Zobrazeno 1 - 10
of 368
pro vyhledávání: '"Roy, Nicholas A"'
Autor:
Cai, Xiaoyi, Queeney, James, Xu, Tong, Datar, Aniket, Pan, Chenhui, Miller, Max, Flather, Ashton, Osteen, Philip R., Roy, Nicholas, Xiao, Xuesu, How, Jonathan P.
Self-supervised learning is a powerful approach for developing traversability models for off-road navigation, but these models often struggle with inputs unseen during training. Existing methods utilize techniques like evidential deep learning to qua
Externí odkaz:
http://arxiv.org/abs/2409.03005
Autor:
Noseworthy, Michael, Tang, Bingjie, Wen, Bowen, Handa, Ankur, Roy, Nicholas, Fox, Dieter, Ramos, Fabio, Narang, Yashraj, Akinola, Iretiayo
We present FORGE, a method that enables sim-to-real transfer of contact-rich manipulation policies in the presence of significant pose uncertainty. FORGE combines a force threshold mechanism with a dynamics randomization scheme during policy learning
Externí odkaz:
http://arxiv.org/abs/2408.04587
When using sampling-based motion planners, such as PRMs, in configuration spaces, it is difficult to determine how many samples are required for the PRM to find a solution consistently. This is relevant in Task and Motion Planning (TAMP), where many
Externí odkaz:
http://arxiv.org/abs/2407.17394
Autor:
Kurtz, Martina Stadler, Prentice, Samuel, Veys, Yasmin, Quang, Long, Nieto-Granda, Carlos, Novitzky, Michael, Stump, Ethan, Roy, Nicholas
We would like to enable a collaborative multiagent team to navigate at long length scales and under uncertainty in real-world environments. In practice, planning complexity scales with the number of agents in the team, with the length scale of the en
Externí odkaz:
http://arxiv.org/abs/2404.17438
Recent work in the construction of 3D scene graphs has enabled mobile robots to build large-scale metric-semantic hierarchical representations of the world. These detailed models contain information that is useful for planning, however an open questi
Externí odkaz:
http://arxiv.org/abs/2403.08094
Publikováno v:
EMNLP 2024 Main (The 2024 Conference on Empirical Methods on Natural Language Processing )
Prompt optimization aims to find the best prompt to a large language model (LLM) for a given task. LLMs have been successfully used to help find and improve prompt candidates for single-step tasks. However, realistic tasks for agents are multi-step a
Externí odkaz:
http://arxiv.org/abs/2402.08702
Reinforcement learning has been demonstrated to outperform even the best humans in complex domains like video games. However, running reinforcement learning experiments on the required scale for autonomous driving is extremely difficult. Building a l
Externí odkaz:
http://arxiv.org/abs/2312.15122
Autor:
Cai, Xiaoyi, Ancha, Siddharth, Sharma, Lakshay, Osteen, Philip R., Bucher, Bernadette, Phillips, Stephen, Wang, Jiuguang, Everett, Michael, Roy, Nicholas, How, Jonathan P.
Traversing terrain with good traction is crucial for achieving fast off-road navigation. Instead of manually designing costs based on terrain features, existing methods learn terrain properties directly from data via self-supervision to automatically
Externí odkaz:
http://arxiv.org/abs/2311.06234
Autor:
Pronovost, Ethan, Ganesina, Meghana Reddy, Hendy, Noureldin, Wang, Zeyu, Morales, Andres, Wang, Kai, Roy, Nicholas
Automated creation of synthetic traffic scenarios is a key part of validating the safety of autonomous vehicles (AVs). In this paper, we propose Scenario Diffusion, a novel diffusion-based architecture for generating traffic scenarios that enables co
Externí odkaz:
http://arxiv.org/abs/2311.02738
Autor:
So, Oswin, Serlin, Zachary, Mann, Makai, Gonzales, Jake, Rutledge, Kwesi, Roy, Nicholas, Fan, Chuchu
Control barrier functions (CBF) have become popular as a safety filter to guarantee the safety of nonlinear dynamical systems for arbitrary inputs. However, it is difficult to construct functions that satisfy the CBF constraints for high relative deg
Externí odkaz:
http://arxiv.org/abs/2310.15478