21,159 Hits in 6.5 sec

GoSafeOpt: Scalable Safe Exploration for Global Optimization of Dynamical Systems [article]

Bhavya Sukhija, Matteo Turchetta, David Lindner, Andreas Krause, Sebastian Trimpe, Dominik Baumann
2022 arXiv   pre-print
This work proposes GoSafeOpt as the first algorithm that can safely discover globally optimal policies for high-dimensional systems while giving safety and optimality guarantees.  ...  We demonstrate the superiority of GoSafeOpt over competing model-free safe learning methods on a robot arm that would be prohibitive for GoSafe.  ...  Acknowledgements We would like to thank Kyrylo Sovailo for helping us with the hardware experiments on the Franka Emika panda arm. The bindings developed (primarily by him) were essential for  ... 
arXiv:2201.09562v3 fatcat:hks533e6ynfnbppobvuikgzmqm

Scalable Safe Exploration for Global Optimization of Dynamical Systems [article]

Bhavya Sukhija, Matteo Turchetta, David Lindner, Andreas Krause, Sebastian Trimpe, Dominik Baumann
This work proposes GoSafeOpt as the first algorithm that can safely discover globally optimal policies for complex systems while giving safety and optimality guarantees.  ...  A notable exception is the GoSafe algorithm, which, unfortunately, cannot handle high-dimensional systems and hence cannot be applied to most real-world dynamical systems.  ...  Acknowledgements We would like to thank Kyrylo Sovailo for helping us with the hardware experiments on the Franka Emika panda arm. The bindings developed (primarily by him) were essential for  ... 
doi:10.48550/arxiv.2201.09562 fatcat:paulnuq5rfajhkainfippwyska

Systemwide Power Management with Argo

Daniel Ellsworth, Tapasya Patki, Swann Perarnau, Sangmin Seo, Abdelhalim Amer, Judicael Zounmevo, Rinku Gupta, Kazutomo Yoshii, Henry Hoffman, Allen Malony, Martin Schulz, Pete Beckman
2016 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)  
The Argo project is a DOE initiative for designing a modular operating system/runtime for the next generation of supercomputers.  ...  A key focus area in this project is power management, which is one of the main challenges on the path to exascale. In this paper, we discuss ideas for systemwide power management in the Argo project.  ...  Argo approaches the management of resources through recursive partitioning and facilitates both global, system-level optimizations and local control of resources at a fine granularity.  ... 
doi:10.1109/ipdpsw.2016.81 dblp:conf/ipps/EllsworthPPSAZG16 fatcat:qyu442kubfdlxhn52puoedq65y

A Distributed Protocol for Safe Real-Time Planning of Communicating Vehicles with Second-Order Dynamics

Kostas E. Bekris, Konstantinos I. Tsianos, Lydia E. Kavraki
2007 The Proceedings of First International Conference on Robot Communication and Coordination  
The algorithm is a novel integration of sampling-based motion planners with message-passing protocols for distributed constraint optimization.  ...  Experiments on a distributed simulator built on a cluster of processors confirm the safety properties of the approach in applications such as coordinated exploration.  ...  Scalability: The scalability properties of the algorithm are presented in Fig. 8 , which provides the average running times (10 runs per case) to complete exploration in the three scenes for car-like  ... 
doi:10.4108/icst.robocomm2007.2087 dblp:conf/robocomm/BekrisTK07 fatcat:eerlx3jo6zb7xkxi2hl3m5qspa

A case for secure and scalable hypervisor using safe language

Haibo Chen, Binyu Zang
2012 Proceedings of the 2012 International Workshop on Programming Models and Applications for Multicores and Manycores - PMAM '12  
We point out that the use of safe languages with scalable system design could address the trustworthiness and scalability issues with virtualization.  ...  System virtualization has been a new foundation for system software, which is evidenced in many systems and innovations, as well as numerous commercial successes in desktop, datacenter and cloud.  ...  We argue that it is time to consider the reconstruction of the hypervisor layer using safe languages as well as new scalable system design techniques, with the goal of yielding a secure, scalable, verifiable  ... 
doi:10.1145/2141702.2141717 dblp:conf/ppopp/ChenZ12 fatcat:cgl2husybbhora2ptp34hbhyem

Exploring hardware overprovisioning in power-constrained, high performance computing

Tapasya Patki, David K. Lowenthal, Barry Rountree, Martin Schulz, Bronis R. de Supinski
2013 Proceedings of the 27th international ACM conference on International conference on supercomputing - ICS '13  
While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level.  ...  In a world where power is plentiful and nodes are scarce, this solution is optimal.  ...  Part of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 (LLNL-CONF-634672).  ... 
doi:10.1145/2464996.2465009 dblp:conf/ics/PatkiLRSS13 fatcat:bolcnsbmfngitftns5chx6nexq

Distributed Reinforcement Learning for Robot Teams: A Review [article]

Yutong Wang and Mehul Damani and Pamela Wang and Yuhong Cao and Guillaume Sartoretti
2022 arXiv   pre-print
The community has leveraged model-free multi-agent reinforcement learning (MARL) to devise efficient, scalable controllers for multi-robot systems (MRS).  ...  Purpose of review: Recent advances in sensing, actuation, and computation have opened the door to multi-robot systems consisting of hundreds/thousands of robots, with promising applications to automated  ...  Springer Nature 2021 L A T E X template Distributed Reinforcement Learning for Robot Teams: A Review  ... 
arXiv:2204.03516v1 fatcat:iga6xlexmjbbflvuv5pjhifggy


