AIRoboticsLabEd's profile picture. Advanced Intelligent Robotics Lab @ University of Edinburgh

Advanced Intelligent Robotics Lab (AIR), Edinburgh

@AIRoboticsLabEd

Advanced Intelligent Robotics Lab @ University of Edinburgh

We are happy to announce that our Case Study, in the CDT Conference for Robotics and Autonomous Systems (RAS) 2021, was selected the best among other high quality research projects! 🥳🥳🥳Stay tuned for more. For more information on our paper - DOI: 10.1109/MRA.2021.3090070

AIRoboticsLabEd's tweet image. We are happy to announce that our Case Study, in the CDT Conference for Robotics and Autonomous Systems (RAS) 2021, was selected the best among other high quality research projects! 🥳🥳🥳Stay tuned for more. For more information on our paper - DOI: 10.1109/MRA.2021.3090070

Advanced Intelligent Robotics Lab (AIR), Edinburgh 님이 재게시함

How do you teach an artificial intelligence to move around our world, which is so full of obstacles? At Edinburgh University an AI has been controlling a robotic dog which has learnt to navigate situations it has never seen before. More here. bbc.co.uk/news/business-…

Ben_Morris1's tweet image. How do you teach an artificial intelligence to move around our world, which is so full of obstacles? At Edinburgh University an AI has been controlling a robotic dog which has learnt to navigate situations it has never seen before. More here. bbc.co.uk/news/business-…

A new step towards autonomous motor behaviours, inspired by @ChrisAtkeson back in 2015. Paper links: - Science Robotics: robotics.sciencemag.org/content/5/49/e… & Arxiv arxiv.org/abs/2012.05810

.@InfAtEd, @ZJU_China researchers devised a new deep learning architecture that can combine pre-learned skills to create new skills on the fly. A four-legged robot using the #AI quickly navigated complicated situations, such as recovering from a fall: fcld.ly/spk3rix



We are very excited to share our first glimpse in fully autonomous grasping and docking from real demonstrations. 🥳 More to come, stay tuned🤖🤖


Very proud to announce our paper has been published in @SciRobotics December issue!! We generate adaptive locomotion skills by learning from a set of individual experts which enables robust locomotion on real robots.🤖 Read and watch at bit.ly/SciRobAIRLab

AIRoboticsLabEd's tweet image. Very proud to announce our paper has been published in @SciRobotics  December issue!! 

We generate adaptive locomotion skills by learning from a set of individual experts which enables robust locomotion on real robots.🤖

Read and watch at bit.ly/SciRobAIRLab

We're in the news!!! 🤖 Check out this @newscientist article about our latest work on machine learning for versatile locomotion

Robots learn to get back up after a fall in an unfamiliar environment bit.ly/2VXJSCv

newscientist's tweet image. Robots learn to get back up after a fall in an unfamiliar environment bit.ly/2VXJSCv


Paper accepted to Robotics and Automation Magazine and presentation at #IROS2020 'Decoding Motor Skills of AI and Human Policies: A Study on Humanoid and Human Balance Control' shows analysis of DRL policies can augment other control methods Paper: bit.ly/2WktGvT

AIRoboticsLabEd's tweet image. Paper accepted to Robotics and Automation Magazine and presentation at #IROS2020

'Decoding Motor Skills of AI and Human Policies: A Study on Humanoid and Human Balance Control' shows analysis of DRL policies can augment other control methods

Paper: bit.ly/2WktGvT

Recent publications no.4 at #ICRA2020 'Unified Push Recovery Fundamentals: Inspiration from Human Study' Common elements in human push recovery strategies are formulated as an MPC Paper bit.ly/2Cbg2Uo Video bit.ly/3guFT8L

AIRoboticsLabEd's tweet image. Recent publications no.4 at #ICRA2020

'Unified Push Recovery Fundamentals: Inspiration from Human Study' 

Common elements in human push recovery strategies are formulated as an MPC

Paper bit.ly/2Cbg2Uo
Video bit.ly/3guFT8L

Our #IROS2020 paper 'Optimisation of Body-ground Contact for Augmenting Whole-Body Loco-manipulation of Quadruped Robots' was accepted! Adding additional contact points to quadruped robots opens up new possibilities for manipulation! Preprint: arxiv.org/pdf/2002.10552…


Recent publications no.3 at #ICRA2020 & RA-L 'Learning natural locomotion behaviors for humanoid robots using human bias' Our new learning framework shows how imitation can produce more natural walking in humanoids Paper bit.ly/3dXASnE Video bit.ly/3eVKlgB


Recent publications no.2 at #ICRA2020 'Force-guided High-precision Grasping Control of Fragile and Deformable Objects using sEMG-based Force Prediction' EMG signals are used to help robots manipulate fragile objects Paper bit.ly/3dUYtVX Video bit.ly/2D3rodN


Highlighting our work at ICRA 2020: 'Learning Pregrasp Manipulation of Objects from Ungraspable Poses', which presents methods of learning pregrasp behaviours to help in the manipulation of difficult to grasp objects Paper bit.ly/3idxFUa Video bit.ly/38fuMOf


Feeling a sense of touch -- teleoperation using HaptX gloves

AIRoboticsLabEd's tweet image. Feeling a sense of touch -- teleoperation using HaptX gloves

Received an old photo taken during the UK-Japan workshop organized by the Alan Turing Insitute.

AIRoboticsLabEd's tweet image. Received an old photo taken during the UK-Japan workshop organized by the Alan Turing Insitute.

AI-powered, super robust, autonomous stair climbing against disturbances! Now, full onboard sensing & control, and it never fails!


Autonomous stair climbing for last-mile delivery robot.


Loading...

Something went wrong.


Something went wrong.