Intelligent Robotics Research, University of Texas
The Intelligent Robotics Research (IRR) at the University of Texas performs research on intelligent behavior, behavior learning, and the learning of action and sensor models. This research is part of the university’s Artificial Intelligence lab under the Computer Science Department with the Qualitative Reasoning Research spearheaded by Professor Benjamin Kuipers.
This laboratory is responsible for the following projects:
Evolving Controllers for Physical Multi-legged Robots, the Real Time Sample Efficient Reinforcement Learning or the TEXPLORE, the Evolving Locomotion Controllers for Multi-legged Robots, the Utilizing Symmetry in Evolutionary Design, and the University of Texas Austin Villa Robot Soccer Team. The lab has also conducted research on Autonomous Intersection Management (AIM), Learning Schemas for Robot Perception, and Constructive Learning Architecture (CLA).
With the Evolving Controllers for Physical Multi-legged Robots, this research makes it possible to create robot prototypes. It also makes it convenient for researchers to prototype ideas faster. This bridges the gap between their projects, namely the Utilizing the Evolution of Network Symmetry and mOdularity (ENSO) approach, to Evolve Modular neural network controllers that are sensitive to inconsistencies between simulation and reality.
With this project, the researchers built a real quadruped robot and evolved a manipulator for this robot in simulation. This project was active from 2010-2011.
Also developed at this laboratory is the UT Austin Villa Robot Soccer Team, which has competed at the legged and simulation leagues of the RoboCup Soccer and is headed by Peter Stone, Todd Hester, Samuel Barrett, Katie Genter, Piyush Khandelwal and Patrick MacAlpine. This project started in 2003, and the team was named almost identical to one of a prominent English football club, the Aston Villa Football Club.
Under this robotic research are various laboratories, which include the like of Neural Networks and Learning Agents.