Dilma Da Silva, Orran Krieger, Robert W. Wisniewski, Amos Waterland, David Tam, Andrew Baumann
2006 ACM SIGOPS Operating Systems Review  
much of the system functionality, and a sophisticated set of underlying services that provides a programming model for developing system software in a scalable and modular fashion.  ...  K42 is an open-source scalable research operating system well suited to support systems research.  ...  Acknowledgments A kernel development project is a massive undertaking, and without the efforts of many people, K42 would not be in the state it is today.  ... 
doi:10.1145/1131322.1131333 fatcat:ysdstqd5cvgjleitnwbhg55soy

Localizing globals and statics to make C programs thread-safe

Adam R. Smith, Prasad A. Kulkarni
2011 Proceedings of the 14th international conference on Compilers, architectures and synthesis for embedded systems - CASES '11  
In this work we focus on the use of static and global variables that are commonly employed in C/C++ programs, the languages of choice for developing embedded systems applications.  ...  technique to handle such variables and generate thread-safe code for parallel programs.  ...  ACKNOWLEDGMENTS We thank the anonymous reviewers for their constructive comments and suggestions. This research was supported in part by NSF grant CNS-0953268.  ... 
doi:10.1145/2038698.2038730 dblp:conf/cases/SmithK11 fatcat:qv5nakzobngbtbhrf55l74akje

Verification for Machine Learning, Autonomy, and Neural Networks Survey [article]

Weiming Xiang and Patrick Musau and Ayana A. Wild and Diego Manzanas Lopez and Nathaniel Hamilton and Xiaodong Yang and Joel Rosenfeld and Taylor T. Johnson
2018 arXiv   pre-print
This survey presents an overview of verification techniques for autonomous systems, with a focus on safety-critical autonomous cyber-physical systems (CPS) and subcomponents thereof.  ...  Recently, the formal methods and formal verification community has developed methods to characterize behaviors in these LECs with eventual goals of formally verifying specifications for LECs, and this  ...  In [141] , the authors present an algorithm for iterative adaptive dynamic pro-gramming (ADP) for problems of optimal tracking control on infinite horizon nonlinear system that are discrete-time.  ... 
arXiv:1810.01989v1 fatcat:a5ax66lsxbho3fuxuh55ypnm6m

Survey on path and view planning for UAVs

Xiaohui Zhou, Zimu Yi, Yilin Liu, Kai Huang, Hui Huang
2020 Virtual Reality & Intelligent Hardware  
In addition to the current research status, we analyze and elaborate on advantages, disadvantages, and potential explorative trends for each application domain.  ...  Methods With specific consideration for three popular UAV applications (scene reconstruction, environment exploration, and aerial cinematography), we present a survey that should assist researchers in  ...  During UAV exploration, how to represent the surrounding environment dynamically and accurately, how to plan an optimal trajectory for different missions, and how to avoid obstacles efficiently and safely  ... 
doi:10.1016/j.vrih.2019.12.004 fatcat:fsqldvofdba7lez2vab4tzvaf4

WiseMove: A Framework for Safe Deep Reinforcement Learning for Autonomous Driving [article]

Jaeyoung Lee, Aravind Balakrishnan, Ashish Gaurav, Krzysztof Czarnecki, Sean Sedwards
2019 arXiv   pre-print
We thus present WiseMove, a software framework to investigate safe deep reinforcement learning in the context of motion planning for autonomous driving.  ...  A number of authors have attempted to address this issue, but there are few publicly-available tools to adequately explore the trade-offs between functionality, scalability, and safety.  ...  Acknowledgment This work is partly supported by the Japanese Science and Technology agency (JST) ERATO project JPMJER1603: HASUO Metamathematics for Systems Design.  ... 
arXiv:1902.04118v1 fatcat:fan2yl4m7zgjzcdrpyd7wecxte

A flexible and dynamic mission planning architecture for UAV swarm coordination

Carlos Sampedro, Hriday Bavle, Jose Luis Sanchez-Lopez, Ramon A. Suarez Fernandez, Alejandro Rodriguez-Ramos, Martin Molina, Pascual Campoy
2016 2016 International Conference on Unmanned Aircraft Systems (ICUAS)  
In this paper a scalable and flexible Architecture for real-time mission planning and dynamic agent-to-task assignment for a swarm of Unmanned Aerial Vehicles (UAV) is presented.  ...  missions with the ability of dynamically re-adapt the mission in real-time.  ...  ., require the use of a complex and coordinated system for optimizing the resources while completing the mission.  ... 
doi:10.1109/icuas.2016.7502669 fatcat:lm3mrg45jbgxxnvjgl4swlkezy

GLAS: Global-to-Local Safe Autonomy Synthesis for Multi-Robot Motion Planning with End-to-End Learning [article]

Benjamin Rivière, Wolfgang Hoenig, Yisong Yue, Soon-Jo Chung
2020 arXiv   pre-print
We present GLAS: Global-to-Local Autonomy Synthesis, a provably-safe, automated distributed policy generation for multi-robot motion planning.  ...  Our approach combines the advantage of centralized planning of avoiding local minima with the advantage of decentralized controllers of scalability and distributed computation.  ...  System Safety We adopt the formulation of safe sets used in control barrier functions and define the global safe set X as the super-level set of a global safety function g : S → R.  ... 
arXiv:2002.11807v2 fatcat:4scyzbknvbbttjdaot3ljakbzm


Luke Dalessandro, Michael F. Spear, Michael L. Scott
2010 Proceedings of the 15th ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '10  
To the best of our knowledge, no extant STM system combines this set of features.  ...  Our experience suggests that NOrec may be an ideal candidate for such a software system.  ...  We see significant opportunities for further work in language integration, TMaware compiler optimizations, and dynamic adaptation to workload characteristics.  ... 
doi:10.1145/1693453.1693464 dblp:conf/ppopp/DalessandroSS10 fatcat:kfqjsfijyrbibnoi4cffmwyrci
« Previous Showing results 1 — 15 out of 21,159 results