Skip to Main content Skip to Navigation
Theses

Planification de mission pour un véhicule aérien autonome

Abstract : Autonomous vehicles follow a mission plan, sometimes updated by an operator. Decisional autonomy is necessary for long endurance missions with limited communication. This work deals with on-line replanning, illustrated by an observation mission for a UAV. The goal is to develop a mission planner integrated in an on-board architecture.

The formalism describes the objectives selection. Each objectif has a reward and a cost, functions of dates and resources. The planner chooses the best way to achieve each objective in time and space, while optimizing rewards and costs and meeting constraints.
The planning algorithm, based on the A*, and several methods of criterion evaluation, pruning and search guidance are described.
The planner is integrated in an on-board hybrid and hierarchized architecture.
36 scenario, simulated for 16 combinations of methods, test the algorithmic part. Results analysis highlights methods that best balance quality and computation time.
Document type :
Theses
Complete list of metadatas

Cited literature [63 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00174631
Contributor : Elodie Chanthery <>
Submitted on : Monday, September 24, 2007 - 4:52:43 PM
Last modification on : Monday, August 26, 2019 - 1:58:01 PM
Long-term archiving on: : Friday, April 9, 2010 - 2:46:03 AM

Identifiers

  • HAL Id : tel-00174631, version 1

Collections

Citation

Elodie Chanthery. Planification de mission pour un véhicule aérien autonome. Automatique / Robotique. Ecole nationale superieure de l'aeronautique et de l'espace, 2005. Français. ⟨tel-00174631⟩

Share

Metrics

Record views

514

Files downloads

1071