Search for: All records

This paper proposes a data-driven optimal tracking control scheme for unknown general nonlinear systems using neural networks. First, a new neural networks structure is established to reconstruct the unknown system dynamics of the form ˙ x(t) = f (x(t))+g(x(t))u(t). Two networks in parallel are designed to approximate the functions f (x) and g(x). Then the obtained data-driven models are used to build the optimal tracking control. The developed control consists of two parts, the feed-forward control and the optimal feedback control. The optimal feedback control is developed by approximating the solution of the Hamilton-Jacobi-Bellman equation with neural networks. Unlike other studies, the Hamilton-Jacobi-Bellman solution is found by estimating the value function derivative using neural networks. Finally, the proposed control scheme is tested on a delta robot. Two trajectory tracking examples are provided to verify the effectiveness of the proposed optimal control approach.

Cloud microphysics is one of the most time‐consuming components in a climate model. In this study, we port the cloud microphysics parameterization in the Community Atmosphere Model (CAM), known as Parameterization of Unified Microphysics Across Scales (PUMAS), from CPU to GPU to seek a computational speedup. The directive‐based methods (OpenACC and OpenMP target offload) are determined as the best fit specifically for our development practices, which enable a single version of source code to run either on the CPU or GPU, and yield a better portability and maintainability. Their performance is first examined in a PUMAS stand‐alone kernel and the directive‐based methods can outperform a CPU node as long as there is enough computational burden on the GPU. A consistent behavior is observed when we run PUMAS on the GPU in a practical CAM simulation. A 3.6× speedup of the PUMAS execution time, including data movement between CPU and GPU, is achieved at a coarse horizontal resolution (8 NVIDIA V100 GPUs against 36 Intel Skylake CPU cores). This speedup further increases up to 5.4× at a high resolution (24 NVIDIA V100 GPUs against 108 Intel Skylake CPU cores), which highlights the fact that GPU favors larger problem size. Thismore »study demonstrates that using GPU in a CAM simulation can save noticeable computational costs even with a small portion of code being GPU‐enabled. Therefore, we are encouraged to port more parameterizations to GPU to take advantage of its computational benefit.

« less

Controlling the deposition of polymer-wrapped single-walled carbon nanotubes (s-CNTs) onto functionalized substrates can enable the fabrication of s-CNT arrays for semiconductor devices. In this work, we utilize classical atomistic molecular dynamics (MD) simulations to show that a simple descriptor of solvent structure near silica substrates functionalized by a wide variety of self-assembled monolayers (SAMs) can predict trends in the deposition of s-CNTs from toluene. Free energy calculations and experiments indicate that those SAMs that lead to maximum disruption of solvent structure promote deposition to the greatest extent. These findings are consistent with deposition being driven by solvent-mediated interactions that arise from SAM-solvent interactions, rather than direct s-CNT-SAM interactions, and will permit the rapid computational exploration of potential substrate designs for controlling s-CNT deposition and alignment.

Allosteric HIV-1 integrase (IN) inhibitors, or ALLINIs, are a new class of antiviral agents that bind at the dimer interface of the IN, away from the enzymatic catalytic site and block viral replication by triggering an aberrant multimerization of the viral enzyme. To further our understanding of the important binding features of multi-substituted quinoline-based ALLINIs, we have examined the IN multimerization and antiviral properties of substitution patterns at the 6 or 8 position. We found that the binding properties of these ALLINIs are negatively impacted by the presence of bulky substitutions at these positions. In addition, we have observed that the addition of bromine at either the 6 (6-bromo) or 8 (8-bromo) position conferred better antiviral properties. Finally, we found a significant loss of potency with the 6-bromo when tested with the ALLINI-resistant IN A128T mutant virus, while the 8-bromo analog retained full effectiveness.

This paper presents an inverse kinematic controller using neural networks for trajectory controlling of a delta robot in real-time. The developed control scheme is purely data-driven and does not require prior knowledge of the delta robot kinematics. Moreover, it can adapt to the changes in the kinematics of the robot. For developing the controller, the kinematic model of the delta robot is estimated by using neural networks. Then, the trained neural networks are configured as a controller in the system. The parameters of the neural networks are updated while the robot follows a path to adaptively compensate for modeling uncertainties and external disturbances of the control system. One of the main contributions of this paper is to show that updating the parameters of neural networks offers a smaller tracking error in inverse kinematic control of a delta robot with consideration of joint backlash. Different simulations and experiments are conducted to verify the proposed controller. The results show that in the presence of external disturbance, the error in trajectory tracking is bounded, and the negative effect of joint backlash in trajectory tracking is reduced. The developed method provides a new approach to the inverse kinematic control of a delta robot.

Remote sensing observational instruments are critical for better understanding and predicting severe weather. Observational data from such instruments, such as Doppler radar data, for example, are often processed for assimilation into numerical weather prediction models. As such instruments become more sophisticated, the amount of data to be processed grows and requires efficient variational analysis tools. Here we examine the code that implements the popular SAMURAI (Spline Analysis at Mesoscale Utilizing Radar and Aircraft Instrumentation) technique for estimating the atmospheric state for a given set of observations. We employ a number of techniques to significantly improve the code’s performance, including porting it to run on standard HPC clusters, analyzing and optimizing its single-node performance, implementing a more efficient nonlinear optimization method, and enabling the use of GPUs via OpenACC. Our efforts thus far have yielded more than 100x improvement over the original code on large test problems of interest to the community.