Showcasing robotics research from Imperial College


Real World Reinforcement Learning

Real World Reinfocement Learning

Fundamental robotics research at Imperial College London is helping to drive the adoption of useful robots into our healthcare, cities and homes.

Showcasing robotics research from Imperial College at the top robotics conferences of 2020 

Fundamental robotics research at Imperial College London is helping to drive the adoption of useful robots into our healthcare, cities and homes. 2020 has so far seen 17 papers from the labs that make up the?Imperial Robotics Forum?accepted into the top robotics conferences IROS and RSS. 

IEEE International Conference on Intelligent Robots and Systems 2020 (IROS) 

Department of Aeronautical Engineering 

TiltDrone: a Fully-Actuated Tilting Quad-Rotor Platform. Peter Zheng, XinKai Tan, Ba?aran Bahad?r Koçer, Erdeng Yang, Mirko Kovac – Aerial Robotics Lab 

MEDUSA: a Multi-Environment Dual-robot for Underwater Sample 
Acquisition. Diego Debruyn, Raphael Zufferey, Sophie Franziska Armanini, Crystal 
Winston, Andre Farinha, Yufei Jin, Mirko Kovac – Aerial Robotics Lab 

Aerial Sensor Placement for Cluttered Environments. Andre Farinha, Raphael Zufferey, Sophie Franziska Armanini, Peter Zheng, Mirko Kovac – Aerial Robotics Lab 

Department of Computing 

Real-World Human-Robot Collaborative Reinforcement Learning. Ali Shafti, Jonas Tjomsland, William Dudley, A. Aldo Faisal - Brain and Behaviour Labpaper video

Crossing the Gap: A Deep Dive into Zero-Shot Sim-to-Real Transfer for Dynamics. Eugene Valassakis, Zihan Ding, Edward Johns – Robot Learning Lab  

Physics-Based Dexterous Manipulations with Estimated Hand Poses and Residual Reinforcement Learning. Guillermo Garcia-Hernando, Edward Johns, Tae-Kyun Kim – Computer Vision & Learning Lab and Robot Learning Lab  

BIT-VO: Visual Odometry at 300 FPS using Binary Features from the Focal Plane. Riku Murai, Sajad Saeedi, Paul H. J. Kelly – Software Performance Optimisation Group  

Dyson School of Design Engineering 

On the False Positives and False Negatives of the Jacobian Matrix in Kinematically Redundant Parallel Mechanisms. Nicolas Baron, Andrew Philippides, Nicolas Rojas - REDS Lab. See the video here

Design and Control of SLIDER: An Ultra-lightweight, Knee-less, Low-cost Bipedal Walking Robot. Ke Wang, David Michael Marsh, Roni Permana Saputra, Digby Chappell, Zhonghe Jiang, Bethany Kon, Petar Kormushev – Robot Intelligence Lab 

Model Predictive Control for a Tendon-Driven Surgical Robot with Safety Constraints in Kinematics and Dynamics. Francesco Cursi, Valerio Modugno, Petar Kormushev – Robot Intelligence Lab 

Department of Electrical and Electronic Engineering 

Augmented Reality User Interface for Heterogeneous Multirobot Control, Rodrigo Chacon & Yiannis Demiris - Personal Robotics Lab 

Department of Mechanical Engineering 

Optimal pose estimation method for a multi-segment, programmable bevel-tip steerable needle. Alberto Favaro, Riccardo Secoli, Ferdinando Rodriguez y Baena, and Elena De Momi – Mechatronics in Medicine Lab 

A novel sensing method to detect tissue boundaries during robotic needle insertion based on laser Doppler flowmetry. Vani Virdyawan, Orsina Dessi, and Ferdinando Rodriguez y Baena – Mechatronics in Medicine Lab 

Imperial Academics are also organising the following workshops at IROS 2020 

Robot-Assisted Training for Primary Care  – Organised by Dr Thrishantha Nanayakkara, Morph Lab 

Robotics: Science and Systems 2020 (RSS) 

Department of Computing 

Aerial Manipulation Using Hybrid Force and Position NMPC Applied to Aerial Writing. Dimos Tzoumanikas, Felix Graule, Qingyue Yan, Dhruv Shah, Marija Popovic, Stefan Leutenegger  - Smart Robotics Lab. See the video here!  

Dyson School of Design Engineering 

The RUTH Gripper: Systematic Object-Invariant Prehensile In-Hand Manipulation via Reconfigurable Underactuation. Qiujie Lu, Nicholas Baron, Angus Clark, Nicolas Rojas – REDS Lab See the video here

Imperial Academics are also organising the following workshops at RSS 2020 

Closing the Academia to Real-World Gap in Service Robotics – Co-organised by Dr Petar Kormushev, Robot Intelligence Lab 


Eloise Matheson

Eloise Matheson
Department of Mechanical Engineering

Click to expand or contract

Contact details

Show all stories by this author