Future infrastructures will be adaptive to comply with security policies established at the time of deployment and to allow systems to autonomously learn, by example, about acceptable changes in configuration and applications. Policies and trust inferences will be based on distributed similarity estimates within hierarchical functional groups which are autonomously formed as nodes exchange information about their configuration and state. These self-regulating and self-healing infrastructures will provide the much needed flexibility and robustness to enable system security and information assurance capabilities for future military and civilian tactical infrastructures.
To realize this vision, IHMC research scientists Jeff Bradshaw, Larry Bunch, and Adam Dalton are developing a biologically-inspired security infrastructure to enforce and maintain security policies, configuration and application integrity in networked tactical systems. This approach combines previous research in artificial immune systems with new theories and techniques involving Danger Theory (DT), autonomous damage detection and context-dependent reputation systems.
Our research in the areas of Security and Information Assurance also includes the protection of the nation’s critical infrastructure and cyber infrastructures. For that domain, IHMC is creating distributed security frameworks that use intelligent mobile agents as permanent roaming security guards for plant and network monitoring and protection.
One example is research on the development of the KAoS policy and domain services frameworks for network management and the coordination of human-agent-robot teamwork. KAoS is a policy and domain services framework based on W3C’s OWL ontology standard. KAoS was the first to offer an ontology-based approach and is currently the most successful of such efforts. KAoS has been used in a variety of government-sponsored and commercial projects for network management, secure policy-based governance of complex systems, and for the coordination of human-agent-robot teams.
Domain services supporting this research include Sol Cyber Framework an agent-based framework for distributed sensemaking. A primary implementation framework for the core capabilities of Sol is the Luna Software Agent Framework.