Proceedings of Symposium on Modeling and Simulation and Complexity in Intelligent,

Adaptive and Autonomous Systems

Editorial

Saurabh Mittal, The MITRE Corporation, USA

Jose L. Risco Martin, Complutense University of Madrid, Spain,
Marco Luetzenberger, DAI-Labor, Germany

The increasing popularity of the Internet of Things (IoT) emphasizes that heterogeneous systems
are the norm today. A system deployed in a netcentric environment eventually becomes a part
of a system of systems (SoS). This SoS may also incorporates adaptive and autonomous elements
(such as systems that have different levels of autonomy and situated behavior). This makes
design, analysis and testing for the system-at-hand a complex endeavor. Testing in isolation is
not the same as a real-system operation, since the system’s behavior is also determined by the
input, which evolves from the environment. This exact factor is difficult to predict, due to an
ever-increasing level of autonomy. Advanced Modeling and Simulation (M&S) frameworks are
required to facilitate SoS design, development, testing, and integration. In particular, these
frameworks must provide methods to deal with intelligent, emergent, adaptive behavior,
scalability, reliability and autonomy.

The subject of emergent behavior and M&S of emergent behaviors takes the center stage in such
systems as it is unknown how a system responds in the face of emergent behavior arising out of
interactions with other complex systems. Intelligent behavior is also defined as an emergent
property in some complex systems (Mittal and Zeigler 2014). Consequently, systems that
respond and adapt to such behaviors may be called intelligent systems as well. These complex
systems exist in domains like robotics, unmanned vehicle systems, swarm engineering, semantic
web, and distributed multi-agent systems.

This dual faced problem of complexity and the resulting behavior out of such complexity needs
to be handled through the application of M&S for complex, adaptive and autonomous SoS.

The first problem deals with defining what an SoS is, and how it needs to be studied in a
particular context. Formulating the definition of SoS for a given purpose is a challenge of its
own. Some aspects that help formulate the definition may be:
• Adaptation
• Presence of learning and intelligence-based behavior
• Cognition
• Human-in-the-loop
• Methodologies, processes and mechanisms for implementing control
• Knowledge formulation, generation, and management
• Emergent behavior
• Multi-level structures
• Self-* (organization, explanation, configuration) capability and collaborative behavior
• Usage of available standards for netcentricity
• Live, Virtual and Constructive (LVC) environments
• Modeling, engineering, simulation, testing, evaluation, validation and verification
• Operations, distribution and communications
• Applications to robotics, unmanned vehicles systems, swarm technology, semantic web
technology, and multi-agent systems
The second problem is to tackle the issue of complexity, which requires incorporating
Complexity Science to advance the M&S of such complex, intelligent, adaptive and autonomous
(CIAAS) sytems. Complexity is a multi-level phenomenon that exists at structural, behavioral
and knowledge levels in such SoS. Emergent behavior is an outcome of this complexity.
Understanding emergent behavior as an outcome of this complexity will provide foundation for
resilient intelligent systems, and M&S thereof. Following are some of the areas where
complexity resides:
• Structure: network, hierarchical, small-world, flat, etc.
• Behavior: Micro and macro behaviors, local and global behaviors, teleologic and
epistemological behaviors
• Knowledge: ontology design, ontology-driven modeling, ontology-evaluation, ontology
transformation, etc.
• Human-in-the-loop: artificial agents, cognitive agents, multi-agents, man-in-loop,
human-computer-interaction
• Intelligence-basis: Situated behavior, knowledge-based behavior, mnemonic behavior,
resource-constrained systems, energy-aware systems
• Autonomy
• Adaptive algorithms and heuristics
• Architecture: Flat, full-mesh, hierarchical, adaptive, swarm, transformative
• Awareness: Self-* (organization, explanation, configuration)
• Interactions: collaboration, negotiation, greedy, rule-based, environment-based, etc.
• Heterogeneity in SoS: Live, Virtual and Constructive environment
• Artificiality: social systems, techno-economic-social systems
• Modeling: model engineering of complex SoS
• Languages: model specification using modeling languages and architecture frameworks
such as UML, PetriNets, SysML, DoDAF, MoDAF, UAF, etc.
• Simulation environments: distributed simulation, parallel simulation, cloud simulation,
netcentric parallel distributed environments
• Testing and Evaluation (T&E): tools for SoS engineering
• Heterogeneity at fundamental level: Hardware/Software Co-design, Hardware in the
Loop, Cyber Physical Systems, the Internet of Things
• Metrics for Complexity design and evaluation
• Verification, validation and accreditation of Complexity in SoS
• Application Domains: Financial, Power, Robotics, Swarm, Economic, Policy, etc.
• Failures: SoS failure, robustness evaluation

