Monday, November 30, 2015

TerraSwarm Funded Researcher Roozbeh Jafari's Group Developing Wearable Technology to Translate Sign Language into Text

Using a system of sensors to record and decode the muscle activity associated with each sign, TerraSwarm funded researcher Roozbeh Jafari's group at Texas A&M University is developing a wearable technology to translate sign language into text. Possible applications include decreased communication barriers for those who use sign language.

The current proof of concept device is able to send the resulting translation from the device to a computer or smartphone using Bluetooth.  The system is designed to learn from its user to accommodate the individual movement of each person. Sophisticated algorithms allow this process of adjustment and are also used to enable real time translation. Going forward the team hopes to develop a smaller device that can be worn as a watch, add a synthetic voice, and  translate full sentences rather then individual signs. 

Tuesday, November 17, 2015

TerraSwarm funded paper “Fast Redistribution of a Swarm of Heterogeneous Robots” Receives Best Paper Award at the 9th EAI International Conference on Bio-inspired Information and Communications Technologies

The paper, "Fast Redistribution of a Swarm of Heterogeneous Robots," received best paper award at the 9th EAI International Conference on Bio-inspired Information and Communications Technologies in New York.  

The paper states: "The authors gratefully acknowledge the support of ONR grants N00014-15-1-2115 and N00014-14-1-05-10, ARL grand W911NF-08-2-0004, NSF grant IIS-1426840, and TerraSwarm, one of six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA."

Amanda Prorok, M. Ani Hsieh, Vijay Kumar. Fast Redistribution of a Swarm of Heterogeneous Robots. Proceedings of 9th EAI International Conference on Bio-inspired Information and Communications Technologies, December, 3- 5, 2015, New York.

We present a method that distributes a swarm of heterogeneous robots among a set of tasks that require specialized capabilities in order to be completed. We model the system of heterogeneous robots as a community of species, where each species (robot type) is defined by the traits (capabilities) that it owns. Our method is based on a continuous abstraction of the swarm at a macroscopic level, as we model robots switching between tasks. We formulate an optimization problem that produces an optimal set of transition rates for each species, so that the desired trait distribution among the tasks is reached as quickly as possible. Our solution is based on an analytical gradient, and is computationally efficient, even for large choices of traits and species. Finally, we show that our method is capable of producing fast convergence times when compared to state-of-the-art methods.

Wednesday, November 4, 2015

TerraSwarm Funded PI Vijay Kumar Receives $5.5 Million DARPA Grant

A three year, $5.5 Million dollar grant has been awarded to GRASP Laboratory at The University of Pennsylvania from the Defense Advanced Research Project Agency for the purpose of creating new robots that are capable of navigating unfamiliar environments independently and swiftly.

TerraSwarm funded PI Vijay Kumar and fellow researchers Daniel Lee, Camillo J. Taylor, Kostas Daniilidis and Jianbo Shi will use their various expertise to develop autonomous robots, capable of flying 20 meters per second, and weighing less than three kilograms.  Possible applications include search and rescue and disaster response, particularly in situations that are hazardous for humans.