1,672 Hits in 4.6 sec

Distributed path planning for mobile robots using a swarm of interacting reinforcement learners

Christopher M. Vigorito
2007 Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems - AAMAS '07  
We extend previous work in this area by incorporating reinforcement learning techniques into these methods and show improved performance in simulated, rough terrain environments.  ...  ., office buildings) using highly structured, uniform deployments of networks (e.g., grids).  ...  Rough terrain was modeled as a noisy impedance factor which divided the maximum speed of a robot passing over that portion of the terrain.  ... 
doi:10.1145/1329125.1329273 dblp:conf/atal/Vigorito07 fatcat:qo3acsozibbc7cbi45x6xt2qn4

Table of Contents

2020 IEEE Robotics and Automation Letters  
Hutter 6177 Model-Based Reinforcement Learning for Time-Optimal Velocity Control . . . . . . G. Hartmann, Z. Shiller, and A.  ...  Liu 5637 Defensive Escort Teams for Navigation in Crowds via Multi-Agent Deep Reinforcement Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ... 
doi:10.1109/lra.2020.3030731 fatcat:kwx4xyitfbfuzgugbi5vavx2xu

Adaptive Tensegrity Locomotion on Rough Terrain via Reinforcement Learning [article]

David Surovik, Kun Wang, Kostas E. Bekris
2018 arXiv   pre-print
Guided Policy Search (GPS), a sample-efficient and model-free hybrid framework for optimization and reinforcement learning, has recently been used to produce periodic locomotion for a spherical 6-bar tensegrity  ...  The proposed solution incorporates new processes to ensure effective local modeling despite the disorganized nature of sample data in rough terrain locomotion.  ...  Reinforcement learning is often also associated with excessive data requirements.  ... 
arXiv:1809.10710v1 fatcat:d2c5wafmrvb25ho6makdmy5x7i

Deep Reinforcement Learning for Six Degree-of-Freedom Planetary Powered Descent and Landing [article]

Brian Gaudet, Richard Linares, Roberto Furfaro
2018 arXiv   pre-print
In this paper, we present a novel integrated guidance and control algorithm designed by applying the principles of reinforcement learning theory.  ...  Specifically, we use proximal policy optimization, a policy gradient method, to learn the policy.  ...  The authors used policy iteration [33] in a model-based formulation, where a model was learning from flight data using weighted least squares.  ... 
arXiv:1810.08719v1 fatcat:sdn6iibeu5eulkkshkbzfh33ry

Computational Aspects of Cognition and Consciousness in Intelligent Devices

Robert Kozma, Hrand Aghazarian, Terry Huntsberger, Eddie Tunstel, Walter Freeman
2007 IEEE Computational Intelligence Magazine  
Continuous adaptation and learning is a key component of computationally intelligent devices, which is achieved using dynamic models of cognition and consciousness.  ...  for control of mobile robots and demonstrating robust navigation capabilities in challenging real life scenarios.  ...  Wavelet-based visual processing converted the raw image data into an array of wavelet coefficients that characterized the roughness of the terrain.  ... 
doi:10.1109/mci.2007.385369 fatcat:jnvmwtgnrfcjnclzqaocot54ry

Monocular Visual Teach and Repeat Aided by Local Ground Planarity [chapter]

Lee Clement, Jonathan Kelly, Timothy D. Barfoot
2016 Springer Tracts in Advanced Robotics  
We validate our system over 4.3 km of autonomous navigation and demonstrate accuracy on par with the conventional stereo pipeline, even in highly non-planar terrain.  ...  the Space and Terrestrial Autonomous Robotic Systems (STARS) lab for their assistance with field testing, the Autonomous Space Robotics Lab (ASRL) for their guidance in interacting with the VT&R code base  ...  Conclusions This paper has described a Visual Teach and Repeat (VT&R) system capable of autonomously repeating kilometre-scale routes in rough terrain using only monocular vision.  ... 
doi:10.1007/978-3-319-27702-8_36 fatcat:gqxzo2v62rc4zblx4chvpr3bxq

Learning Image-Conditioned Dynamics Models for Control of Under-actuated Legged Millirobots [article]

Anusha Nagabandi, Guangzhao Yang, Thomas Asmar, Ravi Pandya, Gregory Kahn, Sergey Levine, Ronald S. Fearing
2018 arXiv   pre-print
We present an approach for controlling a real-world legged millirobot that is based on learned neural network models.  ...  Using less than 17 minutes of data, our method can learn a predictive model of the robot's dynamics that can enable effective gaits to be synthesized on the fly for following user-specified waypoints on  ...  This work: While our prior work evaluated model-based reinforcement learning with neural network models [45] , to our knowledge, the present work is the first to extend these model-based learning techniques  ... 
arXiv:1711.05253v3 fatcat:5kjcnssyzjcq5gacqb2hsrqlre

Recurrent Transition Networks for Character Locomotion [article]

