Auxiliary System Sensor Fusion Navy SBIR FY2011.2 Sol No.: Navy SBIR FY2011.2 Topic No.: N112-159 Topic Title: Auxiliary System Sensor Fusion Proposal No.: N112-159-0259 Firm: Technical Documentation Inc 1150 First Avenue Suite 610 King of Prussia, Pennsylvania 19406 Contact: Richard Kiefer Phone: (321) 480-2597 Web Site: www.tdicorporation.net Abstract: The objective of DoD SBIR 11.2 Solicitation Topic N112-159, Auxiliary System Sensor Fusion, is to develop methods and algorithms that allow sensor information from disparate auxiliary systems to be intelligently fused to provide enhanced situational awareness (SA). A proposal in response to the soliciation topic has been jointly prepared by Technical Documentation Incorporated (TDI), a wholly-owned small business based in King of Prussia, PA and the Center for Data Analytics and Biomedical Informatics, Computer and Information Sciences Department, Temple University based in Philadelphia, PA. The initial six-month period will involve receipt of software, data and documentation from the Government, which will define the notional system simulation of a reduced scale hardware implementation of a shipboard chilled water system and an electrical system. This software, data and documentation, in conjuction with the MATLAB, Simulink and Toolboxes software tool license to be purchased by TDI, will be used to build intelligent algorithms, for the fusion of data obtained from the simulated remote sensors. The intelligent algorithms will be based on proven techniques such as Bayesian belief networks, linear and nonlinear classifiers, Kalman filtering, and Dempster-Schafer. Other techniques may be investigated as they are identified and time and material resources permit. Benefits: Currently, Navy ships monitor sensors (temperature, current, pressure, etc) to determine Situational Awareness (SA) by sensing faults and/or events that can cause failures to onboard systems. These sensors individually provide data on the particular system to which they belong or are monitoring and cannot accurately predict the likely result of these potential failures on other systems. Similarly, no method exists to offer an accurate prediction in the case where there are large gaps in data, such as during a catastrophic event. Therefore, the current technology cannot aid the ship's crew in determining all prudent corrective actions. An opportunity exists to develop a technology that will present an enhanced capability to predict the likely impact of sensor failure from one system to another, with large gaps in sensor data, and to accurately predict possible future impacts or failures. This will aid in fault detection and isolation, provide the user with an increased overall situational awareness and aid in determining any corrective actions. In addition, this will increase the Operational Availability (Ao), reduce the Total Ownership Cost (TOC) and improve safety of complex systems. These factors are largely driven by the reliability and maintainability of highly integrated equipment and by the cost of the systems and operator and maintenance personnel. Equipment reliability can be improved by increasing Mean Time Between Failure (MTBF) and maintainability can be improved by reducing Mean Time to Repair (MTTR). The TOC to the owners can be reduced by increasing Ao and reducing manpower cost. All of these factors are impacted by enhanced Situational Awareness (SA) and by reducing the number of support personnel required. A means to investigate how to achieve these results through modifications to existing systems hardware and software is proposed in this response to Navy Phase I SBIR Topic N112-159, Auxiliary Systems Sensor Fusion. Significant opportunities for commercialization, as discussed in greater detail in Section 6.0, exist with regard to both Department of Defense and private sector applications. For example, this technology could be integrated into the Navy's Integrated Condition Assessment System (ICAS) to allow retrofit into existing shipboard systems. In the private sector, many industries have infrastructures that require the observation of multiple sensors that would also benefit from this technology. For example, Government (TVA) and commercially-owned and operated nuclear power plants use remote sensing and application software for fault detection and fault isolation, and condition-based maintenance, which is very similar to the ICAS software used on US Navy ships. Return