Follow Us:

Alan Christiansen (left) and Alex Meng check out a Pioneer 2-AT robot.

"For teams of military robots, the human-robot interaction and collaboration must be well-defined for reliability."

Platoons of robots may take the place of live soldiers on the battlefield in the future. That's one of the goals of the Future Combat Systems program, a joint research effort by the Defense Advanced Research Projects Agency and the Army. Alan Christiansen, principal engineer in the Cognitive Science and Artificial Intelligence Department, leads a team that is exploring approaches to the command and control of robot platoons that could be used in the battlefield.

The ultimate battlefield robots may be able to think for themselves: to plan their travel routes; negotiate over, under, or around obstacles; and identify targets. They will also be able to coordinate among themselves and interact with humans on an advanced level.

Getting to such well-behaved autonomous robots will take a lot of work, especially in the area of command and control where reasoning is important. Christiansen notes that fully autonomous robots probably won't be realized for at least 10 to15 years. In the meantime, Christiansen and his colleagues are working on a variety of company-sponsored projects that may be fielded in the near term. One application involves developing robots that do search and rescue tasks in urban settings. Here, robots must find a victim or target in a collapsed building, create a map of the victim's location, and report the location. Humans can then create an extraction plan and perform the extraction.

"There are several current robotics activities at MITRE," says Christiansen. "MITRE is investigating how to command robot teams through personal digital assistants using multiple interaction modes, such as voice and gesture. Another project combines robotics with nanotechnology to develop a tiny six-legged walking robot about 1/4 in. long. Other work involves assessing the readiness of robotics technology for intelligence work."

MITRE's robotics activities go back to the late 1980s when researchers worked on robot arm teleoperation for the NASA Space Shuttle. MITRE also worked on underwater robots and ground robots that performed autonomous search and retrieval.

Alan Christiansen uses a laptop computer to issue commands to the robots behind him. In this demonstration, the robots search for a simulated victim of a disaster (such as an earthquake) and report their findings back to Christiansen.

"Our challenge," says Christiansen, "is to strengthen MITRE's robotics program and become ready to help our sponsors with this technology. I came to MITRE in May 2000 when there was quite a bit of interest in the Future Combat Systems program. We expect to see other network-centric warfare concepts emerge in the near future, and MITRE should prepare now to support those efforts." Christiansen's initial research on how MITRE could contribute to Future Combat Systems led to his current internal research into the command and control of robot platoons.

The science of command and control of multiple robots is relatively new. "The robotics field hasn't emphasized coordinating teams of robots on the battlefield," notes Christiansen. "However, for teams of military robots, the human-robot interaction and collaboration must be well-defined for reliability. We must also fully understand how autonomous robots will react in complex, hostile environments."

Not only is the work at MITRE interesting, but so are the people, says Christiansen. When he interviewed with MITRE's Cognitive Science and Artificial Intelligence Department in 2000, he was impressed by the quality of the people he met. "The people, combined with the variety of interesting projects, provided a very good opportunity," he says. Prior to coming to MITRE, Christiansen worked in academia, at national labs, and at both profit and non-profit corporations. His work was in the areas of robotics, artificial intelligence, modeling and simulation, and computer graphics.

When asked about the significance of MITRE's work in robotics, Christiansen replies, "I think MITRE has a fairly significant advice-giving role in the area of robotics. There's a middle ground between the research-oriented institutions like the universities, which may not have the same kind of long-term view that MITRE has, and the actual systems development work sponsored by the military. If we can keep a foot in each camp, then we serve a liaison function that's quite useful to the government."