Trending News
Home / Technology / AI & IT / DARPA Causal Exploration of Complex Operational Environments developing modelling and exploration tools for hybrid Warfare

DARPA Causal Exploration of Complex Operational Environments developing modelling and exploration tools for hybrid Warfare

Over the last 15 years, the U.S. military has increasingly been called upon to face complex operational environments (OE) and diverse enemies. The modern hybrid or irregular conflicts are dominated by complex human dynamics with intertwining political, territorial, economic, ethnic, and/or religious tensions. DARPA explains, “The US military increasingly operates in remote and unstable parts of the world where mission success depends heavily on cooperation with a wide variety of stakeholder groups on civil, economic, and military matters. These groups typically include host nation government organizations, local civilian groups, and non-governmental organizations, each of which has priorities, sensitivities and concerns that may differ significantly.

 

Such complex OEs require the actions of U.S. forces and host-nation or coalition partners to be based on a common holistic understanding of the OE (e.g., governments, population groups, security forces, and violent non-state actors) and, in particular, the causal dynamics that can manifest as unanticipated and often counter-intuitive outcomes. The objective of the Causal Exploration program is to develop modelling and exploration tools to aid military planners in understanding and addressing underlying causal factors that drive regional hybrid conflicts.

 

The U.S. military has a methodology to ensure that planning is adequately informed by this holistic understanding. Referred to as “Operational Design” in joint doctrine , this methodology comprises the interdependent activities of framing the operational environment, framing the problem, and developing an operational approach. Today, these activities are largely manual, and labor and time intensive. Co-located teams – equipped with whiteboards, Post-it Notes, and PowerPoint – manually assemble knowledge about the OE, generate and subjectively evaluate hypotheses about the problem/situation, and flesh out two to three options for a recommended operational approach. The selected operational approach supports the commander’s initial guidance to the planning team.

 

 

While a range of modeling and simulation tools exist within military commands, most are special-purpose and extensive time and effort are often required to configure and use these tools. Existing tools are also heavily dependent on databases with no automated mechanism for keeping them current. These tools are not generally suitable for use directly by operational planners as they require expert modelers to assemble, configure, run, and interpret the outputs.

 

The lack of time, resources, automation, and capabilities for in-depth analysis leads to less effective planning and sub-optimal outcomes of operations. The inability of planners to rapidly and effectively achieve a shared understanding of a situation/problem, systematically explore underlying causal factors driving conflict, and objectively assess potential outcomes of a broad space of actions can severely hamper the potential impact of operations. To overcome these limitations, Causal Exploration will develop, integrate, and evaluate technologies for the development, maintenance, and use of problem-tailored causal models to inform military planning.

 

DARPA said the program is working to develop “technology for analyzing data and information arising from: intelligence networks; open and other external sources; sensors and signal/image processors; and collection platforms and weapon systems,” noting that, “Technical challenges include the need to process huge volumes of diverse, incomplete, and uncertain data in tactically-relevant timeframes.”

 

“Efforts address problems related to causal modeling, automated model construction, media integrity, graph matching, biometrics-based health assessment, domain-specific search, enterprise network defense, social media analysis, and visualization,” DARPA said, pointing out that, “Operational benefits include deeper understanding of the evolving operational environment tailored to the needs of commanders at every echelon.”

 

Causal Exploration seeks to develop a modeling platform to aid military planners in understanding and addressing underlying causal factors that drive complex conflict situations. The technologies embodied in the Causal Exploration platform will enable users to rapidly create, maintain, and interact with a causal model that has been tailored for the operational environment they are facing. Interaction with the model will allow users to explore the causal dynamics driving the conflict, and gain in-depth understanding of the operational environment to support and inform their planning efforts.

 

System Concept

In pursuit of this objective, the program will develop novel techniques for automated extraction of causally relevant knowledge, semi-automated assembly of causal models, intuitive exploration and manipulation of causal models to gain understanding, and assessment of the suitability and maturity of dynamically generated causal models

 

The envisioned result of the Causal Exploration program is an integrated system that enables military planners to rapidly formulate a causal model from source information, including human knowledge.

 

The centerpiece of the system is an Integrated Causal Model (ICM) that combines qualitative and quantitative analysis capabilities. It provides capabilities for reasoning backward from observed or hypothesized conditions to underlying causal factors, and for projecting forward from proposed actions to provide estimates of ranges of outcomes. The ICM also offers qualitative results such as prioritized or time-ordered lists of likely outcomes or causal factors, as well as quantitative projections of outcomes with appropriate measures of uncertainty.

