Our military’s current and future forces are operating in a highly distributed, network centric environment which relies heavily on multidisciplinary, mixed-initiative, distributed heterogeneous teams to accomplish missions. This capability requires a high degree of situational awareness and coordination between team members. For combined arms operations, these teams include dismounted and vehicle based soldiers whose platforms can be manned or unmanned (robotic). The robotic entities range from remotely operated to fully autonomous agents. Although there has been much research on human team performance, there still remain many questions for optimizing operational systems with performance support for these “mixed-initiative” teams. For instance, the design of today’s human-robot team interfaces focuses on the skills required to operate an unmanned system. However, such designs do not address methods for operating within a distributed, mixed-initiative team. There are critical gaps in understanding the technology and methodologies needed to support the human-robot team as part of a larger, heterogeneous team. This paper will discuss research targeted at these gaps. We present the design of a distributed simulation, a virtual environment, and a human-robot interaction (HRI) team test bed. The HRI test bed includes reconfigurable multimodal interfaces to explore collaborative HRI challenges within combined arms missions. Exemplar scenarios and experimental designs for future planned studies will also be presented.