Artificial Intelligence & Machine Learning
You can train our Shadow Dexterous Hand to manipulate physical objects and complete goal-orientated tasks intuitively
Shadow is a proud winner of Awards.
AI 2020 and the AIconics Award 2019 for Best Innovation in AI Hardware
We’re here if you want to learn more about our product's capabilities in Deep Learning.
Researchers have used neural networks to train our robot Hand in simulation using the trial-and-error principle (reinforcement learning).
The data is then transferred to our robot Hand so that it can perform the desired action in a real-life setting and in real-time.
Since our system has no prior information, it removes bias and enables researchers to explore the system freely for revolutionary results.
Case studies
OpenAI
We’ve worked with OpenAI, founded by business tycoons, Elon Musk and Sam Altman to advance research within AI and machine learning using the Shadow Dexterous Hand.
- OpenAI used the Shadow Hand to study ways of learning dexterity using the complex task of solving a Rubik’s cube single-handedly.
- They trained their system entirely in simulation via reinforcement learning.
- Shadow Dexterous Hand demonstrated the learnings in the real world. No fine-tuning needed.
- Featured on BBC News and the front page of the New York Times.
Case studies
Human Brain Project (HBP)
HBP and Maastricht University successfully integrate and simulate the Shadow Hand on the HBP Neurorobotics Platform which connects a physics simulator to a variety of neural networks (or brains).
- The Human Brain Project added the virtual Shadow Hand to their EBRAINS infrastructure to learn more about how the brain coordinates complex hand movements.
- Each Shadow Hand joint can be individually linked to both the input and/or output of the brain.
- Contact-based feedback is forwarded to the brain. No need for time-consuming physical setups.
Case studies
Google Brain
Google Brain used the Shadow Hand to learn how to manipulate multiple objects using just a few hours of real-world data.
- The team used a novel robot task planning technique involving deep dynamics models (DDM).
- They were able to manipulate multiple objects with just four hours of real-world data.
- More info about the impressive results in this article.
Testimonials
OpenAI
“People are able to perform a wide range of dexterous manipulation tasks in a diverse set of environments, making the human hand a grounded source of inspiration for research into robotic manipulation”
OpenAI
Testimonials
Human Brain Project & Maastricht University
“The Shadow Hand exhibits human-level dexterity which allows a research team at Maastricht University to study how the brain coordinates complex hand movements… By providing a highly realistic model of the human hand, the Shadow Hand allows neuroscientists to develop more realistic models around how the human brain works.”
Mario Senden, Assistant Professor
Faculty of Psychology and Neuroscience at Maastricht University
Contact us
A quick and easy way to contact us is through our online form
