You are here

Advanced Machine Learning Target Recognition in Munitions

Description:

TECHNOLOGY AREA(S): Weapons 

OBJECTIVE: Apply Machine Learning/Artificial Intelligence to target recognition algorithms in gun launched munitions. 

DESCRIPTION: The Army requires advancement in Autonomous Target Recognition (ATR) algorithms for seekers in gun launched applications. Currently, seekers are capable of target detection in low clutter environments. To field a fully effective weapon that is also safe for use in conditions where there is high fratricide or collateral damage concern, the ability to discriminate between target types and between friend and foe rapidly (within minutes) and under extremely dynamic conditions is required. This topic will apply advanced and innovative machine learning and/or artificial intelligence to current and future target sensor packages that will be used in artillery, tank and mortar munitions among others. This includes but is not limited to new algorithmic approaches and/or sensor fusion approaches to improve ATR capability at extended slant ranges (3-7km), and while searching large Field of Views (FOV) (up to 3000m radius). The ability to conduct ATR in relatively inexpensive (<$10K unit at 1000 units/year) seeker architectures is critical. The munitions will experience high shock (up to 45,000 g’s) throughout a range of temperature extremes (- 25 to +145 degrees F operating range). The algorithms shall be capable of operating on emerging commercial GPU products suitable to 155mm artillery SWaP-C constraints. Detailed requirements will be provided after contract award. 

PHASE I: Phase I will consist of development of prototype algorithms on representative hardware demonstrated in laboratory simulated environments. A final report will document testing results and present the top level plan to continue development in Phase II. 

PHASE II: Phase II will continue the success of Phase I and integrate the hardware/firmware solution into a representative gun fired munition and tested at a government test range to demonstrate the ability to discern multiple disparate targets within the timing required in multiple environmental conditions. The result of Phase II will be a prototype design, including applicable technical data, which will be integrated into current and future munition designs for advanced target recognition. 

PHASE III: Upon success of Phase II, these technologies would be transitioned to munitions currently in development. Commercial applications could include law enforcement, boarder patrol/control, wildlife tracking or any other application requiring aerial identification of specific items on the ground. 

REFERENCES: 

1: CONVOLUTIONAL NEURAL NETWORKS AS FEATURE EXTRACTORS FOR DATA-SCARCEVISUALSEARCHES, Hichem ben Abdallah, September 2016. http://www.dtic.mil/dtic/tr/fulltext/u2/1029659.pdf

2:  A large-scale controlled object dataset to investigate deep learning, Ali Borji, Saeed Izadi, Laurent Itti. http://www.dtic.mil/dtic/tr/fulltext/u2/1019864.pdf

3:  PROCEEDINGS OF THE GOVERNMENT NEURAL NETWORK APPLICATIONS WORKSHOP, 24-26 August 1992, http://www.dtic.mil/dtic/tr/fulltext/u2/a259638.pdf

4:  computer vision algorithms based on biological, mathematical, and computational principles that are relevant to automatic target recognition, Steven W. Zucker, November 2004. http://www.dtic.mil/dtic/tr/fulltext/u2/a435709.pdf

KEYWORDS: Machine Learning, Artificial Intelligence, Algorithm, Munitions, Ammunition, Extended Range, Artillery, Mortars, Precision, Target Recognition, Target Tracking 

US Flag An Official Website of the United States Government