Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ARMS”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,369 records · Page 76
All sourcesNASA NTRSDOE OSTI
DOE OSTI2012

Development and Deployment of the Mobile Arm Retrieval System (MARS) - 12187

Washington River Protection Solutions (WRPS) is developing and deploying Mobile Arm Retrieval System (MARS) technologies solutions to support retrieval of radioactive and chemical waste from underground single shell storage tanks (SST) located at the Hanford Site, which is near Richland, Washington. WRPS has developed the MARS using a standardized platform that is capable of deploying multiple retrieval technologies. To date, WRPS, working with their mentor-protege company, Columbia Energy and Environmental Services (CEES), has developed two retrieval mechanisms, MARS-Sluicing (MARS-S) and MARS-Vacuum (MARS-V). MARS-S uses pressurized fluids routed through spray nozzles to mobilize waste materials to a centrally located slurry pump (deployed in 2011). MARS-V uses pressurized fluids routed through an eductor nozzle. The eductor nozzle allows a vacuum to be drawn on the waste materials. The vacuum allows the waste materials to be moved to an in-tank vessel, then extracted from the SST and subsequently pumped to newer and safer double shell tanks (DST) for storage until the waste is treated for disposal. The MARS-S system is targeted for sound SSTs (i.e., non leaking tanks). The MARS-V is targeted for assumed leaking tanks or those tanks that are of questionable integrity. Both versions of MARS are being/have been developed in compliance with WRPS's TFC-PLN-90, Technology Development Management Plan [1]. TFC-PLN-90 includes a phased approach to design, testing, and ultimate deployment of new technologies. The MARS-V is scheduled to be deployed in tank 241-C-105 in late 2012. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Finite-Time Analysis of Whittle Index based Q-Learning for Restless Multi-Armed Bandits with Neural Network Function Approximation

Whittle index policy is a heuristic to the intractable restless multi-armed bandits (RMAB) problem. Although it is provably asymptotically optimal, finding Whittle indices remains difficult. In this paper, we present Neural-Q-Whittle, a Whittle index based Q-learning algorithm for RMAB with neural network function approximation, which is an example of nonlinear two-timescale stochastic approximation with Q-function values updated on a faster timescale and Whittle indices on a slower timescale. Despite the empirical success of deep Q-learning, the non-asymptotic convergence rate of Neural-Q-Whittle, which couples neural networks with two-timescale Q-learning largely remains unclear. This paper provides a finite-time analysis of Neural-Q-Whittle, where data are generated from a Markov chain, and Q-function is approximated by a ReLU neural network. Our analysis leverages a Lyapunov drift approach to capture the evolution of two coupled parameters, and the nonlinearity in value function approximation further requires us to characterize the approximation error. Combing these provide Neural-Q-Whittle with convergence rate, where is the number of iterations.

reinforcement learning, structured learning, conve↗