Výsledky vyhledávání - "Khodeir, Mohamed"

Report

Policy-Guided Lazy Search with Feedback for Task and Motion Planning

Autor: Khodeir, Mohamed, Sonwane, Atharv, Hari, Ruthrash, Shkurti, Florian

Publikováno v: 2023 IEEE International Conference on Robotics and Automation (ICRA), London, United Kingdom, 2023, pp. 3743-3749

PDDLStream solvers have recently emerged as viable solutions for Task and Motion Planning (TAMP) problems, extending PDDL to problems with continuous action spaces. Prior work has shown how PDDLStream problems can be reduced to a sequence of PDDL pla

Externí odkaz: http://arxiv.org/abs/2210.14055

Zobrazit plný text záznamu

Report

TASKOGRAPHY: Evaluating robot task planning over large 3D scene graphs

Autor: Agia, Christopher, Jatavallabhula, Krishna Murthy, Khodeir, Mohamed, Miksik, Ondrej, Vineet, Vibhav, Mukadam, Mustafa, Paull, Liam, Shkurti, Florian

Publikováno v: PMLR 164 (2022) 46-58

3D scene graphs (3DSGs) are an emerging description; unifying symbolic, topological, and metric scene representations. However, typical 3DSGs contain hundreds of objects and symbols even for small environments; rendering task planning on the full gra

Externí odkaz: http://arxiv.org/abs/2207.05006

Zobrazit plný text záznamu

Report

Learning to Search in Task and Motion Planning with Streams

Autor: Khodeir, Mohamed, Agro, Ben, Shkurti, Florian

Publikováno v: IEEE Robotics and Automation Letters, vol. 8, no. 4, pp. 1983-1990

Task and motion planning problems in robotics combine symbolic planning over discrete task variables with motion optimization over continuous state and action variables. Recent works such as PDDLStream have focused on optimistic planning with an incr

Externí odkaz: http://arxiv.org/abs/2111.13144

Zobrazit plný text záznamu