Modeling and Classifying Six-Dimensional Trajectories for Teleoperation Under a Time Delay

摘要

Within the context of tele-operating the JSC Robonaut humanoid robot under 2-10 second time delays, this paper explores modeling and classifying human motions represented as six-dimensional (position and orientation) trajectories. A dual path research agenda is reviewed exploring both deterministic and stochastic approaches using Hidden Markov Models. Finally, recent results are shown from a new model that integrates these two research paths. In future work it will be possible to automatically generate autonomous actions by reusing these same predictive models of human behavior to be the source of autonomous control. This approach may change the role of tele-operation from being a stand-in for autonomy into the first step of mentoring generative models capable of autonomous robotic control.