Air force 14. 1 Small Business Innovation Research (sbir) Proposal Submission Instructions

OBJECTIVE: Develop and demonstrate hyperspectral processing algorithms capable of detection and reacquisition of user-designated surface targets on land (or sea) under various illumination/atmospheric conditions with varying sensor/target viewing geometries.

DESCRIPTION: HSI sensors have the unique ability to identify objects on the earth’s surface based on their unique material composition. This may allow detection of designated targets among objects that appear similar to the naked eye. Users of hyperspectral imagery products have a requirement to detect movements of designated targets using subsequent images collected hours to days later. Therefore, a target selected for discrimination in an initial image should be identifiable if it appears in subsequent images, even with changing atmospheric/illumination conditions and under varying observation geometries. Observed target spectral signatures can vary significantly for non-Lambertian objects, such as vehicles, making target detection and/or reacquisition challenging when illumination conditions and/or viewing geometry changes exist between successive images.
Two separate problems should be examined for this effort. In the first problem, the user estimates/extracts a target signature from the scene itself. This signature must then be used to reacquire the same target in subsequent images that may have differences in viewing geometry and/or illumination. The second problem would incorporate a priori bi-directional reflectance distribution function (BRDF) information into the detection/reacquisition algorithms using physical models that can incorporate the BRDF information to achieve improved target detection performance over baseline algorithms that assume Lambertian targets. Additionally, the subsequent images could be acquired by different hyperspectral sensors that may have differences in signal-to-noise ratio (SNR), spectral sampling, ground sample distance (GSD), etc. To address this issue, the algorithms developed must accommodate BRDF characteristics of the target and/or develop methods that are robust to changes in target spectral signatures resulting from these BRDF effects.
The expected development program will make use of available HSI sensor data to explore techniques and algorithms that could enable detection and reacquisition of hyperspectral targets. It would investigate the effects of changes to viewing geometry and target illumination for target materials with reflectance/emissivity characteristics ranging from diffuse to specular.
Investigation of procedures and algorithms for hand-off of targets from one HSI sensor to another is also of interest. A unique spectral signature may allow operators to acquire and specifically identify a given target using more than one sensor. The algorithms developed for robust target reacquisition must be able to accommodate differences in sensor performance, such as spectral resolution, radiometric sensitivity and calibration artifacts.

PHASE I: Develop techniques and algorithms for estimating user-designated target information (i.e., BRDF) from the hyperspectral image itself and develop methods for reacquisition of the target(s) in subsequent images. Demonstrate these techniques on existing HSI data.

PHASE II: Further refine and develop those techniques investigated during Phase I to apply to airborne imagery. Develop techniques and algorithms capable of incorporating a priori BRDF information into the detection and reacquisition of ground targets. Develop and demonstrate an experimental HSI processing system, including a user interface that is easy to learn and operate. Demonstrate the ability to do cross-sensor target reacquisition using airborne imagery.

PHASE III DUAL USE APPLICATIONS: Further refine the Phase II algorithms to produce a prototype HSI software application that can be demonstrated with an operational air or ground system. The prototype software application should be able to operate in real-time in accordance with the sensor data rates.

REFERENCES:

1. Eismann, Michael T., Hyperspectral Remote Sensing, SPIE 2012.

AF141-184 TITLE: RF Photonic Multiple, Simultaneous RF Beamforming for Phased Array Sensors

OBJECTIVE: Develop an integrated photonic TTD Unit for RF receive-only phased array to enable 8 simultaneous, independent beams with high bandwidth and linearity, low loss, sufficient delay resolution for scanning over +/- 60° with potential for reduced C-SWaP.

PHASE I: Design the architecture and components for an RF photonic, receive-only, simultaneous multi-beamforming TDU and evaluate with modeling and simulation. Perform a detailed analysis of the components, their impact on system performance and key technical challenges. Provide a conceptual design of the integrated system including the components and interfaces, C-SWaP, and system performance metrics.

PHASE II: Complete the design as well as address the key technical challenges of the photonic, simultaneous, multi-beamforming receive-only array. Design and fabricate test chips to perform RF and optical tests to characterize performance of the essential components and subsystems which can be used to evaluate the essential metrics necessary to accomplish transition to a Phase III prototype packaged system.

PHASE III DUAL USE APPLICATIONS: Fabricate and test a prototype TTD subsystem for receive-only array. Integrated photonics true-time delay can benefit consumer wideband, wired and wireless gigabit services via wideband tunable delay and phase shifting in synchronous optical networks (SONET) used in the telecommunications industry.

REFERENCES:

1. H. L. Chi (Editor), “Microwave Photonics,” CRC Press, Boca Raton, FL, 2006.

AF141-185 TITLE: Methodologies for Predicting Dormant Missile Reliabilities

OBJECTIVE: Develop an accurate line replaceable units (LRU) level reliability prediction model for dormant aging weapon systems and limited maintenance test data.

