Related Experiment Video
Updated: Dec 26, 2025
11:18
Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
Published on: March 2, 2015
10.7K
Spatial Properties of STDP in a Self-Learning Spiking Neural Network Enable Controlling a Mobile Robot
Sergey A Lobov1,2, Alexey N Mikhaylov1, Maxim Shamshin1
1Neurotechnology Department, Lobachevsky State University of Nizhny Novgorod, Nizhny Novgorod, Russia.
Frontiers in Neuroscience
|March 17, 2020
Summary
This study introduces a spiking neural network (SNN) that enables a robot to learn and adapt using spike-timing-dependent plasticity (STDP). The SNN allows the robot to exhibit classical and operant conditioning, demonstrating robust learning capabilities.
Area of Science:
- Computational Neuroscience
- Artificial Intelligence
- Robotics
Background:
- Spiking neural networks (SNNs) offer greater computational potential than traditional artificial neural networks (ANNs).
- Developing robust learning algorithms for SNNs to control mobile robots is a significant challenge.
- Biological neural networks inspire SNNs, aiming for more advanced computational capabilities.
Purpose of the Study:
- To propose a simple SNN with spike-timing-dependent plasticity (STDP) for mobile robot control.
- To demonstrate associative learning and conditioning in a robot using SNNs.
- To explore the potential of SNNs for creating adaptive, learning robotic systems.
Main Methods:
- Implementation of a simple SNN incorporating a Hebbian learning rule in the form of STDP.
- Utilizing the spatial properties of STDP for associative learning.
- Controlling a LEGO robot with the developed SNN to observe learning behaviors.
Main Results:
- The SNN-controlled robot successfully exhibited classical and operant conditioning.
- Competition between spike-conducting pathways in the SNN was crucial for forming and updating neural associations.
- The robot demonstrated the ability to relearn associations when environmental stimuli changed.
Conclusions:
- The proposed SNN with STDP enables associative learning and adaptive behavior in mobile robots.
- The SNN facilitates relearning by replacing irrelevant associations, mimicking biological adaptability.
- Future work can involve testing this SNN in neuronal cultures and exploring memristive hardware implementations.
