Robotic Coverage and Exploration as Sequential Decision-Making Problems

Nassim Kaldé 1, 2
Abstract : The ability to intelligently navigate in an unknown environment is essential for mobile robots (Obstacle Avoidance (OA)). This is needed to explore and build a map of the environment (Active Mapping (AM)); this map will then support other tasks such as patrolling (Active Coverage (AC)). In this thesis, we focus on decision-making to plan the moves of autonomous robots in order to navigate, cover, or explore the environment. Therefore, we rely on the framework of Sequential Decision-Making (SDM) in Artificial Intelligence to propose two contributions that address: (1) decision processes for AC and AM and (2) long-term planning for AC. Furthermore, mobile robots recently started sharing physical spaces with humans to provide services such as cleaning the house. In such cases, robot behavior should adapt to dynamic aspects of the world. In this thesis, we are interested in deploying autonomous robots in such environments. Therefore, we propose two other contributions that address: (3) short-term AM in crowded environments and (4) clearest path OA in ambient environments
Complete list of metadatas
Contributor : Abes Star <>
Submitted on : Tuesday, February 6, 2018 - 9:48:10 AM
Last modification on : Tuesday, December 18, 2018 - 4:40:22 PM
Long-term archiving on: Saturday, May 5, 2018 - 3:16:33 AM


Version validated by the jury (STAR)


  • HAL Id : tel-01701729, version 1


Nassim Kaldé. Robotic Coverage and Exploration as Sequential Decision-Making Problems. Robotics [cs.RO]. Université de Lorraine, 2017. English. ⟨NNT : 2017LORR0276⟩. ⟨tel-01701729⟩



Record views


Files downloads