Félix G. Harvey, Christopher Pal
2021 arXiv   pre-print
We both quantitatively and qualitatively evaluate our system and show that making the network terrain-aware by adding a local terrain representation to the input yields better performance for rough-terrain  ...  navigation on long transitions.  ...  with advances in reinforcement learning gave rise to deep reinforcement learning methods that addressed many limitations of these previous approaches.  ... 
arXiv:1810.02363v5 fatcat:2rqyzsmrxnhwngmpg72d5csxtm

Learning to Jump from Pixels [article]

Gabriel B. Margolis, Tao Chen, Kartik Paigwar, Xiang Fu, Donghyun Kim, Sangbae Kim, Pulkit Agrawal
2021 arXiv   pre-print
DIC affords the flexibility of model-free learning but regularizes behavior through explicit model-based optimization of ground reaction forces.  ...  Today's robotic quadruped systems can robustly walk over a diverse range of rough but continuous terrains, where the terrain elevation varies gradually.  ...  Other works have applied model-based control to terrain-aware navigation of a mapped environment, typically with complete information about the terrain [8, 34] .  ... 
arXiv:2110.15344v1 fatcat:fr3wyu4udfhvjbqod6hog33l6u

2020 Index IEEE Robotics and Automation Letters Vol. 5

2020 IEEE Robotics and Automation Letters  
., +, LRA Oct. 2020 6342-6349 Deep Reinforcement Learning for Safe Local Planning of a Ground Vehicle in Unknown Rough Terrain.  ...  ., LRA April 2020 2365-2371 Deep Reinforcement Learning for Safe Local Planning of a Ground Vehicle in Unknown Rough Terrain.  ... 
doi:10.1109/lra.2020.3032821 fatcat:qrnouccm7jb47ipq6w3erf3cja

Learning from Demonstration for Autonomous Navigation in Complex Unstructured Terrain

David Silver, J. Andrew Bagnell, Anthony Stentz
2010 The international journal of robotics research  
Rough terrain autonomous navigation continues to pose a challenge to the robotics community.  ...  Using expert examples of desired navigation behavior, mappings from both online and offline perceptual data to planning costs are learned.  ...  These correspondences can be used to learn to model the robot's interactions with terrain by predicting various terramechanical properties, such as roughness [Stavens and Thrun, 2006 ], vehicle slip  ... 
doi:10.1177/0278364910369715 fatcat:b7bizl65yzbsjge5bxqb5d72d4

Finding and transferring policies using stored behaviors

Martin Stolle, Christopher Atkeson
2010 Autonomous Robots  
In the Little Dog terrain, a quadruped robot has to navigate quickly across rough terrain.  ...  We present several algorithms that aim to advance the state-of-the-art in reinforcement learning and planning algorithms.  ...  Acknowledgements This material is based upon work supported in part by the National Science Foundation (NSF) under NSF Grant ECS-0325383 and the Defense Advanced Research Projects Agency Learning Locomotion  ... 
doi:10.1007/s10514-010-9191-2 fatcat:r7rof7k2hnadxnlx3cvgs6bnbe

RLOC: Terrain-Aware Legged Locomotion using Reinforcement Learning and Optimal Control [article]

Siddhant Gangapurwala, Mathieu Geisert, Romeo Orsolino, Maurice Fallon, Ioannis Havoutis
2020 arXiv   pre-print
We utilize on-board proprioceptive and exteroceptive feedback to map sensory information and desired base velocity commands into footstep plans using a reinforcement learning (RL) policy trained in simulation  ...  When ran online, the system tracks the generated footstep plans using a model-based controller. We evaluate the robustness of our method over a wide variety of complex terrains.  ...  We used the standard RLlib TD3 implementation for reinforcement learning.  ... 
arXiv:2012.03094v1 fatcat:ynp2g3ng6rbe3jsmqopz6rrwu4

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning [article]

Viktor Makoviychuk, Lukasz Wawrzyniak, Yunrong Guo, Michelle Lu, Kier Storey, Miles Macklin, David Hoeller, Nikita Rudin, Arthur Allshire, Ankur Handa, Gavriel State
2021 arXiv   pre-print
This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU  ...  Isaac Gym offers a high performance learning platform to train policies for wide variety of robotics tasks directly on GPU.  ...  We would like to thank the following for additional hard work helping us with this work.  ... 
arXiv:2108.10470v2 fatcat:jcaxg7alkbb73gphg26j2k27fy

Language and other complex behaviors: Unifying characteristics, computational models, neural mechanisms

Shimon Edelman
2017 Language Sciences  
An analysis of the functional characteristics shared by complex sequential behaviors suggests that they all present a common overarching computational problem: dynamically controlled constrained navigation  ...  With this conceptual framework in mind, I compare and contrast computational models of language and evaluate their potential for explaining linguistic behavior and for elucidating the brain mechanisms  ...  socially constrained and constructed spaces Applying the constrained navigation metaphor to social behaviors raises several questions.  ... 
doi:10.1016/j.langsci.2017.04.003 fatcat:kisve5cuqfb2nl42wioegqwvl4
« Previous Showing results 1 — 15 out of 1,672 results