Lifting is as much a task about muscle strength as it is about communication. Coordinating two or more people to carry the same heavy object in the same way is vital but, as anyone who\u2019s ever gotten a couch stuck at a weird angle on a landing between staircases can attest, is not as intuitive as it should be. A new project by MIT\u2019s Computer Science & Artificial Intelligence Laboratory uses sensors placed on a human bicep to coordinate a robot lifting partner.<\/p>\n\n\n\n
\u201cDubbed RoboRaise, the system involves putting electromyography (EMG) sensors on a user\u2019s biceps and triceps to monitor muscle activity,\u201d said MIT<\/a>. \u201cIts algorithms then continuously detect changes to the person\u2019s arm level, as well as discrete up-and-down hand gestures the user might make for finer motor control.\u201d<\/p>\n\n\n\n The research, like much of what MIT\u2019s CSAIL does, is first and foremost an academic exploration, a way of marking what existing technology and sensors can do for others to build upon. Yet it\u2019s not hard to imagine how such a sensor input on a robot could be helpful in a military capacity. Delicately moving and loading ammunition in a warehouse or on a vehicle? What if, instead of a human partner, the lifting was done with a human and a robot reading signals directly from the human\u2019s bicep?<\/p>\n\n\n\n Demonstrated with a human partner, the RoboRaise tool explicitly modeled the kind of activities involved in airplane manufacture. (The research was funded, in part, by Boeing.) In the workforce, companion robots could maybe not double the work a human can do, but could certainly augment the lifting capacity of an individual. That same task can be carried over to hangars and motor pools, as assistant robots aid on-duty mechanics in the vital task of patching up what needs patching.<\/p>\n\n\n\n It isn\u2019t a Star Wars-esque astromech droid, but it may in many cases be more useful. When was the last time R2-D2 was depicted doing any heavy lifting?<\/p>\n\n\n\n More broadly, the kind of control system could have implications for military robots in other fields, like silently guiding the lifting arm of a bomb squad robot through a difficult task with more precision than a joystick. New sensors can lead to new control schemes and advance new capabilities, as we\u2019ve already seen from the hands-free navigation of the brain drone racing league<\/a>. As military planners and designers look for new ways to achieve manned-unmanned teaming, the novel control schemes from academia are worth consideration.<\/p>\n\n\n\n Watch a video of the RoboRaise in action below:<\/p>\n\n\n\n