Our Neurorobotics Platform, NRP, is a simulation platform that connects brain models (ranging from spiking neural networks to deep networks) to body models. It enables its users to simulate agents interacting in closed-loop with their virtual environment, thus connecting brain dynamics to behavioural outputs. This approach sheds a new light on the brain mechanism involved in the perception-cognition-action loop, and will eventually support the creation of a new generation of robots displaying situational awareness, adaptation to evolving experimental parameters, ability to plan complex sequences of actions, and more.

The NRP is an open access and open source integrative simulation framework that provides opportunities for innovation in both the robotics and neuroscience spaces. It facilitates experiments in cognitive research by implementing experimental workflows that make it possible for researchers to test and compare brain models in embodied settings. The NRP allows its users to create scenarios where a brain model controls a virtual agent interacting with a realistic simulated environment. Testing procedures that can otherwise be difficult to implement on a physical setup - like testing of safety-critical situations or scenarios that could result in damage to the robot - can now be performed in the NRP without incurring any risk to a costly physical plant. By providing extensive control over experimental conditions, scripting capabilities and physically realistic interactions with the simulated environment, the NRP offers opportunities for closed-loop neuroscience that are unmatched by any competing simulation framework.

Roboticists using the NRP can further their ideas of bio-inspired, safe and adaptable robotic systems. Novel robotic controllers can easily be designed and tested on the NRP, paving the way for enhanced cognitive abilities such as contextual awareness and decision-making.

With the NRP, neuroscientists can observe and analyse emergent behavioural patterns in virtual agents controlled by models of brain architecture and functions. This provides the unprecedented ability to test and refine these models iteratively, without the many approximations entailed by simulating the brain decoupled from the body.

Thanks to its direct connection to both Neuroscience research in HBP and the EBRAINS HPC infrastructure, the NRP is the best available platform for scientific exploration and development of neuromorphic computing applications.