 1: Obtain the material delivery task requirements: including the total weight of the materials , transportation distance and maximum delivery time specified in the mission ;
[0014] Step 2: Obtain the performance parameters of the drone swarm: including the single delivery volume of the drone swarm V=[ V 1 , V 2 ,…, V N ] , UAV swarm single delivery consumes resources L =[ l 1 , l 2 ,…, l N ] , UAV swarm single delivery time consumption T =[ t 1 , t 2 ,…, t N ] , the maintenance time of a single drone delivery and the maximum number of drone deliveries ; Indicates The single delivery volume of a drone, Indicates The resources consumed by a single delivery of a drone within the required transportation distance for the material delivery mission are Indicates The time it takes for a drone to deliver a single material at the required transportation distance. , N is the number of drones;
[0015] Step 3: Set up the drone delivery sequence planning scheme;
[0016] Step 4: Based on the material delivery mission requirements and the performance parameters of the drone swarm, set the drone delivery frequency constraints;
[0017] Step 5, construct the initial UAV delivery planning scheme population, set the crossover rate, mutation rate, maximum number of iterations and profit function;
[0018] Step 6, iteratively evolving the population of UAV delivery frequency planning schemes based on a genetic algorithm;
[0019] Step 7: Combine the UAV delivery frequency planning scheme output by the genetic algorithm with the UAV delivery sequence planning scheme in step 3 to complete the final material delivery mission planning.
[0020] Further, for the first The resources consumed by a single delivery of a drone within the required transportation distance for the material delivery mission , and its specific calculation method is:
[0021] ;
[0022] in, Indicates The energy consumption of a single delivery of a UAV at the required transportation distance for the material delivery mission, Indicates The maintenance cost of a single delivery of a drone at the required transportation distance for the material delivery mission, Indicates The lifespan of a drone in a single delivery at the required transportation distance for the material delivery mission is consumed. and They are and The super parameter of and The value range is 0 to 1.
[0023] Furthermore, the drone delivery sequence planning scheme in step 3 is specifically as follows:
[0024] The time consumed by a single delivery of the drone under the required transportation distance for the material delivery mission and the maintenance time of the drone for a single delivery are summed to obtain the drone delivery interval time; a queue to be delivered is constructed. If a drone has not completed its planned number of deliveries and the delivery interval time has been reached, it will be added to the queue to be delivered. The drone with the largest single delivery volume will be selected from the queue to be delivered for delivery preparation, and the queue to be delivered will be updated to select the next drone until the material delivery mission is completed.
[0025] Furthermore, the constraints on the number of drone delivery times in step 4 are specifically:
[0026] ;
[0027] ;
[0028] ;
[0029] in, For the The number of drone launches, for The minimum value in ; The time it takes for the drone swarm to complete the material delivery mission according to the drone takeoff sequence planning plan in step 3.
[0030] Furthermore, the specific method of constructing the population of initial drone delivery frequency planning schemes in step 5 is:
[0031] Step 501, setting the population size P;
[0032] Step 502: randomly generate n×N binary codes, wherein the n-bit binary codes are grouped into a group, and the decimal number corresponding to the n-bit binary code in the i-th group represents the i-th group. The number of drone delivery times, the value of n satisfies: ;
[0033] Step 503, if the n×N-bit binary code meets the UAV delivery frequency constraint, add it as a population individual to the initial UAV delivery frequency planning scheme population, otherwise discard the current n×N-bit binary code; repeat steps 502 to 503 until the initial UAV delivery frequency planning scheme population contains P population individuals.
[0034] Furthermore, the profit function in step 5 is specifically:
[0035] ;
[0036] Among them, f is the profit value of a certain drone delivery frequency planning scheme, λ is the hyperparameter of T, for The hyperparameters, λ and The value range is 0 to 1.
