Headquarters In a Backpack
Program puts battle management tools at commanders’ fingertips.
Researchers are developing a prototype technology that may replace traditional command posts. The system consists of manportable, lightweight computers loaded with battle management software and collaboration tools. The devices will permit commanders to conduct highly mobile operations while maintaining situational awareness and connectivity to superiors and subordinates across the battlefield.
Network-centric warfare is swiftly becoming a reality. However, many legacy systems and concepts must be addressed before the U.S. military can fully embrace maneuver warfare unlimited by the need for static facilities on or near the battlespace. Technologies that sever these ties while offering increased collaboration and information sharing would greatly enhance warfighters’ effectiveness at all echelons.
Moving away from this stationary architecture is the goal of the Arlington, Virginia-based Defense Advanced Research Projects Agency’s (DARPA’s) Command Post of the Future program. The original intent of the effort was to create a better information environment in the existing infrastructure, explains Program Manager Ward Page. The project developers envisioned using large screen displays and digital sand tables for commanders to move data. However, senior military advisers rejected this approach. “They basically beat us up and said, ‘Big rooms are not the way to go. What we really want to do is get rid of command posts entirely and replace them with truly mobile, distributed operations,’” Page says.
The U.S. Army wanted a system to fit its concept of battle command on-the-move. To fulfill this vision, engineers developed portable computers called battle boards. These devices are pen-driven tablet computers of various sizes. “Basically, anywhere you are, you’ve got full connectivity to information and you’ve got distributed collaborative access with all your subordinates and commanders. You also have the ability to talk to anybody on the battlefield,” Page explains.
The program currently is using Fujitsu 3400 and 3500 pen-based computers for its experiments. The devices are roughly half the size of an open laptop computer. In addition, a number of foam and paper mock-ups are used for sizing and interaction studies. The first battle board concept model is roughly the size of a map case 18 x 24 inches when closed and 36 x 24 inches open. A number of smaller models also were developed.
The size of the model will depend on an officer’s echelon, Page believes. Commanders at the battalion, brigade and division levels will use the larger map-case-size versions because they are not participating in direct combat. Lower-echelon officers would use a smaller device roughly six inches on a side, whereas the smallest units would be the size of a personal digital assistant and fit on a soldier’s wrist or forearm. Another solution is to attach the larger and mid-size units to the back of a radio operator’s kit so the commander can use the device on the move.
Once it was decided to develop the battle boards, several key issues needed to be resolved. The first was the exact shape of the interface. Page notes that this is a critical step. “It’s really the long pole in the tent. Once you solve that, most of the other hardware issues go away,” he says.
Accurately representing data on the screen is a major area of research. Page explains this as a situational awareness task—how to present information in a manner that is quickly understood by the commander. “How do you let him interact with the information? How do you let him interact with subordinates to do collective planning, collaborative situational awareness and situational understanding?” he asks.
DARPA experiments indicate commanders currently understand only about 27 percent of what takes place on a battlefield. This represents factors such as knowledge of the forces involved and their relationships, operational patterns, possible futures, weaknesses and strengths, and understanding the enemy’s intent. This understanding can be boosted above 85 percent with visualization techniques and roughly to 92 percent through collaboration.
Another side of this approach is shared awareness in which all friendly forces in the battlespace have a common understanding of the situation. “This is a very important aspect of command—knowing what your boss wants, what your subordinates need and what your peers are planning to do. We can boost understanding—not just of the battlespace but what they know and understand—to above 90 percent consistently across the board,” Page maintains.
Understanding how commanders mentally process information may affect how the system uses bandwidth. Although Page cautions that this was not based on a formal analysis, program scientists discovered that officers define and share data visually. Commanders essentially look at maps of the battlefield and turn that image into a stream of words that another officer then decrypts back into a picture. “It’s a very inefficient communication method. If you can actually share the picture directly—and there are smart ways to do this so you are not transmitting all the pixels—then you actually get a more parsimonious communication that shares more information,” he explains.
Approximately 70 percent of battlefield communications is basic location information, such as where a unit is and what it sees. Pictures can eliminate a large amount of this type of communication, he says. Although the Army currently operates on an analog bandwidth, the adoption of the joint tactical radio system and its digital messaging will solve a number of bandwidth issues. However, this does not address large blocks of intelligence data. “We’re not passing imagery, maps or raw sensor data around yet. Although when those things come online and we have to start dealing with those, we’ll need to be smart about how we do that,” Page says.
