Skip to Main content Skip to Navigation
Theses

Path planning on a high performance heterogeneous CPU/FPGA architecture

Résumé : Security issues are today a key-differentiator in the aviation sector. Indeed, it comes to ensure the safety of expensive equipments but above all to save human lives. In this context, it is necessary to offer an important level of autonomy to helicopters. Although some studies have been carried out in this area, the dynamic generation of a sequence of maneuvers under hard time constraints in an unknown environment still represents a major challenge for many academic and industrial working groups. AIRBUS HELICOPTERS as a leader of helicopters manufacturing, looks forward to integrate an assistance system for mission re-planning in the next generation of aircrafts.The work conducted in this PhD thesis falls within a collaboration between AIRBUS HELICOPTERS and UNIVERSITE DE VALENCIENNES ET DU HAINAUTCAMBRESIS. One of the main purposes of this work is efficient flight plan generation. Indeed, for intelligent assistant systems we need to generate a new path planning inorder to face emergency events such as an equipment failure or adverse weather conditions. The second major objective of this work is the deployment of mission planning tasks onto a high performance architecture CPU/FPGA in order to meet real-time requirements for the dynamic optimization process. In the present work, we first studied efficient flight plan generation. Indeed, we developed efficient and effective algorithms for helicopter path planning. Then, in order to obtain a real-time system, we resolved the problem of scheduling optimization on a heterogeneous architecture CPU / FPGA by proposing several scheduling methods including exact approaches and heuristics.
Document type :
Theses
Complete list of metadata

https://hal-uphf.archives-ouvertes.fr/tel-03356784
Contributor : Frédéric Pruvost Connect in order to contact the contributor
Submitted on : Tuesday, September 28, 2021 - 1:59:07 PM
Last modification on : Tuesday, October 19, 2021 - 6:38:14 PM

Identifiers

  • HAL Id : tel-03356784, version 1

Collections

Citation

Omar Souissi. Path planning on a high performance heterogeneous CPU/FPGA architecture. Computer Science [cs]. Université de Valenciennes et du Hainaut-Cambrésis, 2015. English. ⟨NNT : 2015VALE0003⟩. ⟨tel-03356784⟩

Share

Metrics

Record views

13