[0037] Furthermore, the specific method of step 6 is:
[0038] Step 601, according to the profit function, calculate the profit value of each individual in the current UAV delivery frequency planning scheme population, and obtain the individual with the largest profit value;
[0039] Step 602, according to the crossover rate, crossover operations are performed on the individuals of other populations and the individuals of the population with the largest benefit value;
[0040] Step 603, performing mutation operation on all individuals in the population according to the mutation rate;
[0041] Step 604, determine whether each population individual after crossover mutation meets the drone delivery frequency constraint condition, if so, retain the population individual after crossover mutation, if not, retain the population individual before crossover mutation, and obtain the evolved drone delivery frequency planning scheme population;
[0042] Step 605, determine whether the current number of iterations has reached the maximum number of iterations. If so, calculate the benefit value of each individual in the current UAV delivery planning population based on the benefit function, and use the individual with the largest benefit value as the UAV delivery planning output by the genetic algorithm; otherwise, enter the next round of iteration process.
[0043] Due to the adoption of the above technical solution, the beneficial effects of the present invention compared with the prior art are:
[0044] 1. The present invention proposes a material delivery task planning method based on multi-objective optimization. The method comprehensively considers the load capacity, resource consumption and overall resource scheduling of the drone group. The method is simple, effective and reasonable.
[0045] 2. The present invention solves the material delivery task based on the genetic algorithm and uses a 4-bit binary code to represent the number of delivery times of a UAV. The method is simple, effective and reasonable.
[0046] 3. The present invention provides a reasonable UAV delivery sequence planning, which is sorted according to the UAV's single delivery volume priority, task completion, and delivery interval time rules. The rules are simple and effective. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 The present invention provides an overall flow chart of a material delivery task planning method based on multi-objective optimization in an embodiment of the present invention.
[0048] Figure 2 Schematic diagram of gene coding in an embodiment of the present invention. DETAILED DESCRIPTION
[0049] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.
[0050] A material delivery task planning method based on multi-objective optimization is applied to a swarm of drones that deliver materials, and plans the delivery times and delivery sequence of each drone in the swarm; Figure 1 As shown, the following steps are included:
[0051] Step 1: Obtain the material delivery task requirements: including the total weight of the materials , transportation distance and maximum delivery time specified in the mission ;
[0052] Step 2: Obtain the performance parameters of the drone swarm: including the single delivery volume of the drone swarm V=[ V 1 , V 2 ,…, V N ] , UAV swarm single delivery consumes resources L =[ l 1 , l 2 ,…, l N ] , UAV swarm single delivery time consumption T =[ t 1 , t 2 ,…, t N ] , the maintenance time of a single drone delivery and the maximum number of drone deliveries ; Indicates The single delivery volume of a drone, Indicates The resources consumed by a single delivery of a drone within the required transportation distance for the material delivery mission are Indicates The time it takes for a drone to deliver a single material at the required transportation distance. , N is the number of drones;
[0053] Step 3: Set up the drone delivery sequence planning scheme;
[0054] Step 4: Based on the material delivery mission requirements and the performance parameters of the drone swarm, set the drone delivery frequency constraints;
[0055] Step 5, construct the initial UAV delivery planning scheme population, set the crossover rate, mutation rate, maximum number of iterations and profit function;
[0056] Step 6, iteratively evolving the population of UAV delivery frequency planning schemes based on a genetic algorithm;
[0057] Step 7: Combine the UAV delivery frequency planning scheme output by the genetic algorithm with the UAV delivery sequence planning scheme in step 3 to complete the final material delivery mission planning.
[0058] Further, for the first The resources consumed by a single delivery of a drone within the required transportation distance for the material delivery mission , and its specific calculation method is:
[0059] ;
[0060] in, Indicates The energy consumption of a single delivery of a UAV at the required transportation distance for the material delivery mission, Indicates The maintenance cost of a single delivery of a drone at the required transportation distance for the material delivery mission, Indicates The lifespan of a drone in a single delivery at the required transportation distance for the material delivery mission is consumed. and They are and The super parameter of and The value range is 0 to 1.
[0061] Furthermore, the drone delivery sequence planning scheme in step 3 is specifically as follows:
[0062] The time consumed by a single delivery of the drone under the required transportation distance for the material delivery mission and the maintenance time of the drone for a single delivery are summed to obtain the drone delivery interval time; a queue to be delivered is constructed. If a drone has not completed its planned number of deliveries and the delivery interval time has been reached, it will be added to the queue to be delivered. The drone with the largest single delivery volume will be selected from the queue to be delivered for delivery preparation, and the queue to be delivered will be updated to select the next drone until the material delivery mission is completed.
