Effective communication between teammates is critical to the success of collaboration, including human-robot collaboration. For enabling human robot communication, several modalities are actively being researched — such as, text, speech, visual signals, and legible motion. The design of these modalities is necessary to achieve effective communication; however, it is not sufficient. Communication typically requires both the human and robot to expend resources; and too much or too little communication has the potential to adversely affect task performance. Here, we focus on the complementary but relatively less explored problem of decision-making for communication — i.e., deciding if and when to use an available communication modality during human-robot collaborative tasks. We briefly discuss the modeling and algorithmic challenges for communication decision-making in human-robot and human-agent teams. The article concludes with an overview of our past and on-going research
in developing algorithms for communication decision-making in sequential human-robot collaborative tasks.