A multi-robot system for unconfined video-conferencing

Abstract

Abstract-Telepresence or tele-immersion technologies allow people to attend a shared meeting without being physically present in the same location. Commercial telepresence solutions available in the market today have significant drawbacks -they are very expensive, and confine people to the area covered by stationary cameras. In this paper, we present a mobile tele-immersion platform that addresses these issues by using robots with embedded cameras. In our system, the users can move around freely because robots autonomously adjust their locations. We provide a geometric definition of what it means to get a good view of the user, and present control algorithms to maintain a good view. The algorithms are validated both in simulation and in real experiments.

A Review of Mobile Robotic Telepresence

“…The system automatically halts and tells the pilot on which side the robot hits an obstacle and an ultrasonic sensor is used to measure the distance to the nearest obstacles for each command. A system with multiple off-the-shelf to iRobots attached robots attached to laptops for unconfined video conferencing and contributed with motion planning algorithms that maintains a good frontal view of the local user for as long as possible was introduced in [90]. The WU Telepresence Robot, a low cost robot using off-the-shelf parts and open source software [91], can be ordered online and be assembled within a couple of hours.…”

Modeling human motion patterns for multi-robot planning

“…We demonstrated the utility of our motion model and our planning algorithm in simulation, using a variant of the visibility-based multi-robot target-tracking problem from our previous work [8]. We constructed a map from a realworld office environment, collected data from students that frequently traversed its hallways, and demonstrated using simulations a two-robot strategy that maximizes the expected total time during which the person is tracked by at least one robot.…”

“…We show how to incorporate this model into a general combinatorial motion planning algorithm. We demonstrate the utility of both our motion model and our planner for a multi-robot visibility-based target-tracking problem introduced in our previous work [8]. In particular, we present results for a real-world dataset consisting of trajectories frequented by students in a university building.…”

“…In previous work, we studied a continuous-space version of this objective [8]. In this paper, our environment representation is discrete and we define instead the following POMDP reward function.…”