Inspecting the supports and wires of power transmission lines is increasingly being entrusted to drones. But if you launch not one drone but a whole group from a common base, the routes must be planned so that the total flight time is as short as possible. This problem is addressed by two new algorithms described in the International Journal of Systems, Control and Communications.
The first algorithm is intended for inspecting individual elements on the supports, such as insulators. It is based on an improved genetic algorithm. This is the name for a computational method that imitates natural selection: it gradually "crosses" and improves candidate solutions. To it were added the 2-opt technique - which improves a route by reversing individual segments of the path - and adaptive mutation rates. Mutations introduce random changes into solutions so that the algorithm does not get stuck on poor routes.
The second algorithm - also an adaptive genetic one - is needed for surveying the lines themselves. It distributes segments among the drones so as to avoid unnecessary flights.
The authors tested the work on maps of different sizes and compared it with a conventional genetic algorithm. According to their report, the new methods produced higher-quality and more efficient solutions. For element inspection the results were more stable, and when surveying lines it was possible to reduce the total flight distance and select the optimal number of drones.
It is important to understand: this refers to computer tests on maps, not field launches over real power lines. The study does not prove that the algorithms will immediately reduce costs or risk to people - these are only possible consequences that remain to be verified. In addition, the comparison was with earlier approaches, not with ready-made commercial systems.
Nevertheless, the work shows a direction: instead of one expensive fly-around - the coordinated work of several machines that divide zones and do not duplicate routes. If such algorithms prove themselves in the field, they could be useful to energy workers, who currently have to send people to heights and hard-to-reach places.