 

The ICM is envisioned as a computational structure, accessible through a set of Application Programmer Interfaces (APIs) by the other components of the system, specifically:

  •  Knowledge Organization (KO), which automatically extracts and organizes  operational environment knowledge from heterogeneous and semantically diverse sources;
  • Causal Model Assembly (CMA), which creates the ICM. With input and feedback from the user, CMA assembles the computational causal structure of the ICM based on the knowledge provided by KO and populates, configures, and maintains the ICM to support reasoning about causal factors and projection of outcomes; and
  • Human Model Interaction (HMI), which provides the interface to enable users to explore and manipulate the ICM to answer complex “why?” and “what if?” questions.

This system provides a capability for military planners and similar users to gain an in-depth understanding of the causal dynamics of the operational environment to support and inform planning operations. Potential use cases for Causal Exploration include:

  • Exploration of the underlying causal factors driving the visible symptoms of a  conflict in order to scope the problem being addressed;
  • Exploration of causal factors impacting the desired end state;
  •  “What if?” analysis to examine hypotheses about causal factors that are not well understood;
  • “Design of Experiment” style comparative analyses of a space of options to assess the relative strengths and weaknesses of different approaches;
  •  Sensitivity analysis to assess the relative impacts of different elements of an approach or the robustness of an approach to a range of conditions or assumptions; and
  •  “System as Briefing” to explain and substantiate a problem statement and recommended operational approach to a commander.

The envisioned notional user is a member of the planning staff supp supporting the commander of a Joint Task Force (JTF)

 

The FY 2018 plans include:

• Developing technologies for populating knowledge bases with extracted entities, events and relationships in selected operational environments;

• Developing information integration and scenario modeling frameworks and interfaces to support operational design and planning for complex hybrid warfare environments;

• Developing interfaces for rapidly visualizing and evaluating models and likely outcomes of alternative courses of action; and

• Implementing, executing, and assessing models that support the design of representative hybrid missions.

 

BAE Wins DARPA Contract to Develop Conflict-Modeling Software

(DARPA, the research lab awarded BAE a $4.2 million contract for the Causal Exploration of Complex Operational Environments program, BAE said in a release.

 

The Air Force Research Lab and BAE Systems are partnering on a program that could help map out future conflicts as they occur. The concept uses technologies to “model different political, territorial and economic tensions that often lead to conflicts” and will help “planners to avoid unexpected outcomes,” the release states.

 

“Military planners often conduct manual research and use limited modeling tools to generate models and evaluate conflict situations, which are extremely time consuming and labor intensive,” said Chris Eisenbies, product line director of the Autonomy, Controls, and Estimation group at BAE Systems.

 

BAE Systems autonomy, controls and estimation group product line director Chris Eisenbies said: “Military planners often conduct manual research and use limited modelling tools to generate models and evaluate conflict situations, which are extremely time-consuming and labour-intensive.

 

“To break down these barriers, CONTEXTS will use reasoning algorithms and simulations with the goal to give planners a quicker and deeper understanding of conflicts to help avoid unexpected and counterintuitive outcomes.” BAE said the Defense Department’s need is based on the lack of time, expert resources and automated tools available as a conflict emerges.

 

Raytheon using artificial intelligence to help military planners understand what causes conflict

Raytheon BBN Technologies, based in Cambridge, MA, announced on December 6 that it will explore artificial intelligence and machine learning techniques to develop tools that may enable military planners to understand how cultural and other factors combine to cause conflicts.

 

Raytheon BBN aims to use its machine reading system – a computer system that reads prose and converts it to language understandable by artificial intelligence – to analyze the causal factors behind events and the relationships between them. These factors and relationships will be represented in semantic graphs, providing information in ways that both people and computers can understand. This, in conjunction with other modeling and visualization tools, may provide military planners a clear view of the potential repercussions of various courses of action.

 

“Using artificial intelligence, we can potentially understand what causes conflict,” said David Lintz, vice president, Raytheon BBN Technologies. “When military planners understand the root causes, they may be better able to recommend the best course of action for any given situation.”

 

The work is being done under an award from the Defense Advanced Research Projects Agency’s Causal Exploration of Complex Operational Environments program.

 

 

References and resources also include:

https://www.darpa.mil/program/causal-exploration

https://www.military.com/dodbuzz/2018/06/11/bae-wins-darpa-contract-develop-conflict-modeling-software.html

http://intelligencecommunitynews.com/raytheon-using-artificial-intelligence-to-help-military-planners-understand-what-causes-conflict/

https://www.biometricupdate.com/201806/darpa-seeks-continued-funding-for-biometric-analytic-tech-for-warfighters

About Rajesh Uppal

Check Also

Cloud Computing technology and trends driven by AI & Machine learning, IoT and 5G

Cloud computing is the on-demand delivery of IT resources over the internet with pay-as-you-go pricing. …

Leave a Reply

Your email address will not be published. Required fields are marked *

error: Content is protected !!