What is agent-based modeling and simulation?
Agent-based modeling and simulation (ABMS) is a way to model the dynamics of complex systems made up of autonomous, interacting agents in order to assess their effects on the whole system. When a number of agents (human or otherwise) are working together, they tend to self-organize. Using sophisticated behavioral models, we can observe and understand how various agents are behaving and interacting with each other.
As agent modeling tools have developed, agent-based applications have sprouted up across many industries. Increasing availability of microdata and continuing computational advances have sped up this process. One such area of focus includes validating new employment concepts for unmanned and autonomous systems (UAS).
The potential of collaborative autonomy for UAS
Advances in unmanned and autonomous research are focusing on a concept called collaborative autonomy – how groups of UAS work together under a single operator’s control. Using collaborative autonomy, each unmanned vehicle would evaluate its own state relative to its environment. With this data, each UAS would make recommendations to the operator – not only for its own actions, but for the team’s actions as well. The operator would be responsible for approving or disapproving these recommendations.
Collaborative autonomy would allow unmanned aircraft to find targets and engage them as appropriate, after applying previously established rules of engagement to the situation. They could also leverage nearby systems with minimal supervision and adapt to dynamic situations, such as attrition of friendly forces or the emergence of unanticipated threats, making them more resilient.
Using ABMS to develop and validate operational concepts
Developing operational concepts for missions and validating their effectiveness through discrete-event and ABMS will help transform UAS operations. Whereas current technology often requires multiple operators for each UAS, collaborative autonomy opens the door to having one mission commander simultaneously directing all of the mission’s unmanned vehicles. Commanders would also have the flexibility to combine different UAS to best suit individual missions. This represents a dramatic improvement over the current state, in which missions depend on a single UAS with integrated capabilities.
Using multiple, coordinated UAS also insulates the mission from the impact of losing any one UAS, since others could be plugged into the system easily. This flexibility could significantly increase the mission- and cost-effectiveness of legacy assets, reduce development times and costs for future systems, and enable new deployment concepts.
Herren has extensive experience in designing discrete event and agent-based simulations of unmanned and autonomous mission scenarios, using customized software, as well as creating original models using standard programming languages. Contact us to learn more.