Active Perception in Human-Swarm Collaboration
We are interested in improving human-swarm team performance by including the human directly into the control loop. While fully autonomous robotic exploration requires less supervision, it neglects the operator’s intuition. We have developed a shared control algorithm that leverages the capabilities of the human and the swarm by. This project focuses on determining how autonomy allocation affects exploration efforts, thus task performance of the human-swarm collaboration. Ongoing efforts utilize virtual reality (VR) environment build in Unity Software and HTC VIVE headset before transferring to field tests. At the same time, we are collecting biometric data such as eye gaze (Pupil Labs), EEG (Emotiv) and EKG (SOMNOmedics) to determine operator’s cognitive state.
Tommy Berrueta (Ph.D. Student)
Joel Meyer (Ph.D. Student)
Katarina Popovic (Ph.D. Student)
Ahalya Prabhakar (Ph.D. Student)
Milli Schlafly (Ph.D. Student)
Annalisa Taylor (Ph.D. Student)
Allie Pinosky (Ph.D. Student)
Real-time area coverage and target localization using receding-horizon ergodic exploration
A. Mavrommati, E. Tzorakoleftherakis, I. Abraham, and T. D. Murphey
IEEE Transactions on Robotics, vol. 34, no. 1, pp. 62–80, 2018. PDF, Video 1, Video 2
This project is funded by DARPA: Interaction & Perception: Multi-Source Spectral Framework for Human-Swarm Collaboration.