Human control of cooperating robots

摘要

Advances in robotic technologies and artificial intelligence are allowing robots to emerge fromresearch laboratories into our lives. Experiences with field applications show that we haveunderestimated the importance of human-robot interaction (HRI) and that new problems arise inHRI as robotic technologies expand. This thesis classifies HRI along four dimensions - human,robot, task, and world and illustrates that previous HRI classifications can be successfullyinterpreted as either about one of these elements or about the relationship between two or moreof these elements. Current HRI studies of single-operator single-robot (SOSR) control andsingle-operator multiple-robots (SOMR) control are reviewed using this approach.Human control of multiple robots has been suggested as a way to improve effectiveness inrobot control. Unlike previous studies that investigated human interaction either in low-fidelitysimulations or based on simple tasks, this thesis investigates human interaction with cooperatingrobot teams within a realistically complex environment. USARSim, a high-fidelity game-enginebasedrobot simulator, and MrCS, a distributed multirobot control system, were developed forthis purpose. In the pilot experiment, we studied the impact of autonomy level. Mixed initiativecontrol yielded performance superior to fully autonomous and manual control.To avoid limitation to particular application fields, the present thesis focuses on commonHRI evaluations that enable us to analyze HRI effectiveness and guide HRI design independentlyof the robotic system or application domain. We introduce the interaction episode (IEP), whichwas inspired by our pilot human-multirobot control experiment, to extend the Neglect ToleranceHUMAN CONTROL OF COOPERATING ROBOTSJijun Wang, Ph.D.University of Pittsburgh, 2007vmodel to support general multiple robots control for complex tasks. Cooperation Effort (CE),Cooperation Demand (CD), and Team Attention Demand (TAD) are defined to measure thecooperation in SOMR control. Two validation experiments were conducted to validate the CDmeasurement under tight and weak cooperation conditions in a high-fidelity virtual environment.The results show that CD, as a generic HRI metric, is able to account for the various factors thataffect HRI and can be used in HRI evaluation and analysis.