Organisation/Company: CNRS
Department: Laboratoire d'analyse et d'architecture des systèmes
Research Field: Engineering, Computer Science, Mathematics
Researcher Profile: First Stage Researcher (R1)
Country: France
Application Deadline: 2 Dec 2024 - 23:59 (UTC)
Type of Contract: Temporary
Job Status: Full-time
Hours Per Week: 35
Offer Starting Date: 1 Jan 2025
Is the job funded through the EU Research Framework Programme? Not funded by a EU programme
Is the Job related to staff position within a Research Infrastructure? No
Offer Description The thesis will be carried out within the RIS team at LAAS-CNRS as part of the scientific developments of the ANR HumFleet project.
This project focuses on multi-robot/human mixed-initiative manipulation task planning for structure assembly. Motion planning is an important and active field of research in Robotics, which has seen rapid progress in recent years, particularly due to the development of probabilistic exploration methods. The planning of object handling tasks involves the planning of collision-free robot movements and the sequencing and interdependencies of various elementary actions (e.g., picking up, setting down, and transferring an object) required to complete the task.
Work at LAAS has led to original and complementary planning techniques for solving these problems at both geometric and symbolic planning levels. This topic concerns the combination of these geometric/symbolic planning techniques in a hybrid approach based on the "Combined task and Motion Planning" (CTAMP) theme. This aims to better cope with the high complexity of manipulation task planning problems, where LAAS has recognized expertise in both geometric (motion) and symbolic (task) components.
The contributions available today in the literature do not provide satisfactory answers in a reasonable time when confronted with complex situations. The novelty here is twofold: we are considering a multi-robot system capable of carrying out multi-robot co-manipulation or co-transport tasks, and interaction with a human operator who collaborates in defining and supervising the task.
A promising avenue will be the study of the coupling of learning methods with motion and/or task planning algorithms for more efficient solutions. We will also explore refined models of interaction constraints (robot-robot or human-robot), particularly through criteria to be optimized that take into account the configuration of the human-robot system, trajectory, and dynamics of movement, along with more sophisticated cost functions derived from learning interaction constraints on the task.
#J-18808-Ljbffr