ever, many of these approaches and efforts are not in full coordination with others in the same area, and so researchers have not been able to leverage the lessons learned and emerging results of the other efforts. Relevant robotics research in the civilian market continues to dwarf ARL’s efforts, which need to be applied to filling the military gaps left in the civilian research, not to duplicating that research. A better justification of the research approaches being pursued is needed in many areas of the intelligence research. A further justification in terms of military requirements is also needed.
Some of the approaches are becoming well developed enough for standardization—for example, mapping inside a building should now be standardized, and additional research should be aimed at object identification inside the building. ARL should lead a mapping effort in this area to ensure that all areas are covered and that relevant results are being leveraged. Incremental advances also include modeling the multiple-robot patrol problem in a new way and using machine learning in a variety of ways to improve robot intelligence.
ARL is currently employing simulation in its HRI research; this should be extended. There are additional roles for HRI beyond the testing of swarming robots. HRI should be considered in all aspects and stages of robotic research. Systematic tools for doing analysis of HSI needs should be used to drive definitions of mission and scenarios. ARL should take advantage of knowledge about human cognition in perception and intelligence applications. ARL should use more real robots and consider testing at Fort Benning, Georgia, with intended user groups.
Although the long-term vision is for soldiers to have robotic teammates, in the midterm, robots could be used as tools, with functions and tasks allocated according to supporting analyses of human and robot capabilities. The utilization of a robot as a trusted team member rather than as a tool is a noble goal that the ARL programs have embraced. HRI research should be given priority very early in all robotics programs.
ARL is to be complimented for the range of robotics sizes and the different types of mobility devices in its robot research portfolio. Also, the research portfolio for robotic mobility shows a good balance of analysis and physics-based modeling and experiments. It addresses real-world effects and has great focus on meeting specific needs. Metrics were fairly well defined, and the inherent requirement of a test vehicle drove the system thinking and approach. The work on the micro flyers and that on legged robotic systems are best in class. There was a good portfolio of vehicles and platforms from small scale to mesoscale and microscale. Certain efforts were judged to be leading the state of the art in their areas. Staff were aware of the system perspective and addressed it. One area that could be emphasized is more discrete awareness of mission, sensors, and power requirements to meet the application vision and scenario. However, the efficiency of existing robotic systems in transferring energy from the engine to the environment is still several orders of magnitude worse than that of biological devices, and therefore, continued work in this area is required. Because of the burden imposed on soldiers by battery packs and the limited time on mission for robots, research that improves the overall energy density and efficiency of converting energy to motion is required. In particular, research on the combustion of small JP-8 combustion engines and research on the efficient creation and transfer of force to the environment should be added to the research portfolio.
ARL leadership is in transition. At several levels, from that of the ARL Director through individual directorates, “acting leadership” is the watchword of the day. The hard work and significant accomplishments of the current acting leaders are acknowledged, but instability introduces uncertainty, which in