Related Experiment Video
Updated: Mar 7, 2026
11:18
Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
Published on: March 2, 2015
10.9K
Connecting Artificial Brains to Robots in a Comprehensive Simulation Framework: The Neurorobotics Platform
Egidio Falotico1, Lorenzo Vannucci1, Alessandro Ambrosano1
1The BioRobotics Institute, Scuola Superiore Sant'Anna , Pontedera , Italy.
Frontiers in Neurorobotics
|February 10, 2017
Summary
The Neurorobotics Platform enables brain model validation through simulated environments and robots. This tool facilitates neurorobotic research by connecting complex neural networks with robotic systems for in silico experiments.
Area of Science:
- Neuroscience
- Computer Science
- Robotics
- Computational Biology
Background:
- Biologically realistic brain models based on spiking neural networks require embodiment in dynamic sensory environments for validation.
- Current brain models, due to complexity, cannot operate in real-time, necessitating simulated environments for realistic tasks.
- A gap exists in tools that facilitate communication between simulated brain models and simulated robotic bodies.
Purpose of the Study:
- To introduce the Neurorobotics Platform, a web-based environment designed to bridge the gap between brain models and robotic simulations.
- To provide scientists and developers with a software infrastructure for connecting spiking neural network models to simulated robots and environments.
- To enable in silico experimentation with neurorobotic systems, simplifying workflows and reducing programming requirements.
Main Methods:
- Development of a web-based Neurorobotics Platform with editors for experimental sequences, environments, robots, and brain-body connectors.
- Integration of existing robot and environment simulations within the platform.
- Testing the platform's capabilities with three example experiments: Braitenberg task, sensory-motor learning, and visual tracking.
Main Results:
- The Neurorobotics Platform architecture has been developed and its first release is presented.
- The platform successfully enables the design and execution of basic neurorobotic experiments using simulated components.
- Example experiments demonstrate the platform's applicability for both robotic tasks and neuroscientific investigations.
Conclusions:
- The Neurorobotics Platform effectively addresses the need for a unified tool to connect brain models with simulated robotic systems.
- It facilitates in silico validation of spiking neural network models in realistic, albeit simulated, sensory-motor tasks.
- The platform supports diverse neurorobotic research, from basic robotic control to complex neuroscientific experiments.
