“Human-robot teamwork vastly improves when they swap roles”

A team of roboticists has shown that by getting robots and humans to carry out simple tasks together, then swap roles, synchronicity increases significantly because the machines gain a better understanding of what’s required of them. Julie Shah and Stefanos Nikolaidis of MIT’s Interactive Robotics Group in the Computer Science and Artificial Intelligence Laboratory (CSAIL) trialled the approach, known as cross-training, in light of how successful it has been among humans in the military — not just in improving efficiency in teamwork, but in improving trust between the involved parties compared with simplistic task and reward motivator scenarios..