Computational Sciences

Research concentrates on understanding and exploiting the fundamental aspects of programming models and methodology, algorithms, and  hardware for emergent and future computing architectures for tactical, scientific, and data intensive applications. Computing systems include both mobile and fixed architectures optimized for limited communications, lower power consumption, hierarchical memory, distributed algorithms, resiliency.

AIML Enabled Adaptive Computing (APG)

Real-time computational states, light weight intelligent decision making functions and a set of common interface abstractions are essential to realizing an optimized computational efficiency of resource constrained tactical computing platforms. AIML enabled hardware independent resource constraints-aware adaptive computing engine (ACE) with optimization and decision making functions will assist tactical computing platforms in distributed cooperative coded computing and in making intelligent computational off-loading decisions when there are insufficient computational resource to complete the input task. While there are several intelligent AIML frameworks and optimization algorithms, many of them are unsuitable for tactical computing environments due to their computational complexity. We are researching into the development of mission deployable light weight AIML enabled adaptive computing strategies, optimization and decision making algorithms to solve tactical computing problems at the edge

ARL is seeking collaborations in

  • Development of efficient run-time algorithms and AIML frameworks suitable for adaptive computing in heterogeneous tactical environment.
  • Algorithms to predict the appropriate mobility parameters for successful computational off-loading in resource constrained environments.
  • Generalized computing and network abstractions to realize efficient distributed heterogeneous computing in tactical environments.

Principal Investigator:

Dr. Venkat Dasari, Venkateswara.r.dasari.civ@mail.mil, (410) 278-2846

Domain Specific & Emergent Computing Architectures (APG)

Emerging computing architectures can provide revolutionary computational capabilities which enable greater autonomy and adaptability at the tactical edge. ARL is providing a core capability to understand, innovate and integrate emergent computing architectures that may provide revolutionary capabilities over other well-understood evolutions of current architectures. ARL seeks collaborations in developing methodology and implementations using heterogeneous architectures for Army applications such as perception, control, online learning, and autonomy.

Of interest are computing architectures that are SWaP, architectures that support computational intelligence and neuromorphic computing, architectures for embedded processing, architectures for high-performance computing which improve performance & enable increased system functionality, tools to increase the productivity of developing applications, methods and architectures that can provide dramatic improvements in the performance/cost of systems. Other novel hardware paradigms of interest include exploring alternatives to Von Neumann computing architectures.

Principal Investigator

Mr. Dale Shires, dale.r.shires.civ@mail.mil, 410-248-5006