A Multi-objective Optimization Method For Efficient Data Transmission By UAV-assisted Automated Guided Vehicle Collaboration

Automated Guided Vehicle(AGV)has the advantages of intelligence and high mobility,and has a broad prospect in the field of data collection and wireless transmission.To improve the transmission range of AGVs,Unmanned Aerial Vehicles(UAVs)can act as aerial Base Stations to receive data from AGVs by virtue of their mobility and flexibility.In order to transmit data to a distant UAV Base Station,this paper uses AGV to form an AGV-based Virtual Antenna Array(AVAA)for efficient data transmission with UAVs.In order to enhance the transmission performance and efficiency of the considered system,three objectives of maximizing the total transmission rate of AGVs,minimizing the total formation time of AGVs,and minimizing the total motion energy consumption of AGVs are proposed to construct a multi-objective optimization problem for efficient data transmission between AGV and UAV,and the corresponding optimization algorithms are proposed.(1)Based on the Line-of-Sight(LoS)channel,a multi-objective optimization problem for AGV-UAV communication based on LoS channel is constructed by simultaneously optimizing the position of AGVs,the weight of excitation currents,the movement speed,and the communication order with different UAVs,using UAVs as flying base stations with known positions and stationary positions(Based on Line-ofSight Channel AGV-UAV Communication Multi-objective Optimization Problem,LAUCMOP)and prove that the problem is NP-hard.An improved Multi-objective Ant Lion Optimization Algorithm(EMOALO)is proposed for LAUCMOP,which introduces Chebyshev chaotic mapping and a backward learning strategy OppositionBased Learning(OBL)for solution initialization;and Lévy flight for continuous solution updating;Partially-matched Crossover(PMX)and Exchange Mutation(EM)for discrete solution updating.Simulation experiments are conducted at different array sizes,and the results show that the proposed EMOALO algorithm has optimal performance in solving the LAUCMOP.(2)In the scenario of probabilistic LoS channel,the AGV-UAV efficient data transmission multi-objective optimization problem(Based on Probability Line-of-Sight Channel AGV-UAV Communication Multi-objective,PLAUCMOP)is constructed using UAVs as stationary and known position flying base stations.By simultaneously optimizing the position of AGVs,the weight of excitation currents,the moving speed and the communication sequence with different UAVs to achieve the multi-objective solution of the three objectives proposed in this paper.Further,an improved Multiobjective Dragonfly Algorithm(EMODA)algorithm is proposed,which introduces Tent chaotic mapping for solution initialization based on the traditional algorithm;uses a nonlinear method to change the inertia weights;and introduces an order crossover operator(Order Crossover,OX)and Simple Inversion Mutation(SIM)are introduced for the update of discrete solutions.The simulation experiments verify that the EMODA algorithm has the optimal performance in solving the PLAUCMOP at different array element sizes.