Projecte llegit
Títol: Incorporació d'objectius temporals en la planificació de trajectòries de drons
Estudiants que han llegit aquest projecte:
PANISELLO MORENO, RUT (data lectura: 30-10-2026)- Cerca aquest projecte a Bibliotècnica
PANISELLO MORENO, RUT (data lectura: 30-10-2026)Director/a: PASTOR LLORENS, ENRIC
Departament: DAC
Títol: Incorporació d'objectius temporals en la planificació de trajectòries de drons
Data inici oferta: 04-02-2026 Data finalització oferta: 04-10-2026
Estudis d'assignació del projecte:
GR ENG SIS TELECOMUN
GR ENG SIST AEROESP
GR ENG TELEMÀTICA
| Tipus: Individual | |
| Lloc de realització: EETAC | |
| Paraules clau: | |
| U-space, drones, U-plan, time objectives | |
| Descripció del contingut i pla d'activitats: | |
| Drone operations in U-space are planned via the negotiation of U-plans. One key aspect is the time-bounded expectation associated with U-plans, typically derived from flight predictions.
However, drone flight plans currently lack the possibility to specify such time-bounded expectancies. The project will explore methods to add such capacity and evaluate the positive impact on U-plan predictions. |
|
| Overview (resum en anglès): | |
| The aim of this final degree project is to develop a temporal supervision model for 4D trajectories of unmanned aircraft systems (UAS), capable of detecting deviations from planned timings and determining which corrective actions can be applied to maintain the temporal conformance of a mission.
In the context of U-space operations, 4D trajectories incorporate the temporal dimension into the spatial representation of a flight, since it is important to know not only the position a vehicle will occupy, but also when it will do so. However, an analysis of systems such as PX4, ArduPilot and DJI shows that, although they provide specific timing functions, they do not necessarily include global supervision of planned timings or automatic correction of deviations. To address this issue, a temporal supervision layer is proposed as a complement to the autopilot. It compares the estimated temporal evolution of the vehicle with the planned reference. The model establishes criteria for detecting and classifying delays and advances, defining tolerance margins and determining whether deviations can be recovered, taking into account the vehicle's kinematic limitations and the remaining trajectory. When corrective action is required, the supervisor modifies the target speed without replacing the autopilot's control mechanisms. The proposed model is implemented in a Java-based simulation environment, which allows different temporal recovery strategies to be compared across five trajectories with different characteristics and timing deviations. The evaluation includes execution without correction and three strategies: recovery distributed across segments, aggressive recovery using the maximum allowable speed, and recovery limited to the current segment. The results show that corrective strategies reduce timing deviations, although their effectiveness depends on the characteristics of the trajectory and the time available to recover the delay. The aggressive strategy achieves the lowest mean absolute timing errors, but it can lead to arrivals earlier than planned. In contrast, the strategy limited to the current segment offers the best overall balance between temporal recovery and compliance with planned timings. Distributed recovery produces less favourable results for the trajectories studied, particularly when speed constraints make the planned corrections difficult to apply. In conclusion, the simulations demonstrate that introducing a temporal supervision layer can improve compliance with planned timings in 4D trajectories. Future work includes extending the evaluation to more trajectories, studying the energy consumption associated with corrective actions and developing a supervisor capable of selecting the most appropriate strategy for each situation. |
|