Call49-Topic1-Basic-SOO.doc

Statement of Objectives (SOO) for Object Physics for Recognition Advancements (OPERA) Topic 1

1.0 Background

Current exploitation algorithm technology applies state of the art algorithm technology from pattern recognition and learning theory communities to sensor data products. Bayesian decision theory is applied to develop decision boundaries in feature space. Data driven methods are validated and characterized through extensive evaluation. This effort develops methods for isolating and characterizing the object centric physical mechanisms that can be exploited for recognition functions. This effort applies controlled phenomenology measurements and physics modeling methods to develop low uncertainty data sets required for physics based feature saliency method development. Modeling and Simulation methods for distributed sensing (phenomenology-based, physics based, statistical techniques, and data driven approaches) to maximize target detection, tracking, and identification/imaging performance under the constraints of severely heterogeneous environments, difficult targets, and a rapidly diminishing EM spectrum for DoD missions.

2.0 Objectives

The objective of OPERA Topic 1 (T1) is to develop a physics-based capability to support multi-modal sensing object recognition algorithms for development and support of next generation multi-mode sensing concepts and systems. OPERA T1 aims to obtain the capability to isolate and model physical mechanisms responsible for exploitation performance earlier in the exploitation development process for effective and efficient signature exploitation systems. An approach incorporating a working systems theory is critical to allow the propagation of various sources of uncertainty in the sensing process to the impact on system performance. The objective of this program is to develop methods to discover salient physics characteristic to an object across RF and EO sensing domains with emphasis on civilian vehicles, dismounts improvised explosive devices (IED) and indicators of IED. This effort seeks to develop exploitable EO/RF such as Lidar/Radar complementary signature features using systems theory models and physical saliency methods that relate exploitable features to causal physics across wavelength. The results of which can become an important enabler for advancing distributed sensing, fully adaptive radar (FAR) and other future sensing paradigms, whose eventual military applications will include contested environments, urban operations missions, space-based radar, space situational awareness, missile defense and subsurface sensing. Methods developed will be applied to controlled measurement data for initial validation and field measurement data as appropriate for specific applications. Designed experiments will be used to quantify information gain provided by multi-sensor features within a system theory context. Current methods for saliency analysis rely on resolved data products. This effort also develops methods for extracting salient physical observables and analyzing the recognition potential of features from non-resolved data products.

3.0 Needs

Extend Radar Based Saliency Methods and System Theory via emphasis on exploiting signature data in Bi-static and Distributed radar configurations for object classification and recognition. The approach should utilize previous literature and findings on physics based models with strong consideration for integration of uncertainty sources that propagate to the final determination.

Lead to and develop new technology designed to identify salient EM scattering signatures and physics-based feature attributes (i.e. such as velocity, material properties, volume, shape, etc.) that are essential for separating targets in the exploitation algorithm’s feature space. In addition, work under this program should develop companion technology that links these salient components in signature or feature space to physical attributes on the targets chosen by the bidder for ease of access and testing. These technologies should also enable signature and feature extraction target model optimization for efficient signature prediction.

4.0 Tasks

4.1 Develop novel methods for identifying uncertainty components of the classification and recognition processes and for quantifying the uncertainty within an information theoretic framework. Such a framework should lead to better algorithm and system assessment methods and increased utility of information provided to the warfighter.

4.2 Define the exploitable physics of objects of interest such that sensor dimensions of polarization and waveform can be tailored to enhance signature collection for exploitation in bistatic and multistatic scenarios.

4.3 Space Objects Feature Extraction: Develop methods to extract structural information of a satellite from unresolved data, both electro-optical and radar. In this effort, we will test the 4.4 hypothesis that fused optical and radar data will provide more information about structural features than can be determined from either data set alone. Previous work has focused primarily on resolved signature development. New applications including Space Situational Awareness and Dismounts require that we develop a better understanding and characterization of the exploitation potential of unresolved signature data.

4.41. Detect structural features from unresolved EO data of objects in deep space.

4.42.

Create modeling and simulation capability for ground-based RF sensing of objects in deep space.

4.43.

Develop methods to analyze and exploit non-resolved (0-D and 1-D) radar signature data of space objects.

4.44.

Explore the capability to detect structural features by exploiting the complementary nature (fusing) of radar and EO data.

Unresolved data from existing electro-optic (EO) and radar sensors in the ground-to-space scenario have not been fully exploited. More work has been done in the EO arena than in the radar arena in the ground-to-space scenario. Most automatic identification algorithms in the radar arena use a statistical matching methodology to determine object identity. In this project, shape and volume estimates will be computed from the available data. This will provide structural feature information about an object, as well as improved identification capability.

4.5 Develop new signature database synthesis methodologies to support physics based saliency cueing.

4.6 Develop strategy for classification/recognition by fusion of LIDAR/RADAR data or features derived from physical phenomenology and characterizing the uncertainty in the system. The aim of this effort is to develop a loosely coupled capability for multi-sensor measurement, processing, modeling and analysis methods to support present and near term recognition database development efforts of civilian vehicle ISR applications.

5.0 Required Expertise

The Offeror shall accomplish multiple separate and independent research projects related to specific Air Force needs in the areas of sensor/sensing phenomenology, target knowledge and recognition. Individual Task Ordered research projects may be independent or complementary to one or more other already ongoing and separately funded research contracts.

6.0 Management Objectives

Management objectives include maintaining clear government visibility into program schedule, performance, and risk while allowing the contractor maximum flexibility and innovation to manage the effort. This would include program schedule, cost, performance, risks, associate contractor agreements, subcontracts and data to meet contract objectives that satisfies all technical requirements.

6.1 Meetings. Host a post-award conference within 30 days of contract award. Program management reviews and technical interchange meetings for each individual delivery order will be required. Agendas for all meetings will be developed jointly.

6.2 Systems Engineering. Disciplined systems engineering approach tailored to meet the Program and Contract Objectives is required to provide an integrated development program.

6.3 Materials. The contractor shall procure materials if they are not available through government channels and are required for the performance of this program. The materials shall consist of test equipment, supplies, replacement and new parts, subassemblies, manufacturer's literature, and manuals which are not available through government channels and/or not available in sufficient time to meet the urgency of requirements. Flight/ground testing requirements may involve the use of contractor-owned/operated or government-owned/operated testbed aircraft.

7.0 OPSEC

General OPSEC procedures, policies and awareness is required in an effort to reduce program vulnerability from successful adversary collection and exploitation of critical information. OPSEC will be applied throughout the lifecycle of the contract. The Critical Information List will be provided upon request by the RYOY Security Office. While working on the government installation OPSEC will be provided by the RYOY Security Office.

8.0 Deliverables-

8.1 CDRLS-See CDRLS attached to the solicitation.

8.2 Note: Hardware and software data items will require the addition of CLINs:

CLIN 00xx – Software Database: Software database developed and/or modified during the course of the effort.

CLIN 00xx – Computer Software: All software developed and/or modified during the course of the effort.