In the recent edition of Proceedings of Symposium on M&S and Complexity in Intelligent,

Adaptive and Autonomous Systems, held within the Spring Simulation Multi-conference 2018,
Tolk et al. (2018) discussed the issue of hybrid simulation for cyber physical systems and where
the community needs to go regarding complexity, intelligence and adaptability of CPS using
simulation. Zeigler, Mittal and Traore (2018) discussed some fundamental requirements and a
Discrete Event Systems (DEVS) approach for M&S of complex adaptive system of systems in
the context of the healthcare system. Perez-Vilarelle, Martin and Ayala (2018) employed
complexity principles to the subject of M&S of wind energy production in the smart-grid
scenario. Xin et al. (2018) presented the planning problem in a multi-agent system based on
sequential allocation method.

The MSCIAAS 2018 edition brings together the subjects of Complexity Science, SoS
engineering and M&S across diverse subject areas of cyber physical systems, healthcare reform,
smart-grid, and planning and control in a multi-agent system. Another recently edited text by
Mittal, Diallo and Tolk (2018) highlights the significance of M&S in exploring the emergent
behavior for complex systems engineering.

Together, these advances push the envelope in bringing Complexity Science, M&S and SoS to
develop an engineering methodology for complex adaptive system of systems.

References
Mittal, S., Diallo, S., Tolk, A., (2018) Emergent Behavior in Complex Systems Engineering: A
Modeling and Simulation Approach, Wiley & Sons.
Mittal, S., Zeigler, B.P., (2014) Context and Attention in Activity-based Intelligent Systems,
Proceedings of Activity-based Modeling and Simulation (ACTIMS’14), ITM Web of
Conferences (3), DOI: http://dx.doi.org/10.1051/itmconf/20140303001
Perez-Vilarelle, L., Martin, J.L.R., Ayala, J.L. 2018. “Modeling and simulation of wind energy
production in the smart-grid scenario”, Symposium on M&S and Complexity in Intelligent,
Adaptive and Autonomous Systems, Spring Simulation Multi-conference 2018, Baltimore,
MD, USA
Tolk, A., Barros, F., D’Ambrogio, A., Rajhans, A., Mosterman, P., Shetty, S., Traore, M.,
Vangheluwe, H., Levent Yilmaz. 2018. “Hybrid simulation for Cyber Physical Systems – A
panel on where we are going regarding complexity, intelligence and adaptability of CPS
using simulation”, Symposium on M&S and Complexity in Intelligent, Adaptive and
Autonomous Systems, Spring Simulation Multi-conference 2018, Baltimore, MD, USA
Zeigler, B.P., Mittal, S., Traore, M., 2018. “Fundamental requirements and DEVS approach for
modeling and simulation of complex adaptive system of systems: Healthcare reform”,
Symposium on M&S and Complexity in Intelligent, Adaptive and Autonomous Systems,
Spring Simulation Multi-conference 2018, Baltimore, MD, USA
Zhou, Xin, Wang, W., Wang, T., Li, X., 2018. “Patrolling task planning for the multi-layer multi-
agent system based on sequential allocation method”, Symposium on M&S and Complexity
in Intelligent, Adaptive and Autonomous Systems, Spring Simulation Multi-conference
2018, Baltimore, MD, USA

Disclaimer
The author’s affiliation with the MITRE Corporation is provided for identification purposes
only, and is not intended to convey or imply MITRE’s concurrence with, or support for, the
positions, opinions, or viewpoints expressed by the author. Approved for Public Release,
Distribution Unlimited [Case: PR_17-3254-15].