Percorrer por autor "Ferreira, Diogo Joaquim Santos"
A mostrar 1 - 1 de 1
Resultados por página
Opções de ordenação
- Design of a Collision Avoidance System for Small Fixed-wing UAVsPublication . Ferreira, Diogo Joaquim Santos; Bousson, KouamanaThis dissertation addresses the development and evaluation of a conflict detection and resolution algorithm inspired by the principles of the Traffic Collision Avoidance System (TCAS) — with the reference to TCAS concerning specifically its conflict detection logic and separation thresholds — and tailored to the operation of Unmanned Aerial Vehicles (UAVs). The proposed approach employs Model Predictive Control (MPC) to generate evasive trajectories that preserve flight safety and efficiency while ensuring conflict-free operation. To properly evaluate the algorithm, the simulation environment was designed so that conflict situations are always present. Specifically, two aircraft are initially assigned colliding trajectories, deliberately creating a reference conflict. In addition, three extra aircraft are randomly generated within a predefined range that reflects the maximum distance at which inter-aircraft information exchange is considered feasible. This ensures that the evasive trajectory is shaped not only by the conflicting pair but also by the surrounding simulated traffic. The stochastic placement of additional aircraft guarantees diversity across scenarios, enabling a more comprehensive and rigorous validation of the algorithm. The results show that the algorithm successfully modifies the evasive aircraft’s trajectory to avoid conflicts while accounting for nearby traffic to prevent the creation of new ones. Although computational cost remains a relevant limitation, the study confirms the feasibility of applying MPC-based strategies to UAV conflict resolution in three-dimensional scenarios. Overall, the work establishes a solid foundation for future research, including extensions to multi-aircraft encounters, integration of more realistic environmental conditions, and improvements in computational efficiency.