Adversarial Machine Learning in the Physical Domain
With deep neural networks (DNNs) being used increasingly in many applications, it is critical to improve our understanding of their failure modes and potential mitigations. A Johns Hopkins University Applied Physics Laboratory (APL) team successfully inserted a backdoor (train-time attack) into a common object detection model. In conjunction with this research, they developed a principled methodology to evaluate patch attacks (test-time attacks) and the factors impacting their success. Their approach enabled the creation of a novel optimization framework for the first-ever design of semitransparent patches that can overcome scale limitations while retaining desirable factors with regard to deployment and detectability.
Quantum Matched Filtering—Signal Processing in the Quantum Age
Optimal quantum control theory identifies the quantum equivalent of a matched filter, which maximizes the signal-to-noise ratio, enabling exploitation of extremely high sensitivity of quantum sensors to detect known signals of interest. This article describes a Johns Hopkins University Applied Physics Laboratory (APL) team’s work in this field.
Predicting Failure in Additively Manufactured Parts—“The Effects of Defects”
While the use of metal additive manufacturing (AM) has grown immensely over the past decade, there still exists a gap in understanding of process defects in AM, which often inhibit its use in critical applications such as flight hardware. The Johns Hopkins University Applied Physics Laboratory (APL) is developing novel techniques to replicate authentic surrogate defects in AM parts and characterize their effect on mechanical response. Advanced data processing methods, such as machine learning, are being leveraged to develop predictive failure models, which will help enhance our understanding of the effects of defects.
Multifunctional Hypersonic Components and Structures
This article describes a Johns Hopkins University Applied Physics Laboratory (APL) strategic independent research and development project exploring multifunctional hypersonic components and structures. The project was envisioned to develop transformational materials technologies and expertise that could be applied to relevant hypersonic vehicle programs supported at APL.
Metal Matrix Composites Synthesized with Laser-Based Additive Manufacturing
Metal matrix composites (MMCs), with their unique property combinations, have the potential to enable disruptive capabilities for extreme environment applications that require high performance from materials. A Johns Hopkins University Applied Physics Laboratory (APL) team successfully produced an aluminum-silicon carbide system with additive manufacturing (AM). The team also demonstrated the ability to grade the metal and ceramic three-dimensionally to form tailored material gradients. This effort merely scratches the surface of what is possible; future advances in AM materials development could result in materials with properties that are currently impossible to achieve with any other manufacturing process. These materials could benefit many applications.
Developing Complex Shape-Morphing Metallic Structures for Space Applications
This article describes an ongoing Johns Hopkins University Applied Physics Laboratory (APL) fundamental additive manufacturing study to fabricate large-scale (up to 10 × 10 × 13 in.3) shape-memory alloy components with locally tailored actuation stroke, force, and activation temperature.
Simplifying Digital Array Architectures with Multifunctional Metasurface Apertures
Holographic metasurfaces, tailored to exhibit a precise electromagnetic response from a low profile, are a powerful platform for wavefront manipulation and present the possibility to substantially simplify the architecture of increasingly popular (and increasingly complex) digital phased arrays. This article describes the work a Johns Hopkins University Applied Physics Laboratory (APL) team is doing in this area.
Infrared Polarization-Sensitive Imaging with Meta-Technology
A Johns Hopkins Applied Physics Laboratory (APL) team developed infrared (IR) metasurface imaging lenses designed to selectively focus specific states of polarized light (linear and circular) to different locations on a detector array. The lenses’ operational characteristics make them well suited to miniaturize future optical sensor systems planned for deployment on small platforms or personnel that cannot support the volume or mass of large optical sensor systems.
Epitaxial Chalcogenide Deposition for Optical Phase Change Devices
Because of their low power requirement and fast switching, Van der Waals layered chalcogenide superlattices have performed well in dynamic resistive memories in what is known as interfacial phase change memory devices.
Trustworthy Synthetic Biology: Plant-Based Biosensing
After establishing the four principles of Trustworthy Synthetic Biology—safety, assuredness, efficiency, and robustness—a team of researchers at the Johns Hopkins University Applied Physics Laboratory (APL) generated trustworthy plant sensor and reporter systems. Their work, initially funded as an APL independent research and development project, has since transitioned to a sponsor-funded project.