[0063] Furthermore, the constraints on the number of drone delivery times in step 4 are specifically:
[0064] ;
[0065] ;
[0066] ;
[0067] in, For the The number of drone launches, for The minimum value in ; The time it takes for the drone swarm to complete the material delivery mission according to the drone takeoff sequence planning plan in step 3.
[0068] Furthermore, the specific method of constructing the population of initial drone delivery frequency planning schemes in step 5 is:
[0069] Step 501, setting the population size P;
[0070] Step 502: randomly generate n×N binary codes, wherein the n-bit binary codes are grouped into a group, and the decimal number corresponding to the n-bit binary code in the i-th group represents the i-th group. The number of drone delivery times, the value of n satisfies: ;
[0071] Specifically, Figure 2 As shown, in this embodiment =15, using 4-bit binary code to represent the number of times a drone is delivered. , , ,......, They correspond to the 1st, 2nd, 3rd, ..., Nth drones respectively;
[0072] Step 503, if the n×N-bit binary code meets the UAV delivery frequency constraint, add it as a population individual to the initial UAV delivery frequency planning scheme population, otherwise discard the current n×N-bit binary code; repeat steps 502 to 503 until the initial UAV delivery frequency planning scheme population contains P population individuals.
[0073] Furthermore, the profit function in step 5 is specifically:
[0074] ;
[0075] Among them, f is the profit value of a certain drone delivery frequency planning scheme, λ is the hyperparameter of T, for The hyperparameters, λ and The value range is 0 to 1.
[0076] Furthermore, the specific method of step 6 is:
[0077] Step 601, according to the profit function, calculate the profit value of each individual in the current UAV delivery frequency planning scheme population, and obtain the individual with the largest profit value;
[0078] Step 602, according to the crossover rate, crossover operations are performed on the individuals of other populations and the individuals of the population with the largest benefit value;
[0079] Step 603, performing mutation operation on all individuals in the population according to the mutation rate;
[0080] Step 604, determine whether each population individual after crossover mutation meets the drone delivery frequency constraint condition, if so, retain the population individual after crossover mutation, if not, retain the population individual before crossover mutation, and obtain the evolved drone delivery frequency planning scheme population;
[0081] Step 605, determine whether the current number of iterations has reached the maximum number of iterations. If so, calculate the benefit value of each individual in the current UAV delivery planning population based on the benefit function, and use the individual with the largest benefit value as the UAV delivery planning output by the genetic algorithm; otherwise, enter the next round of iteration process.
[0082] In summary, the present invention proposes a material delivery task planning method based on multi-objective optimization. The method comprehensively considers the load capacity, resource consumption and resource scheduling of the drone group as a whole, and the method is simple, effective and reasonable.
[0083] Those skilled in the art will appreciate that the embodiments described are intended to help readers understand the principles of the present invention, and should be understood that the scope of protection of the present invention is not limited to the embodiments described. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of the claims of the present invention.
Claims
1. A material delivery task planning method based on multi-objective optimization, applied to a drone swarm for delivering materials, and plans the delivery times and delivery sequence of each drone in the drone swarm; characterized in that: The following steps are involved: Step 1: Obtain the material delivery task requirements: including the total weight of the materials , transportation distance and maximum delivery time specified in the mission ; Step 2: Obtain the performance parameters of the drone swarm: including the single delivery volume of the drone swarm , UAV swarm single delivery consumes resources , UAV swarm single delivery time consumption , the maintenance time of a single drone delivery and the maximum number of drone deliveries ; Indicates The single delivery volume of a drone, Indicates The resources consumed by a single delivery of a drone within the required transportation distance for the material delivery mission are Indicates The time it takes for a drone to deliver a single material at the required transportation distance. , N is the number of drones; Step 3: Set up the drone delivery sequence planning plan; Step 4: Based on the material delivery mission requirements and the performance parameters of the drone swarm, set the drone delivery frequency constraints; Step 5, construct the initial UAV delivery planning scheme population, set the crossover rate, mutation rate, maximum number of iterations and profit function; Step 6, iteratively evolving the population of UAV delivery frequency planning schemes based on a genetic algorithm; Step 7: Combine the UAV delivery frequency planning scheme output by the genetic algorithm with the UAV delivery sequence planning scheme in step 3 to complete the final material delivery mission planning.
