41,386 Hits in 6.1 sec

A Deep Reinforcement Learning Environment for Particle Robot Navigation and Object Manipulation [article]

Jeremy Shen, Erdong Xiao, Yuchen Liu, Chen Feng
2022 arXiv   pre-print
While its control is currently limited to a hand-crafted policy for basic locomotion tasks, such a multi-robot system could be potentially controlled via Deep Reinforcement Learning (DRL) for different  ...  Further development of DRL algorithms is necessary in order to accomplish the proposed tasks.  ...  This research is supported by the NSF CPS program under CMMI-1932187.  ... 
arXiv:2203.06464v1 fatcat:hyto64ijf5fr3f3njjwbddxu2a

MESA: Offline Meta-RL for Safe Adaptation and Fault Tolerance [article]

Michael Luo, Ashwin Balakrishna, Brijen Thananjeyan, Suraj Nair, Julian Ibarz, Jie Tan, Chelsea Finn, Ion Stoica, Ken Goldberg
2021 arXiv   pre-print
Safe exploration is critical for using reinforcement learning (RL) in risk-sensitive environments.  ...  Simulation experiments across 5 continuous control domains suggest that MESA can leverage offline data from a range of different environments to reduce constraint violations in unseen environments by up  ...  Authors were also supported by the SAIL-Toyota Research initiative, the Scalable Collaborative Human-Robot Learning (SCHooL) Project, the NSF National Robotics Initiative Award 1734633, and in part by  ... 
arXiv:2112.03575v1 fatcat:uhmj46nkbbg4dmv4ou77csrhxq


Y. Han, A. Yilmaz
2021 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
In this work, we propose an approach for an autonomous agent that learns to navigate in an unknown map in a real-world environment.  ...  We demonstrate that the agent can learn navigating to the destination kilometers away from the starting point in a real world scenario and has the ability to respond to environment changes while learning  ...  learning progress in the case environment dynamically changes.  ... 
doi:10.5194/isprs-annals-v-1-2021-145-2021 fatcat:ur3qavm25bfklfqpcdaeed5gme

Deep Learning for Embodied Vision Navigation: A Survey [article]

Fengda Zhu, Yi Zhu, Vincent CS Lee, Xiaodan Liang, Xiaojun Chang
2021 arXiv   pre-print
"Embodied visual navigation" problem requires an agent to navigate in a 3D environment mainly rely on its first-person observation.  ...  The remarkable learning ability of deep learning methods empowered the agents to accomplish embodied visual navigation tasks.  ...  [128] propose to jointly learn a navigation agent by imitation learning and reinforcement learning.  ... 
arXiv:2108.04097v4 fatcat:46p2p3zlivabbn7dvowkyccufe

UAV Maneuvering Target Tracking in Uncertain Environments Based on Deep Reinforcement Learning and Meta-Learning

Bo Li, Zhigang Gan, Daqing Chen, Dyachenko Sergey Aleksandrovich
2020 Remote Sensing  
We consider a multi-task experience replay buffer to provide data for the multi-task learning of the DRL algorithm, and we combine meta-learning to develop a multi-task reinforcement learning update method  ...  unmanned aerial vehicle (UAV), allowing a UAV to quickly track a target in an environment where the motion of a target is uncertain.  ...  We combine meta-learning to propose a multi-task reinforcement learning update method to ensure the generalization of the policy.  ... 
doi:10.3390/rs12223789 fatcat:ldccvrk5ejaiffhhl43k2vygge

Socially-Aware Multi-Agent Following with 2D Laser Scans via Deep Reinforcement Learning and Potential Field [article]

Yuxiang Cui, Xiaolong Huang, Yue Wang, Rong Xiong
2021 arXiv   pre-print
The multi-agent following problem is tackled by utilizing the complementary strengths of both reinforcement learning and potential field, in which the reinforcement learning part handles local interactions  ...  Target following in dynamic pedestrian environments is an important task for mobile robots.  ...  By utilizing potential field and reinforcement learning, we can navigate a team of robots through unseen environments while following the target in multiple dynamic environments, ensuring both safety and  ... 
arXiv:2109.01874v1 fatcat:4nbqyrdqkzbtzid46bfy5aob5i

A Parallel Evolutionary Algorithm with Value Decomposition for Multi-agent Problems [chapter]

Gao Li, Qiqi Duan, Yuhui Shi
2020 Lecture Notes in Computer Science  
In order to solve multi-agent problems by EA, a value decomposition method is used to decompose the team reward.  ...  Second, most multi-agent environments only provide a shared team reward as feedback. As a result, agents may not be able to learn proper cooperative or competitive behaviors by traditional RL.  ...  We evaluated our algorithm in the Cooperative Navigation task of MAPE We compare our algorithm with RS, ES, REINFORCE, Actor-Critic, DQN, VDN in this environment.  ... 
doi:10.1007/978-3-030-53956-6_57 fatcat:gowdjtn77fdzvfejrtuuk4j2ua

Motion planning for mobile Robots–focusing on deep reinforcement learning: A systematic Review

