Robot Videos of Week: Potentials of DeepLearning in Robotics

Robot Videos of Week: Potentials of DeepLearning in Robotics

Jul 31, 2016 @ 18:07 | 

RSS Conference 2016 happened in June and along with that one workshop also organized on topic “Limits and Potentials of DeepLearning in Robotics”. Videos are available now on YouTube:

Introduction by Niko Sünderhauf (Australian Centre for Robotic VIsion, QUT)

End to End learning for self-driving cars by John Leonard (MIT), Larry Jackel (nVidia)

DeepLearning for Tracking and Intuitive Physics | Dieter Fox (University of Washington)

What is there to be skeptical about? | Oliver Brock (TU Berlin)

DeepLearning for Robotics | Pieter Abbeel (UC Berkeley)

Progressive Nets for simulation to robot transfer | Raia Hadsell (Google DeepMind)

Learning and sharing knowledge for Robots | Ashutosh Saxena (BrainOfThings)

Panel Discussion

Moderator: Ben Upcroft

Panelists: Pieter Abbeel, Walter Scheirer, Raia Hadsell, Larry Jackel, Dieter Fox, Ashutosh Saxena, Oliver Brock

How to Teach Autonomous Vehicles to Take Driving Decisions like Human?

Stanford engineers are conducting experiments to translate social behavior into algorithms so that self-driving cars will maintain vehicle safety and passenger comfort.

A robot as an assistant nurse? | University of Louisville

A new UofL engineering research team is building the next generation of robots. The robots appear almost human and will be able to do jobs that humans can’t or don’t want to do in military or health care settings.

Crazy Robot Mech Goats

Grand Challenges of Human-Robot Interaction in Space |  Microsoft Research

From the capture and return of an asteroid, to the orbital assembly of space vehicles, robots play a significant role in nearly every scenario of future space exploration. Given the costs and risks involved in space exploration, space agencies look to solve these problems by creating systems where humans and robots can effectively collaborate. In this talk, I describe efforts to develop control systems for two NASA robots. ATHLETE is a hexapod robot designed to support multiple mission scenarios, including asteroid exploration and capture, and mining, lunar and Martian roving. Robonaut 2 is an anthropomorphic robot designed with the manual dexterity to conduct astronaut work. An R2 is currently deployed at the ISS. From across the variety of systems developed, I identify a common set recurring, wicked problems. I describe these as the grand challenges of the robotic exploration of space, and consider the ways research in Human-Robot interaction could contribute to their solution.

Development of Speech Perception and Vocalization of Robots | Microsoft Research

One of the big issues of human cognitive development is language acquisition process, and constructive approaches have been attacking the issue to understand this process by building vocal robots. This talk gives a survey on these approaches. First, cognitive developmental robotics is briefly introduced to show a general idea of these approaches. Then, learning of vowel vocalization by parrot-like teaching is shown as one of the model of the interaction between an infant and a caregiver. Next, caregiver’s anticipation during the interaction process is analyzed, and as a result, it is shown how caregiver shapes the infant vowels. Finally, future issues are discussed towards language acquisition by cognitive developmental robotics.

iCub performing highly dynamic Tai Chi while interacting with humans | IIT

This video shows the latest results on the whole-body control of humanoid robots achieved by the Dynamic Interaction Control Lab at the Italian Institute of Technology.

The control of the robot is achieved by regulating the interaction forces between the robot and its surrounding environment. The force and torque exchanged between the robot’s feet and the floor is regulated so that the robot keeps its balance even when strongly perturbed.

CMU Leg Lab | Robotics@Georgia Tech

The Institute for Robotics and Intelligent Machines (IRIM) at Georgia Tech has launched a new series of robotics research videos on YouTube. Donated by Dr. Clinton W. Kelly III, a member of the College of Computing’s advisory board and a longtime benefactor of Georgia Tech, the diverse collection of videos covers an extended period of time and offers a behind-the-scenes look at different aspects of the earlier days of unmanned vehicle and other robotics research conducted across multiple institutions, companies, and funding agencies.

The first video in the series features the Leg Lab, established at CMU by Marc Raibert and later moved to MIT. The Leg Lab developed robots that ran and maneuvered like animals and formed the basis for the company Boston Dynamics, which was founded by Raibert in 1992. The video shows some of the early examples of legged robots from monopods to quadropeds.

Drexel’s Microswimmer Robots

While developing microswimmer robots for surgical and targeted drug delivery applications, Drexel researchers demonstrated the chain-like robots ability to split apart, operate individually, then link back together.
This video shows the use of magnetohydrodynamics for assembly and disassembly of robots.

Leave a Reply

Your email address will not be published. Required fields are marked *