Bug-Sized ‘Bots for the Urban Battle

See that spider crawling along the sidewalk? In five years, you might want to take a closer look to see if it has a nanocomposite exoskeleton or cameras and infrared sensors for its eyes.

No, you haven’t stumbled upon a Hollywood set filming the sequel to “Minority Report,” the Tom Cruise sci-fi flick in which tarantula-like police ‘bots scuttle through buildings to identify people by scanning their irises. But you will have happened upon a technology that was inspired in part by the movie.

The Army Research Laboratory in April awarded a $37 million contract to BAE Systems to develop biologically based surveillance and reconnaissance robots to help soldiers conduct urban warfare. The terrestrial and aerial unmanned systems are part of a recent spate of Defense Department initiatives to spur miniature robotics innovations for troops on the ground.

Officials believe these insect- and bird-sized robots will help close the gap on surveillance needs not being met by the larger drones flying in the skies over Iraq and Afghanistan.

“You can’t always get a Predator over there fast enough or with the right sensors on it to provide the surveillance, particularly if troops want to know what is inside a building,” says Aaron Penkacik, chief technology officer for electronics and integrated solutions at BAE Systems. The company will lead an alliance of scientists and researchers from government, academic and industry laboratories to design and build collaborative robots that will provide troops intelligence whenever and wherever they require it, he says.

Imagine a Marine or soldier patrolling a city block when he suspects there might be insurgents in one of the buildings ahead. He stops, pulls several small robots out of his backpack and deploys them into the air and on the ground. They fly and scramble ahead, sending back images and audio to a handheld device monitored from the safety of his vehicle or under protection of his comrades.

Depending on what sensors they carry, the robots may be able to map out interior hallways, detect the transmissions from a radio and track a departing vehicle near the building. Based upon that information, troops can make better decisions on how to handle the situation.

Much of the technology already exists piecemeal in commercial and academic research labs, says Penkacik. The collaboration will allow engineers to mine previous work and couple those technologies with new developments in micro-electronics, signal processing and algorithms that will allow the small robots to be realized. But there are numerous challenges.

“It’s not a slam dunk. None of these things are — they’re hard. This is hard science,” says Penkacik.

Translating biological systems to robotic hardware will require significant advances in lightweight materials and mechanical engineering techniques. Though academic researchers have replicated how a bee flies and hovers on a rudimentary scale, much work remains to make such systems viable outside of the lab. Ongoing work in miniaturization processes and power consumption will help to alleviate the challenges in electronics packaging and heat dissipation. The team will try to leverage technologies from the Defense Advanced Research Projects Agency’s Urban Challenge last fall, Penkacik adds. The competition required teams to develop autonomous cars and trucks capable of driving through a closed suburban course on a former military base. While the race was deemed a success by the robotics industry, it also highlighted the difficulties in programming vehicles to react to moving obstacles. The BAE team will have the additional burden of developing algorithms to allow collaborative behavior between its robots.

Once researchers address those challenges, they will conquer the task of converting the gathered data into useful knowledge for the war fighter. Most of the drones’ battlefield data today is sent to a central processing site, which analyzes and then disseminates the information back to the fighters. Penkacik says the team intends to provide such knowledge directly to soldiers by having the robots operate in an ad hoc network to distribute the information to their PDAs.

Troops have noted that the ability to rapidly deploy surveillance technologies and create timely knowledge of their surroundings is of utmost importance, followed by the persistence and the duration of surveillance. The team is taking those requirements into consideration as they define the variety of system configurations needed for urban missions. Once completed, it will begin the design phase. By 2013, the team hopes to have developed demonstrable prototypes of the robots, if not deployable hardware.

Please email your comments to GJean@ndia.org

Topics: C4ISR, Sensors, Robotics, Science and Engineering Technology, DARPA