Van Dang

Evolutionary computation technologies for space systems

The Evolvable Computation Group, at NASAs Jet Propulsion Laboratory, is tasked with demonstrating the utility of computational engineering and computer optimized design for complex space systems. The group is comprised of researchers over a broad range of disciplines including biology, genetics, robotics, physics, computer science and system design, and employs biologically inspired evolutionary computational techniques to design and optimize complex systems. Over the past two years we have developed tools using genetic algorithms, simulated annealing and other optimizers to improve on human design of space systems. We have further demonstrated that the same tools used for computer-aided design and design evaluation can be used for automated innovation and design. These powerful techniques also serve to reduce redesign costs and schedules

Infrared sounding of the trade‐wind boundary layer: AIRS and the RICO experiment

The new generation of remote sensors on board NASAs A-Train constellation offers the possibility of observing the atmospheric boundary layer in different regimes, with or without clouds. In this study we use data from the Atmospheric InfraRed Sounder (AIRS) and of the Rain In Cumulus over the Ocean (RICO) campaign, to verify the accuracy and precision of the AIRS Version 5 Level 2 support product. This AIRS product has an improved vertical sampling that is necessary for the estimation of boundary layer properties. Good agreement is found between AIRS and RICO data, in a regime of oceanic shallow cumulus that is known to be difficult to analyze with other remote sensing data, and also shows a low sensitivity to cloud or land fraction. This suggests that AIRS data may be used for global boundary layer studies to support parameterization development in regions of difficult in-situ observation.

Risk analysis for nondeterministic mission planning and sequencing

In this paper, we address the dilemma of planning in the presence of uncertainty - the problem of scheduling events where some events might have nondeterministic durations. Real world planning and scheduling problems are almost always difficult. Planning and scheduling of events with a mixture of deterministic and nondeterministic durations is particularly challenging. The idea of scheduling events into a conflict-free plan becomes obscure and intangible when event durations are not known in advance - there is no guarantee that when the plan is executed, the scheduled events would not violate any pre-defined rules and constraints, and the resource usages would not exceed their maximum allowable limits. This dilemma of not being able to a priori quantify the likelihood of achieving a conflict-free plan in the presence of uncertainty usually results in an overly conservative plan where resources are under utilized. Making use of some standard communication link analysis techniques to characterize communication system performance, to support tradeoffs, and to manage the operational risks associate with the link usage, we instigate a probabilistic description of event durations and introduce the notion of risk in terms of probability that the plan fails to execute successfully, which we denote as Pp. We attempt to define a rational and systematic approach to weight risk against efficiency by iteratively applying constrained optimization algorithms and Monte Carlo simulations to the plan. We also derive a simple upper bound of PF for a given plan, which is independent of the optimization algorithm. This risk management approach allows planners to quantify the risk and efficiency tradeoff in the presence of uncertainty, and help to make forward-looking choices in the development and execution of the plan. Another emphasis of this paper is to demonstrate that the general criteria of optimality and rules and constraints for event planning can be described mathematically in terms of linear and non-linear functions and inequalities. This allows the use of customized and commercial off-the-shelf (COTS) constraint optimization algorithms to generate conflict-free plans. The results described in this paper are applicable to many general planning and scheduling problems. However the emphasis of this work is on mission planning and sequencing of spacecraft events with a mixture of deterministic and nondeterministic durations. Mission planning and sequencing is a critical component for mission operations. It provides a mechanism for scientists and engineers to operate the spacecraft remotely from the ground. It translates the science intents and spacecraft health and safety requests from the users into mission plans and sequences. After a rigorous process validating the plan, the plan will be transmitted to the spacecraft for its execution. Usually mission planning and sequencing and its validation are time consuming and costly operations. We apply the aforementioned methodology for formulating and optimizing both deterministic and nondeterministic sequence events planning. We demonstrate this approach with examples of scheduling science and engineering activities for mission operations.

Fusion of microwave and infrared data for enhancing its spatial resolution

The images acquired by microwave sensors are blurry and of low-resolution. On the other hand, the images obtained using infrared/visible sensors are of sufficiently high-resolution. In this paper, we develop a data fusion methodology and apply it to enhance resolution of a microwave image using the data from a collocated infrared/visible sensor. Such an approach takes advantage of the spatial resolution of the infrared instrument and the sensing accuracy of the microwave instrument. We tested our method using precipitation scenes captured with the Advanced Microwave Sounding Unit (AMSU) microwave instrument and the Advanced Very High Resolution Radiometer (AVHRR).

Enhanced Resolution of Microwave Sounder Imagery through Fusion with Infrared Sensor Data

The images acquired by microwave sensors are blurry and have low resolution. On the other hand, the images obtained using infrared/visible sensors are often of higher resolution. In this paper, we develop a data fusion methodology and apply it to enhance the resolution of a microwave image using the data from a collocated infrared/visible sensor. Such an approach takes advantage of the spatial resolution of the infrared instrument and the sensing accuracy of the microwave instrument. The model leverages sparsity in signals and is based on current research in sparse optimization and compressed sensing. We tested our method using a precipitation scene captured with the Advanced Microwave Sounding Unit (AMSU-B) microwave instrument and the Advanced Very High Resolution Radiometer (AVHRR) infrared instrument and compared the results to simultaneous radar observations. We show that the data fusion product is better than the original AMSU-B and AVHRR observations across all statistical indicators.

Custom on-demand and data access services for news merged A-Train data

To simplify access to large and complex satellite data sets for climate analysis and model verification, we developed a web service tool that is used to study long-term and global-scale trends in climate, water and energy cycle, and weather variability. A related NASA Energy and Water Cycle Study (NEWS) task has created a merged NEWS Level 2 data from multiple instruments in NASAs A-Train constellation of satellites. We used this data to enable creation of climatologies that include correlation between observed temperature, water vapor and cloud properties from the A-Train sensors. Instead of imposing on the user an often rigid and limiting web-based analysis environment, we recognize the need for simple and well-designed distributed services so that users can perform analysis in their own familiar computing environments. A Federated OpenSearch capability with full text + space + time search of data products also facilitated interoperability with other data systems.

Web services for multiplatform exploratory analysis of level 2 and 3 NEWS merged A-Train data

To simplify access to large and complex satellite data sets for climate analysis and model verification, a service-oriented architecture-based tool was developed to help study long-term and global-scale trends in climate, water and energy cycle, and weather variability. NASAs A-Train satellite constellation set of Level 2 data can be used to enable creation of climatologies that include correlation between observed temperature, water vapor and cloud properties from the A-Train sensors. However, the volume and inhomogeneity of Level 2 data have typically been difficult or time consuming to search and acquire. This tends to result in small-scale or short-term analysis. Instead of imposing on the user an often rigid and limiting web-based analysis environment, we recognize the need for well-designed distributed services so that users can perform analysis in their own familiar computing environments. Voluminous merged Level 2 data containing the various instrument data from the A-Train have recently been generated. Scientists next want to efficiently access selected sets of this merged data and perform their analysis. Server-side capabilities were developed to off-load processing and reduce the amount of data to be transferred to the client. Correspondingly, client-side processing APIs were developed to enable scientists to perform analysis of voluminous server-side data from within their own familiar computing environment (Java, Python, Matlab, IDL, C/C++, and Fortran90).