UAV systems capable of collecting data and monitoring sensitive areas, such as urban traffic, without the need for ground infrastructure and at a low cost, have revolutionized traffic management. For intelligent traffic monitoring, selecting the optimal flight path for the UAV is crucial to ensure timely service delivery and the UAV\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\'s return to its base. This research focuses on the simultaneous optimization of the flight path length and achievable data rate in UAV-assisted traffic monitoring communication channels. First, the problem is divided into two sub-problems: path planning and achievable rate, each of which is weighted and assigned a corresponding fitness function. Then, the optimization is performed using five algorithms: ant colony optimization, particle swarm optimization, refrigeration simulation, genetic algorithm, and deep learning. The proposed algorithms are simulated in an environment with four transmitters placed at the corners of a square. Finally, a bi-objective optimization method is implemented to minimize the path length and maximize the achievable rate without dividing the problem into two parts. The algorithm and simulation results are presented for four transmitters and six obstacles. Numerical results show that for four transmitters in a square with a side length of 1000 m and six obstacles, with a boundary of 50 m, the path length is 3800 m and the total achievable rate is 4.7. It is worth noting that, based on the network needs, the path length can be reduced and the achievable rate can be increased.

کلمات کلیدی