Computer Design Competition National Award Works_2. Application Materials

This series is a related document for the work "Environmental Monitoring UAV Route Optimization" in the 2021 China College Students Computer Design Competition, which won the first prize in the 2021 Northwest Division and the third prize in the National Competition. Student work is for your reference only.

insert image description here

1. Title of work: Route optimization of environmental monitoring drones

2. Content introduction:

2.1 Project background:

The use of air mobile monitoring equipment carried by drones to monitor air pollution in specific areas and pollution sources has the advantages of high efficiency, time saving, flexibility and concealment, and rapid locking of pollution sources. The planning of the UAV flight route needs to design a flight route to complete the assigned task under the premise of meeting the constraints of maneuverability and obstacle avoidance. The route planning of the UAV can be solved by using the simulated annealing algorithm.

2.2 Innovation points:

1) Study the characteristics and interrelations of operators in the existing simulated annealing algorithm, and find that the exchange operation is equivalent to the superposition and composition of two nested reverse order operations;
2) By calculating the intermediate results of the exchange operation, two As a result of the reverse order operation, the three new paths generated by the exchange and reverse order operations can be obtained without increasing the amount of calculation;
3) Therefore, a new exchange-reverse order joint operator simulated annealing algorithm is proposed, which does not increase The large amount of computation increases the search space of the simulated annealing algorithm by 3 times.

2.3 Software features:

1) The improved simulated annealing algorithm is used to expand the search range and improve the optimization performance and efficiency;
2) Completely independently developed the environmental monitoring UAV route optimization software, which can meet the application requirements of UAV flight route planning;
3) Development considerations Route planning function influenced by wind direction.

3. Design ideas:

The research idea of this project is to carry out research and design according to the steps of algorithm research, demand analysis, program design, and test filing:

3.1 Algorithm research:

(1) Literature research and problem analysis: study the characteristics and interrelationships of existing operators such as reverse order, shift and exchange, and analyze the specific implementation process of the algorithm;
(2) Algorithm improvement: first calculate the two nested operators separately. The path difference of the reverse order operation, after adding the two, the path difference of the exchange operation is obtained, that is, the path difference of the two reverse order operations can be obtained by calculating the intermediate result of the path difference of the exchange operation;
(3) Propose an optimization algorithm : A new exchange-reverse order joint operator simulated annealing algorithm is proposed, which can obtain 3 new paths generated by exchange operation and reverse order operation without increasing the amount of calculation, and expand the search space of the simulated annealing algorithm by 3 times. , to improve the optimization performance;
(4) Test the improved algorithm: test with different scales of Benchmark problems, the simulation results show that the performance of the joint operator is better than the existing shift, exchange, reverse order operators, and also better than these operators combination scheme.

3.2 Demand analysis:

(1) Demand research:

Due to the characteristics of air pollution, such as “large area involved, large amount of pollutant transmission between areas, many types of pollution sources, and relatively complex pollution factors”, traditional environmental monitoring methods face problems such as heavy workload, poor flexibility, and difficulty in locating pollution sources. The atmospheric mobile monitoring equipment carried by the drone is used to monitor the air pollution in specific areas and pollution sources, which has the advantages of high efficiency, time saving, flexibility and concealment, and fast locking of pollution sources.

Environmental monitoring drones can achieve high-space, large-area monitoring, and can also achieve accurate monitoring in a low-space and a small range. Through the fusion of various detection methods, environmental data of monitoring points and areas can be obtained. The products and solutions of Shenzhen Kefei Technology, Guangzhou Gesai, Beijing Future Intelligence and other companies have been put into commercial applications, providing a strong decision-making basis for environmental monitoring, emergency rescue, emergency response, smart city and scientific research.

Environmental monitoring drone products and technologies are first considered to meet environmental monitoring tasks. By carrying gas detection sensors and pan-tilt cameras, monitoring gas data, images obtained by cameras, and drone flight data are transmitted to the control center in real time. Most of the existing products do not have the function of flight path planning, and the monitoring sequence and flight path are mainly set manually. Therefore, the development of environmental monitoring UAV route optimization software, combined with constraints such as maneuverability and obstacle avoidance, can automatically plan the route, which can provide customers with a better user experience.

(2) Problem modeling and solving algorithm:

For the planning of the UAV's flight route, it is necessary to design a flight route that completes the assigned task under the premise of meeting constraints such as maneuverability and obstacle avoidance. That is, finding the Hamiltonian circuit with the smallest total weight for a given positive-weight complete graph is a typical traveling salesman problem. The traveling salesman problem is a classical combinatorial optimization problem, which belongs to NP-complete problem.
Research on the large-scale traveling salesman problem has focused on high-performance approximate optimal methods, including various heuristic search algorithms constructed based on feature information (such as position, distance, angle, etc.), and developed by simulating or explaining natural laws. Simulated annealing algorithm, genetic algorithm, ant colony algorithm, neural network algorithm and other intelligent optimization algorithms.

The simulated annealing algorithm is a common method for solving large-scale combinatorial optimization problems. Based on the aforementioned research and improvement results of the simulated annealing algorithm, the path optimization problem of the UAV flight path can be solved.

(3) Analyze software requirements:

Business requirements: Obtain map and location coordinate data of designated environmental monitoring points, obtain constraints such as UAV maneuvering performance and obstacle avoidance, and obtain optimization goals for UAV flight route planning; perform path optimization through simulated annealing algorithm, and output optimization Results output; dynamically adjust flight tasks and flight routes according to real-time monitoring results.

Functional requirements: ①Data input function, including creating new projects, importing projects, and reading data files; ②Settings, including setting optimization parameters, setting constraints, and setting optimization goals; ③Path optimization, according to the input data and conditions, use improved The simulated annealing algorithm optimizes the processing path and displays the optimization results; ④Result output; ⑤Help function.

User interface requirements: ①Following the design principles of the Graphical User Interface (GUI), the interface is intuitive and transparent to users; ②Consistency in interface design, using standard controls and unified information presentation methods; ③The main interface has a menu bar and tool list and shortcut keys to display the optimization result graph of the last run; ④Menu and sub-interfaces include: data input, parameter setting, path optimization, result output, and help.

Operating environment requirements: The software uses Python3.8 programming, and Qt develops GUI interface. It is recommended to use Intel CPU above 1.6GHz, memory above 2G, and Windows7/Windows10 operating system.

3.3 Software Design:

(1) Overall design: ①Data input module: create a new project, or import a project, or read the data of monitoring points from a file; ②Parameter setting: parameter setting of simulated annealing algorithm, optional constraint setting, optional optimization target setting ;③Path optimization: use the improved simulated annealing algorithm to optimize the path, choose other optimization methods to compare the optimization performance, display the optimization process graphically, and display the optimization results; ④Result output: output the optimized path and save it to a file; ⑤Help function.
(2) Program coding: On the basis of the overall design, according to the design requirements of data structure, algorithm analysis and module implementation, write Python programs to achieve the requirements of function, performance, interface and interface.

3.4 Software Testing:

(1) Test with Benchmark problems of different scales, and compare with the original algorithm and other optimization algorithms, mainly to test the optimization performance of the improved algorithm;

(2) Test the actual cases of typical users, mainly test the software interface and functional modules to meet the design requirements.

3.5 Delivery and Summary:

Write project documentation (development documentation, user manual), organize and file process documentation.
Deliver user trials; and collect user comments for product updates and upgrades.

insert image description here

Youcans@xupt original works, reprints must be marked with the original link: (https://blog.csdn.net/youcans/article/details/123979917)