Adaptive Coordinated Control in the Multi-Agent 3D Dynamic Battlefield: Adaptive Coordinated ControL of Intelligent Multi-Agent TEams (ACCLIMATE)

Abstract

ions of Hybrid Systems for Architecture Design and Validation The overall command and control framework has a hierarchy of abstractions: high-level planning, team organization and trajectory planning. A key to this hierarchical decomposition is the ability for higher levels to have consistent abstractions of the lower layers. Our previous work [17] has been on the development of hierarchically consistent abstractions of continuous control systems and the composition of a special class of controllers with guarantees [41, 42]. We also have studied the topology of the space to define control laws, their domains and the arbitration scheme among the control laws [1]. We propose to extend this research in two key directions: 1. Hierarchical abstractions of hybrid systems. Consistent abstractions of hybrid systems are useful since for decision making at the high levels we need discrete state approximations of continuous state trajectory plans. However, the granularity of the discrete abstraction should not be so coarse as to destroy features of the plan to be implemented at the lower levels such as asking for conflicting or infeasible trajectories for UAVs. Thus, while there is an infinity of abstractions of a continuous system, the abstraction driven by specific requirements on the functionality needs to be developed.

14 Figures and Tables

Cite this paper

@inproceedings{SastryAdaptiveCC, title={Adaptive Coordinated Control in the Multi-Agent 3D Dynamic Battlefield: Adaptive Coordinated ControL of Intelligent Multi-Agent TEams (ACCLIMATE)}, author={Subramanya Sastry} }