Skip to Main content Skip to Navigation
Theses

Lifelong Exploratory Navigation : integrating planning, navigation and SLAM for autonomous mobile robots with finite resources

Fabrice Mayran de Chamisso 1
1 LRI - Laboratoire de Robotique Interactive
DIASI - Département Intelligence Ambiante et Systèmes Interactifs : DRT/LIST/DIASI
Abstract : One of the yet unresolved canonical problems of robotics is to have robots move completely autonomously in order to accomplish any mission they are charged with, with time and resource constraints and without prior knowledge of the environment. Reaching a goal requires the robot to perform at least four tasks: maintaining an abstract representation of the environment (map), being able to localize itself within this representation, using the representation to plan paths and navigating on the planned paths while handling dynamics of the environment and avoiding obstacles. Each of these problems has been studied extensively by the robotics community. However, the four components are usually studied separately, and as a result are mostly incompatible with each other. Additionally, since humans as well as robots have to operate with finite memory and computing resources, long running planning, navigation and SLAM algorithms may have to operate on incomplete or compressed data while guaranteeing that the goal(s) can still be reached. In this thesis, planning, navigation and SLAM in arbitrarily large environments with finite computing resources and memory are considered as one single problem, for a new bio-inspired paradigm which we call Lifelong Exploratory Navigation.
Document type :
Theses
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-01674200
Contributor : Abes Star :  Contact
Submitted on : Tuesday, January 2, 2018 - 1:41:04 AM
Last modification on : Wednesday, October 14, 2020 - 3:59:20 AM
Long-term archiving on: : Wednesday, May 2, 2018 - 2:14:58 PM

File

73441_MAYRANDECHAMISSO_2016_di...
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01674200, version 1

Collections

Citation

Fabrice Mayran de Chamisso. Lifelong Exploratory Navigation : integrating planning, navigation and SLAM for autonomous mobile robots with finite resources. Artificial Intelligence [cs.AI]. Université Paris Saclay (COmUE), 2016. English. ⟨NNT : 2016SACLS413⟩. ⟨tel-01674200⟩

Share

Metrics

Record views

933

Files downloads

302