Huihui Sun, Weijie Zhang, Runxiang YU, Yujie Zhang
2021 IEEE Access  
This paper reviews the methods based on motionplanning policy, especially the ones involving Deep Reinforcement Learning (DRL) in the unstructured environment.  ...  INDEX TERMS Mobile robot; Deep reinforcement learning; Motion planning  ...  [121] proposed a deep imitative learning method to perform navigation tasks in 3D simulation environment.  ... 
doi:10.1109/access.2021.3076530 fatcat:53kdh5cfgvcang5xymf4vrqx2e

Reinforcement learning-based group navigation approach for multiple autonomous robotic systems

O. Azouaoui, A. Cherifi, R. Bensalem, A. Farah, K. Achour
2006 Advanced Robotics  
In this paper, a reinforcement learning (RL)-based group navigation approach for multiple ARS is suggested.  ...  Then, simulation results display the ability of the suggested approach to provide ARS with capability to navigate in a group formation in dynamic environments.  ...  The group navigation problem in such a multi-robot environment is solved by interacting with it.  ... 
doi:10.1163/156855306776985531 fatcat:qpz4za6hu5hwhpjrdgbq7amx4a

A Survey on Visual Navigation for Artificial Agents with Deep Reinforcement Learning

Fanyu Zeng, Chen Wang, Shuzhi Sam Ge
2020 IEEE Access  
In this paper, we first present an overview on reinforcement learning (RL), deep learning (DL) and deep reinforcement learning (DRL).  ...  Visual navigation for artificial agents with deep reinforcement learning (DRL) is a new research hotspot in artificial intelligence and robotics that incorporates the decision making of DRL into visual  ...  Therefore, a visual DRL navigation agent with a multi-modal sense can learn a better policy [121] .  ... 
doi:10.1109/access.2020.3011438 fatcat:ie6qvu24qbapbjxtiudh7fumgy

A Survey of Embodied AI: From Simulators to Research Tasks [article]

Jiafei Duan, Samson Yu, Hui Li Tan, Hongyuan Zhu, Cheston Tan
2022 arXiv   pre-print
Instead, they learn through interactions with their environments from an egocentric perception similar to humans.  ...  Lastly, this paper surveys the three main research tasks in embodied AI -- visual exploration, visual navigation and embodied question answering (QA), covering the state-of-the-art approaches, evaluation  ...  Acknowledgments This research is supported by the Agency for Science, Technology and Research (A*STAR), Singapore under its AME Programmatic Funding Scheme (Award #A18A2b0046) and the National Research  ... 
arXiv:2103.04918v8 fatcat:2zu4klcchbhnvmjej5ry3emu4u

Hierarchical Reinforcement Learning Framework towards Multi-agent Navigation [article]

Wenhao Ding, Shuaijun Li, Huihuan Qian
2018 arXiv   pre-print
In this paper, we propose a navigation algorithm oriented to multi-agent environment.  ...  This algorithm is expressed as a hierarchical framework that contains a Hidden Markov Model (HMM) and a Deep Reinforcement Learning (DRL) structure.  ...  We first train the agent in a clear scene with a target, and then change the environment into a multi-agent scene.  ... 
arXiv:1807.05424v2 fatcat:fhp3l7ctqrcltfly7wag6dh7g4

HAMMER: Multi-Level Coordination of Reinforcement Learning Agents via Learned Messaging [article]

Nikunj Gupta, G Srinivasaraghavan, Swarup Kumar Mohalik, Matthew E. Taylor
2021 arXiv   pre-print
Cooperative multi-agent reinforcement learning (MARL) has achieved significant results, most notably by leveraging the representation learning abilities of deep neural networks.  ...  After explaining our MARL algorithm, hammer, and where it would be most applicable, we implement it in the cooperative navigation and multi-agent walker domains.  ...  . * Part of this work has taken place in the Intelligent Robot Learning (IRL) Lab at the University of Alberta, which is supported in part by research grants from the Alberta Machine Intelligence Institute  ... 
arXiv:2102.00824v1 fatcat:u3deetdxwvh6vffrqvueqto2xa

Shared Multi-Task Imitation Learning for Indoor Self-Navigation [article]

Junhong Xu, Qiwei Liu, Hanqing Guo, Aaron Kageza, Saeed AlQarni, Shaoen Wu
2018 arXiv   pre-print
We model each task as a sub-policy and design a multi-headed policy to learn the shared information among related tasks by summing up activations from all sub-policies.  ...  However, in the traditional imitation learning framework, one model only learns one task, and thus it lacks of the capability to support a robot to perform various different navigation tasks with one model  ...  For example, in [24] , reinforcement learning is used to train a siamese neural network to navigate to a target position.  ... 
arXiv:1808.04503v1 fatcat:gwph3rjhfzeuzjcgbsnmbf3xay

CityLearn: Diverse Real-World Environments for Sample-Efficient Navigation Policy Learning [article]

Marvin Chancán, Michael Milford
2020 arXiv   pre-print
Visual navigation tasks in real-world environments often require both self-motion and place recognition feedback.  ...  While deep reinforcement learning has shown success in solving these perception and decision-making problems in an end-to-end manner, these algorithms require large amounts of experience to learn navigation  ...  Recent deep reinforcement learning (RL) approaches have successfully performed active navigation tasks on simulated environments using real-world street imagery [2] or synthetic The work of M.C. was  ... 
arXiv:1910.04335v2 fatcat:kuwwsso3qbbs7lxsiabus3ofum
« Previous Showing results 1 — 15 out of 41,386 results