Research And Implementation Of Scheduling Optimization Technology For Uav Data Collection In Iot Environment

The classical three-layer Io T architecture includes a sensing layer,a network layer and an application layer,where the network layer needs to collect data from the sensing layer and transmit it to the upper layer applications,so complete and efficient data collection from sensing devices is the key to Io T.The traditional static-based collection mode suffers from poor scal-ability,high maintenance cost and high data loss rate,while the UAV-based mobile collection mode is widely used due to its advantages of high mobility,high flexibility,rapid deployment and no need to consider terrain factors.However,in the practical data collection scenarios,in addition to choosing the appropriate collection mode,it is also necessary to consider the impact of different types of sensing devices on the data collection.In this thesis,the sensing device that senses sparse and single data is called light-data sensing device,and the sensing device that senses dense and multiple data is called heavy-data sensing device.For the light-data sensing device,a single UAV is used to collect the data.Because of the limited energy of the UAV,its energy consumption needs to be optimized in order to improve the collection efficiency.According to the practical UAV flight energy consumption model,it is known that operations such as turning and switching make the UAV energy consumption composition intricate,and mathematical programming or machine learning solutions are not applicable to the problem,so it is necessary to design a new flight planning algorithm based on a refined energy model to optimize UAV energy consumption.For heavy-data sensing devices,the multi-UAV mode collection with the assistance of vehicle occupies an advantage,but still faces the following challenges.On the one hand,the deadline of data collection and data volume constraints need to be satisfied; on the other hand,increasing the number of UAVs and increasing the number of vehicles to reduce the number of UAVs both bring about an increase in collection cost,so the optimal scheduling strategy for the collection cost of multiple UAVs needs to be designed.In order to solve the above challenges,this thesis focuses on the problems of existing research such as the lack of refined UAV energy models,the difficulty of applying traditional algorithms and the lack of research on heavy-data collection scenarios,conducts theoretical research on light-data and heavy-data collection scenarios,respectively,and implements a UAV scheduling optimization system for data collection of Io T devices.Specific research results include:(1)Considering the overlapping transmission range of heterogeneous light-data sensing devices and the fact that UAV turning and hovering also affect energy consumption,a graph based algorithm to optimize the energy consumption of a single UAV in light-data collection is proposed,converting straight flight energy consumption,turning energy consumption and switching energy consumption into edge weights of the graph model,which is solved using the classical generalized travel salesman problem.Finally,according to simulations,the proposed algorithm achieves 93% of the optimal solution and saves about 50% energy than the naive solution.(2)Considering the time sensitivity of heavy-data collection,as well as the limited energy of UAVs and flight safety,the vehicle-assisted multi-UAV collection mode is adopted.With the objective of minimizing the collection cost under collection constraints,the optimization algo-rithm for the collection cost of multi-UAVs in heavy-data collection is proposed.The algorithm is first based on dynamic programming to solve a special instance of the original problem,and then extend the solution to the general problem.Finally,according to simulations,the average performance of the proposed algorithm outperforms the greedy algorithm by about 46%.(3)Based on the light/heavy data collection scheduling algorithm proposed in Chapter 3and Chapter 4 of this thesis,we design and implement a UAV scheduling optimization system for data collection of Io T sensing devices,and demonstrate the system operation process and effect to verify the performance of the data collection scheduling algorithm.In summary,this thesis investigates the scheduling optimization technology of UAV data collection in Io T environment,and verifies the correctness and efficiency of the proposed data collection optimization algorithms through a series of experiments and system implementations,to support the practical applications of Io T with the collected complete and efficient data.