Robotics Research

Robotics Research: Engaging with Universities to Enable Breakthroughs in Robotics


If you want to be at the cutting-edge of your research field and publish impactful research papers, you need the most cutting-edge hardware. Our technology is unique (we own the relevant IP) and unrivalled.

Our first ever clients were researchers, and over the past 20 years, we have developed strong links with the academic community along with research companies. We’ve sold our robot Hands to universities, innovation labs and research companies across the globe including:

If you’re putting a proposal together to help fund your research, we can offer an academic discount.

Contact us for a quote or to purchase a Hand for your institute. We also offer consultancy or collaboration services to help with your research.

Popular Research Topics:

  • Artificial Intelligence and Machine Learning
  • Human-Robot Interaction (HRI) / Human-Computer Interaction (HCI)
  • Grasping and manipulation
  • Dexterity
  • Torque control
  • Motion Planning
  • Trajectory Optimisation

 

What Makes Shadow’s Key Products Useful for Researchers?

Shadow’s Dexterous Hand:

Our most in-demand Hand among researchers and developers around the globe. It’s the only robot Hand on the market to have 24 movements and 20 DOF for increased flexibility in grasping and manipulating a range of objects. Along with human-like skilfulness, the Hand is also aesthetically familiar, making it intuitive to handle. It uses an open-source platform giving researchers full access to the code along with editing ability to help researchers develop software for their own specific needs. AI and machine learning researchers have programmed the Hand to provide naturalistic simulations in a way that simple computer-driven algorithms cannot. You can read more about our recent AI advances here.

Teleoperation System:

Our new teleoperation system is a revolutionary way to remotely control robotic hardware, providing distance and safety. Researchers use our system to explore how teleoperation technology can solve their challenges. The results are impressive for theses, part of company research and/or to submit into robotic competitions like the $10M ANA Avatar XPRIZE.

Modular Grasper:

This is our modular configurable platform for grasping and allows researchers to have more control over key variables. It comes with 3 fingers with a total of 9 degrees of freedom; however, researchers have the flexibility to increase or decrease the degrees of freedom by adding or deducting fingers. It also uses an open-source platform and is quick and easy to set up so you can get to work straight away. It’s more affordable than our Dexterous Hand is more robust.

 

All our Hands come complete with:

• Full components (e.g. power supplies, adaptors, robot arm etc.)
• Training
• Delivery
• Installation
• Documentation
• 1-year warranty

 

Case Studies:

University of Washington (UW)

The University of Washington (UW), is one of our first clients and contacted us to purchase a Shadow Dexterous Hand to carry out research as part of their ADROIT project

The ADROIT project looks to explore dexterous manipulation, one of the most complex types of biological movement which has proven very difficult to replicate in robots. UW team demonstrated online planning (or model-predictive control) with a full physics model of a humanoid hand. They then augmented the actuation space with motor synergies which sped up optimisation without removing flexibility.

The results showed nonprehensile object manipulation as well as typing. In both cases, the input to the system is a high-level task description, while all details of the Hand movement emerge online from fully-automated numerical optimisation.

UW won the Best Manipulation Paper at ICRAAt the time, one of their doctoral students, Vikash Kumar, also defended his PhD stating:

I want to take a moment and extend my gratitude towards the help and support I received from you guys and the Shadow Robot Company as a whole – Dr Vikash Kumar, University of Washington

 

Bielefeld University

A team led by Prof. Helger Ritter in the Neuroinformatics group at Bielefeld University used a pair of Shadow Dexterous Hands in various Manipulation tasks such as the paper project where the group use their pair of Shadow Hands to experiment with different strategies to fold a piece of paper.

They also used the Hands as part of vision grasping where they analysed data from a vision system to determine the size shape and position of an object, then matched it to an appropriate grasp primitive in a predetermined database to successfully grasp the detected object. Watch the video of them using a pair of Shadow Dexterous Hands.

With its unique, human-like design, the Shadow Hands serve as the perfect test bed to design, evaluate, and improve computational models of human manual intelligence. Here we particularly need hands of human size and dexterity – Prof. Dr Helge Ritter

 

Paris Descartes University

Selim Eskiizmirliler and his team at Descartes University in Paris used the Hand in their research entitled: Motion control of thumb and index finger of an artificial hand for precision grip using asynchronous decoding of CM cell activity. This project involves using neural signals from the brain of a monkey to train an artificial neural network to control the Shadow Hand system. The team adjusted various parameters during the training of the ANN to achieve optimal control of the robot.

Reach out to discuss more around how our technology could enhance the learning for your students or trainees. You can come and see it in action too.

How Else Do We Support Growth in Robotics?

Some of the ways we do this is through:

Research papers
• Free downloadable virtual Hand in ROS
• Academic discounts
Internships
Consulting and research collaborations