You are here

Plume Characterization and Differentiation in Multimodal Threat Sensing


TECHNOLOGY AREA(S): Chem Bio Defense

OBJECTIVE: Develop analytical formalisms and algorithms to support reconnaissance and surveillance and integrated early warning of possible chemical or biological attacks

DESCRIPTION: Advanced sensor suites incorporated onto reconnaissance and force protection systems including the Nuclear, Biological, and Chemical Reconnaissance Vehicle and on installations, bases, and logistics support areas that support sustained maneuver are increasingly being integrated to develop great volumes of plume data in order to achieve situational awareness and immediate warning in the event of an adversary use of chemical and/or biological threat agents.A key technology component in such sensing systems is a light detection and ranging (LIDAR) system that develops geospatially accurate renderings of airborne particulate plumes in the battlefield environment.In pristine test environments, the plume data generated by elastic backscatter LIDAR measurements provides an intuitive diagnostic and metric to enable the localization and tracking of the challenge aerosol plume as it evolves from the point of release to evolve by mass transport and dispersion and (in the case of volatile aerosols) evaporation. However, in the anticipated land and/or maritime component operational space during ground maneuver and high-intensity conflict, the environment would be expected to be replete with airborne particulates caused by vehicle movement and weapons effects.Against such a backdrop of extensive clutter, the efficacy of a LIDAR plume detection and tracking capability set would be reduced if not rendered ineffective.This issue could be mitigated by exploiting the unique characteristics of a deliberate dissemination mechanism, such as a spraying device, liquid or solid aerosol dispersing shell or bomblet, or a missile or rocket incorporating a release mechanism to disseminate liquid or solid particulate matter.The effectiveness of such a signature or morphological characterization approach would be defined by the performance of a plume characterization algorithm that would tease out certain peculiar features of deliberate dissemination events that help distinguish them from more common kinetic events that would be ubiquitous in the battlefield environment.

PHASE I: The Phase I feasibility/proof-of-concept study will develop a machine learning and/or artificial intelligence-based logic approach that identifies plume features that would be expected to be peculiar to a deliberate dissemination mechanism vis-à-vis the routine and commonplace plumes that would be expected as a result of movement and kinetic events.Referee test data, to include LIDAR plume tracking measurements, from decades of release events at Army and Defense Threat Reduction Agency test facilities will be made available for the purposes of the Phase I study to develop a sufficiently robust volume of training and cross-validation data to demonstrate the bona-fides of the analytic procedure that would recognize subtle characteristics of the deliberate dissemination situation as distinguishable from the more routine plumes generated by movement or kinetic events.Machine learning approaches must be sufficiently efficient to afford real time computational analysis of LIDAR data using ordinary platform-portable computational processing. • Must be operable on a single rack-mounted computer system. • Must operate continuously and generate decisions on plume data in real time when presented with a cluttered rendition of multiple (three or more) simultaneous plumes, one of which is a representative deliberate dissemination event.

PHASE II: The Phase II Period of Performance will engineer, integrate, test, and optimize the performance of a real-time analytic software engine based on the outcome of the Phase I feasibility study.

PHASE III: A real time analytic engine that employs an artificial intelligence approach to recognize and distinguish elements among a highly cluttered LIDAR data stream in real time would be extremely useful for a variety of military unmanned systems.PHASE III DUAL USE APPLICATIONS:The real time analytic recognition and differentiation technology developed in conjunction with this topic would enable a wide variety of applications to include autonomous vehicle sensing system enhancements to recognize obstacles and pedestrians, industrial security, and agricultural autonomy control.Such algorithms may also prove to have applicability to medical diagnostic systems such as computational tomography and magnetic resonance imaging data analysis.Small business offerors should enunciate a clearly-defined commercialization strategy for the data feature recognition algorithms that includes an analysis of the market for such capabilities in industrial autonomy, environmental surveillance, diagnostic, and industrial monitoring applications.

KEYWORDS: light-induced detection and ranging, LIDAR, aerosol discrimination, plume recognition, plume morphology, artificial intelligence, machine learning, mass transport and dispersion, pattern recognition, autonomy


1. Ali H. Omar;Maria Tzortziou;Odele Coddington;Lorraine A. Remer, “Plankton Aerosol, Cloud, ocean Ecosystem mission: atmosphere measurements for air quality applications,” J. of Applied Remote Sensing, 12(4), 042608 (2018). Peter R. Colarco, Edward P. Nowottnick, Cynthia A. Randles, Bingqi Yi, Ping Yang, Kyu-Myong Kim, Jamison A. Smith, and Charles G. Bardeen, “Impact of radiatively interactive dust aerosols in the NASA GEOS-5 climate model: Sensitivity to dust particle shape and refractive index,” J. of Geophys. Res.: Atmospheres, 119(2), 753-786, 27 January 2014.3. A. J. Prata, I. F. Grant, “Retrieval of microphysical and morphological properties of volcanic ash plumes from satellite data: Application to Mt Ruapehu, New Zealand,” Quarterly J. of the Royal Met. Soc. Part B 127(576), 2153-2179 (2001).4. David S. Choi, Adam P. Showman, Ashwin R. Vasavada, Amy A. Simon-Miller, “Meteorology of Jupiter’s equatorial hot spots and plumes from Cassini,” Icarus, Volume 223, Issue 2, April 2013, Pages 832-843.5. Joseph M. Cheben, Yousheng Zeng, Jon Morris, Yanhua Ruan, “Autonomous Detection of Chemical Plumes,” US Patent App. 14/001, 356, 2014.

US Flag An Official Website of the United States Government