DESCRIPTION: The Air Force is interested in advancing methodologies that can accurately predict reliabilities of an existing aging dormant weapon system and its LRU. Most particularly, we are interested in predicting when LRU subsystems can be expected to wear out without any remedial action. “Wear out” is when one can expect to see a significant increase in failure rates and decrease in reliability. Knowing in advance of entering the wear-out phase for subsystems can provide lead time for planning supply chain upgrades and modifications to the system. This is very important when considering systems that sit in storage for the majority of their lifetimes and are required to operate at high reliabilities upon immediate use. Nuclear cruise missiles may reside in a dormant state for many years within a needed service life of decades (approximately 30-50 years) and then be required to operate with high reliabilities. The typical environment is for a cruise missile to be stored in a non-environmentally controlled facility for two years as part of a pylon or launcher package, minimally tested at two years as part of the package, returned to storage for two more years, minimally tested again, returned to storage for two more years, then downloaded from the package and undergo maintenance actions (like engine removal and replacement, pyro change out), extensive system level testing, and returned to storage to start the cycle again. Cruise missiles that require minimum maintenance and inspection throughout their lifetimes do not have a methodology to properly predict the reliability of the subsystems. LRUs are defined as separate boxes or items that have a unique purpose within the cruise missile. LRUs can be cables, actuators, panels, sensors, antennas, navigation boxes, batteries, engine, fuel tanks, motors, etc. Required characteristics, for these methodologies include being able to execute via computer code quickly (< 2 hrs.) with use of an industry standard PC, is based on limited maintenance data availability, can use statistical/probability distribution/confidence level processes, mathematics must be sound (i.e., must consider dissimilar data issues), not require any additional weapon system instrumentation while in storage, and can utilize LRU drawings or subject matter expert information for parts count or stress determinations as needed by the methodology. Limited maintenance data availability is defined as having field maintenance test data on certain subsystems once every two years maximum. This data only assesses pass or fail condition for the LRU.
As Phase I efforts, the Air Force would like the consideration and development of several methodologies and evaluation of the effectiveness/accuracy of the reliability predictions for those methodologies. Also, the Air Force would like a description of an approach for future activity to achieve an accurate model for a dormant cruise missile.

PHASE I: Identify and develop methodologies for predicting reliabilities of subsystems using representative data. Identify and evaluate mechanisms for proofing or validating the methodologies. Deliverable is a paper or papers that explain methodologies and mechanisms for validating the methodologies.

PHASE II: Develop prototype computer code model based on Phase I findings and that meets a set of criteria developed. Criteria will be similar as in description above. Demonstrate the accuracy of the computer code to predict reliability of dormant subsystems using representative data.

PHASE III DUAL USE APPLICATIONS: Other industries with systems/subsystems that remain dormant for extended periods of time require high reliability and are being considered for use past their original designed service life. Usable by other dormant systems/subsystems.

REFERENCES:

1. MIL-HDBK-217F, Reliability Prediction of Electronic Equipment, 2 Dec 1991.

AF141-186 TITLE: Advance Tracking Algorithms to Meet Modern Threats

OBJECTIVE: Investigate and validate the use of advanced non-linear, non-Gaussian tracking algorithms against targets which are either supermaneuverable, have low radar cross section (RCS) and/or have high scintillation properties.

PHASE I: Evaluate tracking algorithms for non-linear targets in non-Gaussian environment. Complete trade-off study for various Kalman, particle, and multi-hypothesis filters. Evaluate each approach against various supermaneuverable targets, targets with low RCS, and/or targets with specular scintillation properties. Evaluate TBD algorithms and determine SNR improvements and computational requirements.

PHASE II: Develop, demonstrate, and validate an assessment tool that addresses capability gaps identified in Phase I and provides future radar architecture design performance requirements to acquisition community for tracking and TBD algorithms. Tool will determine performance parameters and determine various sampling, data bandwidths, and computational requirements. Tool can be used to access difficult commercial tracking applications, such as low-level aircraft and non-linear movement of weather.

PHASE III DUAL USE APPLICATIONS: Construct a prototype system (hardware and analysis software) and validate in production representative environment. Follow-on activities include specific application integration and creation of any customer-unique requirements and documentation. Develop commercialization plan and market analysis.

1. "Particle Filter Theory and Practice with Positioning Applications,” Fred Gustafsson, IEEE Aerospace and Electronics Systems Magazine, 2010.

AF141-187 TITLE: Increased Radio Frequency (RF) Sampling & Radar Architecture Upgrades

OBJECTIVE: Validate the use of analog to digital converters (ADC), with higher sample rates and wider word lengths, to improve radar detection. Explore architecture upgrades which will allow rapid insertion of faster ADC, processor, and bandwidth technology.