2. According to the method of claim 1, the material delivery task planning method based on multi-objective optimization is characterized in that: For step 2 The resources consumed by a single delivery of a drone within the required transportation distance for the material delivery mission , the specific calculation method is: ; in, Indicates The energy consumption of a single delivery of a UAV at the required transportation distance for the material delivery mission, Indicates The maintenance cost of a single delivery of a drone at the required transportation distance for the material delivery mission, Indicates The lifespan of a drone in a single delivery at the required transportation distance for the material delivery mission is consumed. and They are and The super parameter of and The value range is 0 to 1.
3. The material delivery task planning method based on multi-objective optimization according to claim 2 is characterized in that: The specific plan for the drone delivery sequence in step 3 is: The time consumed by a single delivery of the drone under the required transportation distance for the material delivery mission and the maintenance time of the drone for a single delivery are summed to obtain the drone delivery interval time; a queue to be delivered is constructed. If a drone has not completed its planned number of deliveries and the delivery interval time has been reached, it will be added to the queue to be delivered. The drone with the largest single delivery volume will be selected from the queue to be delivered for delivery preparation, and the queue to be delivered will be updated to select the next drone until the material delivery mission is completed.
4. The material delivery task planning method based on multi-objective optimization according to claim 3 is characterized in that: The specific constraints for the number of drone delivery times in step 4 are: ; ; ; in, For the The number of drone launches, for The minimum value in ; The time it takes for the drone swarm to complete the material delivery mission according to the drone takeoff sequence planning plan in step 3.
5. The material delivery task planning method based on multi-objective optimization according to claim 4 is characterized in that: The specific method of constructing the initial UAV delivery frequency planning scheme population in step 5 is: Step 501, setting the population size P; Step 502: randomly generate n×N binary codes, wherein the n-bit binary codes are grouped into a group, and the decimal number corresponding to the n-bit binary code in the i-th group represents the i-th group. The number of drone delivery times, the value of n satisfies: ; Step 503, if the n×N-bit binary code meets the UAV delivery frequency constraint, add it as a population individual to the initial UAV delivery frequency planning scheme population, otherwise discard the current n×N-bit binary code; repeat steps 502 to 503 until the initial UAV delivery frequency planning scheme population contains P population individuals.
6. A material delivery task planning method based on multi-objective optimization according to claim 5, characterized in that: The specific profit function in step 5 is: ; Among them, f is the profit value of a certain drone delivery frequency planning scheme, λ is the hyperparameter of T, for The hyperparameters, λ and The value range is 0 to 1.
7. A material delivery task planning method based on multi-objective optimization according to claim 6, characterized in that: The specific method of step 6 is: Step 601, according to the profit function, calculate the profit value of each individual in the current UAV delivery frequency planning scheme population, and obtain the individual with the largest profit value; Step 602, according to the crossover rate, crossover operations are performed on the individuals of other populations and the individuals of the population with the largest benefit value; Step 603, performing mutation operation on all individuals in the population according to the mutation rate; Step 604, determine whether each population individual after crossover mutation meets the drone delivery frequency constraint condition, if so, retain the population individual after crossover mutation, if not, retain the population individual before crossover mutation, and obtain the evolved drone delivery frequency planning scheme population; Step 605, determine whether the current number of iterations has reached the maximum number of iterations. If so, calculate the benefit value of each individual in the current UAV delivery planning population based on the benefit function, and use the individual with the largest benefit value as the UAV delivery planning output by the genetic algorithm; otherwise, enter the next round of iteration process.
Citation Information
Patent Citations
Delivery task planning method based on hierarchical improved genetic algorithm
CN114281108A
Unmanned aerial vehicle task planning method for comprehensive task
CN114326800A
Air delivery task planning method based on space-time division and conquer
CN114740880A
Unmanned aerial vehicle group emergency material delivery method and simulation system
CN115423219A
Unmanned aerial vehicle group cooperative task planning method considering carrier delivery mode
CN115755963A
Cited By
Task planning optimization method and device for emergency material supply and transportation guarantee
CN119831209A
Task planning optimization method and device for emergency material supply and transportation guarantee
CN119831209B
