Communication of Intent in Distributed Supervisory Control Systems

Abstract

Distributed supervisory control systems are characterized by remote supervisors who work through multiple local actors to control a dynamic process. Coordination normally occurs through the use of predetermined plans and procedures. However, these plans and procedures can be underspecified and brittle when an actor is confronted with an unanticipated situation. In these instances, the local actor must adapt the plan in a manner consistent with the intent of the remote supervisor. This research investigated the communication of intent in military C2 systems by using a mixed-fidelity simulation. US army battalion commanders and subordinate company commanders developed operations orders using their standard planning procedures. Anomalies were introduced that forced them to rely on the senior commander's intent rather than the written plan. In spite of an embedded procedure for communicating intent in the military domain, in most instances, company commanders failed to follow their battalion commander's intent. Explanations for the poor performance and methods for imparting presence and communicating intent are discussed. Implications of the study extend to other types of distributed supervisory control systems in which machine agents can function as either remote supervisor or local actor.