DARPA also is developing a number of battle management technologies for the battle boards. One is a collaborative technology that creates a visualization workspace permitting officers to interactively move data between different displays.
Multimodal interfaces using speech and gesture also are under development to eliminate the need for keyboards and mice. “Commanders like to draw. If they have to type, they’ll basically put a soldier between them and the system, and you really don’t want to do that. You want the system to become an extension of the commander,” he maintains.
Researchers are examining two-dimensional sketching, iconography recognition and speech recognition technologies. Page notes that the majority of these applications are commercially available, and when they are combined, the accuracy of the visual recognition systems approaches 95 percent and voice identification 80 percent. Boosting voice access to the 95 percent range also approaches levels that commanders will find acceptable for use, he says. DARPA will conduct experiments to verify these systems in late spring and summer.
Other applications showing great promise are automated reasoning tools for conducting intent discovery. The software analyzes the enemy’s intent based on how they use their artillery and reconnaissance assets. Page notes that the battle management software for the battle boards includes collaboration and visualization tools for maneuver, intelligence, artillery fires and some logistics. He believes it is possible to conduct distributed planning and execution with these tools, but further experiments are necessary.
The program also is making use of intelligent software agents (SIGNAL, February 2001, page 67). These systems will primarily act as a communications and process management infrastructure. Page does not see immediate applications of software agents in their traditional advisory role, but he does envision automated reasoning systems being included in the computers’ battle management software at some point in the future. “We have not, at this point, wrapped it in an intelligent agent framework. But it is easier to do once we prove the utility of it. The infrastructure is there to throw the stuff in, but we just haven’t pushed on it yet,” he says.
The next step in the program is to interface with other battlefield systems. Page notes that this summer integration work will be conducted with the maneuver control system (MCS); the all source analysis system (ASAS); Force XXI battle command brigade-and- below (FBCB2); and the advanced field artillery tactical data system. He adds that the MCS work is mostly complete. The integration work provides the program with data on the other systems and some tools to aid in additional integration.
Integrating battle board technology into other systems such as MCS, ASAS or FBCB2 represents another step in the program’s development. Page observes that FBCB2 is an ideal candidate for the technology. However, this would mean stripping off the existing FBCB2 interface. “We can actually go in and run FBCB2 with our interfaces on top of their infrastructure and data sources with quite a bit of functionality,” he points out.
Designers encountered a number of cognitive issues in developing the battle boards. An important factor that emerged early in the program was that visualization is a highly individualized experience. “The same combination of elements and display mechanisms—shape, color, motion and animation—if you use them just a little differently, you get radically different results in how people understand what’s going on,” he says.
Noting that displays typically are not designed by soldiers, Page says a deeper understanding of how the human mind interprets visualization elements is necessary. The program has begun this work, but he cautions that it represents an immense long-term research issue. However, DARPA scientists have gleaned some initial results. One major factor is that the symbols for friendly and enemy forces must be radically different because they represent different levels of uncertainty.
Two-way data interaction and collaboration is another area where progress was made. Page notes that research indicates commanders depend on the help of subordinates, peers and staff in very complex ways when they share information. For example, company commanders have the best view of their part of the battlefield. “Even though the boss may be getting good overhead data, because of the way it’s filtered down, it is raw data. The guy on the ground has a better understanding of what the patterns are and has thought through the enemy’s actions in his area and can communicate that to his boss fairly well,” he says.
Currently, the only way for commanders to get this view of a battle is to either talk to their subordinates or to visit the area. A two-way information exchange between front-line officers and their commanders would create a more accurate view of the overall battle that is not delayed by processing and sensing and is connected to immediate events on the ground. Page notes that this would be a huge benefit to the Army’s reaction time and overall understanding of the battle.
The Command Post of the Future program began in 1998 and is in its final year. Unless it is extended, immediate goals are to test all the major technologies in a major experiment this June. Over the summer, the various applications will be integrated into a complete system. In October, a complete battle board prototype equipped with all four of the technologies developed by the program—collaboration, visualization, multimodal interfaces and reasoning—will participate in a demonstration. The device then will be hardened and transitioned to other operational and development groups in December, he says.