Cognitive Multi-Sensor Improvised Explosive Device (IED) Detection Technologies (COMIDT)

ABSTRACT: Nearly half of all U.S. and coalition deaths in Afghanistan are due to improvised explosive devices (IEDs). In addition, more than 600 IED incidents occur monthly throughout the rest of the world. While JIEDDO has introduced a number of sensor technologies that reduce the effectiveness of a given IED, the total number of devices continues to rise. Insurgents adapt and counteract new detection methods, and no one system is currently capable of detecting multiple components of an IED threat. To address this challenge, LGS is teamed with the University of Washington, Ahura, Rajant, and Raytheon to build a system that standardizes and normalizes heterogeneous sensor data in a repository. This data is then pulled into a cognitive engine that uses an artificial neural network to calculate an overall probability of an IED utilizing the sensor data, its known limitations, and outside reference sources. The information is reported in near real time on a laptop, with data sharing to continuously improve the predictive capabilities. A radio interface allows use with sensors that lack communication capabilities as well. This technology will virtually eliminate IED false positives and false negatives, while creating an ever changing sensor platform to which the enemy cannot adapt. BENEFIT: IED detection equipment is currently a $5.6 billion market worldwide providing a large commercialization opportunity for the Phase II prototype in improving Explosive Ordnance Disposal (EOD) and combat engineer mission efficiency, while protecting warfighters, Department of Homeland Security (DHS), and law enforcement personnel worldwide. The prototype system will significantly improve IED predictive capabilities and reduce false positives by an estimated 50-75%. The initial system will work with explosive, Ground Penetrating and Synthetic Aperture Radar (GPR/SAR), Radio Frequency (RF) detectors, video/intelligence feeds, and seismic sensors. Once matured, the technology will also be applicable to the $7.5 billion artificial intelligence and sensing market in applications both within and outside the DoD. These include automatic target recognition and Identification Friend-or-Foe (IFF), as well as generalized multi-sensor decision support in consumer, environmental, agricultural, medical, and safety applications. Other commercialization opportunities include navigation and the artificial vision field, which is itself a $21 billion market.