Politecnico di Torino (logo)

A genetic algorithm for a task allocation problem in an urban air mobility scenario

Cristiano Bicchieri

A genetic algorithm for a task allocation problem in an urban air mobility scenario.

Rel. Giorgio Guglieri, Stefano Primatesta, Marco Rinaldi. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Aerospaziale, 2023

PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (903kB) | Preview

In an aerial package delivery scenario carried out by multiple Unmanned Aerial Vehicles (UAVs), it is important to maximize the collaboration and the resource sharing in the fleet and to satisfy, according to the UAVs' constraints, the largest possible portion of the tasks' requirements' space. In this thesis work, a UAVs-based parcel delivery problem is formulated and a Genetic Algorithm (GA) is proposed for solving the formulated task assignment problem for a fleet of heterogeneous UAVs. GAs are a class type of algorithms inspired by the known evolutionary mechanism of populations. The objective of the proposed task allocation method is to minimize the energy consumed by the fleet for executing the delivery tasks while respecting the time window delivery constraints and the maximum payload capacity of each UAV. Each task in the task set is considered with the parcel's pick-up point and delivery point, the payload mass, and parcel's time delivery deadline. The UAVs are assumed to be able to perform only one task at a time. In order to address the service persistency issues, re-charge tasks can also be assigned and added where the UAV's energy is not sufficient to perform the next task. The initial population of the genetic algorithm is created randomly, and subsequent mutation and crossover operations are performed to increase the diversity of the population and avoid incurring in local minima. The capability of the proposed solution of efficiently handling the formulated problem is demonstrated with ad-hoc defined scenarios, which represent the algorithm's instances.

Relators: Giorgio Guglieri, Stefano Primatesta, Marco Rinaldi
Academic year: 2022/23
Publication type: Electronic
Number of Pages: 47
Corso di laurea: Corso di laurea magistrale in Ingegneria Aerospaziale
Classe di laurea: New organization > Master science > LM-20 - AEROSPATIAL AND ASTRONAUTIC ENGINEERING
Ente in cotutela: Institut Superieur de Mecanique de Paris (FRANCIA)
Aziende collaboratrici: UNSPECIFIED
URI: http://webthesis.biblio.polito.it/id/eprint/26956
Modify record (reserved for operators) Modify record (reserved for operators)