Nengo is a software package for building and simulating large-scale neural systems.
Nengo is a complete neural simulator, whether you like yours with or without spikes, with or without deep learning, and with or without online learning. But, Nengo’s sweet spot is anything with dynamics and spikes.
Nengo is unique in its ability to put your neural application on a wide variety of hardware – including the latest neuromorphic hardware. Compile your model to run on CPUs, GPUs, FPGAs, supercomputers running MPI, the SpiNNaker neuromorphic chip, or the latest Loihi neuromorphic chip from Intel.
The best place to learn more about Nengo is our dedicated Nengo website.
Use Nengo to:
Among other things, Nengo has been used to implement motor control, visual attention, serial recall, action selection, working memory, attractor networks, inductive reasoning, path integration, and planning with problem solving. Nengo was used to create a continuing research project, Spaun, a model which combines many of the previously listed models in a single comprehensive whole. Spaun has been featured in hundreds of media stories from around the world.
Nengo is free for individuals and academic research. For commercial licenses, please contact us.
ABRain board allows you to own a piece of the neuromorphic revolution.
ABRain board implements a learning, spiking neural network that you can directly program from within Nengo.
This inexpensive board is aimed at introducing neuromorphics to a broad audience. Explore and exploit the advantages of neuromorphic computing in the classroom, the innovation lab, or your basement.
Use Nengo’s friendly graphical interface to run parts of your standard Nengo model on power-efficient and fast neuromorphic hardware. This self-contained board outperforms most laptops in terms of both speed and efficiency.
Start by immediately running any of the many demos that come with the board out of the box: adaptive controllers, visual classifiers, voice recognizers, and many more.
To purchase ABRain board, contact us.
Our novel controller uses machine learning and artificial intelligence methods to improve standard industrial control.
These patented techniques build on decades of research from MIT into efficient methods for building robust and adaptive controllers for nonlinear systems.
Regardless of whether you’re using position, force, or any other control target, our addition of a neural network into the controller is mathematically guaranteed to improve the control outcome compared to even a well-tuned PID controller.
Furthermore, if desired, the controller will continue to adapt while in use, to account for unknown or expected deviations from the original tuning conditions, whether that’s because of wear and tear on the plant, or a change in environmental dynamics.
Best of all we can implement the entire controller on state-of-the-art, efficient neuromorphic hardware (or standard hardware, if you already have it).
If you’re not sure that this will work for your application, contact us and we’ll help you get started, or work closely with you to develop the controller to suit your criteria.
For pricing and further information, contact us.