Evolution of Spiking Neural Controllers for Autonomous Vision-Based Robots

By Dario Floreano, Claudio Mattiussi. 2001

170 citations recorded by the source corpus.

We describe a set of preliminary experiments to evolve spiking neural controllers for a vision-based mobile robot. All the evolutionary experiments are carried out on physical robots without human intervention. After discussing how to implement and interface these neurons with a physical robot, we show that evolution finds relatively quickly functional spiking controllers capable of navigating in irregularly textured environments without hitting obstacles using a very simple genetic encoding and fitness function. Neuroethological analysis of the network activity let us understand the functioning of evolved controllers and tell the relative importance of single neurons independently of their observed firing rate. Finally, a number of systematic lesion experiments indicate that evolved spiking controllers are very robust to synaptic strength decay that typically occurs in hardware implementations of spiking circuits.

Subfields: Cellular and Molecular Neuroscience, Cognitive Neuroscience, Electrical and Electronic Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cellular and Molecular Neuroscience

In Cognitive Neuroscience

In Electrical and Electronic Engineering

Citations in the corpus

Deep learning in neural networks: An overview (2014, background): “Spiking nets were evolved to achieve reasonable performance on small face recognition data sets (Wysoski, Benuskova, & Kasabov, 2010) and to control simple robots (Floreano & Mattiussi, 2001; Hagras, Pounds-Cornish, Colley, Callaghan, & Clarke, 2004).”
Evolving spike-timing-dependent plasticity for single-trial learning in robots (2003, methodology): “This approach marks a di® erence with previous work on the evolution of spiking controllers which have used a neural rate estimation method for driving the motor ( Floreano & Mattiussi 2001 ).”
Evolution of Plastic Learning in Spiking Networks via Memristive Connections (2012, background): “As the subject of the paper describes robotics tasks, a short overview of spiking neuroevolutionary robotics follows.”
Spike-Timing Dependent Plasticity for Evolved Robots (2002, methodology): “Recent work in evolutionary robotics has began to explore the use of spike-based neural controllers (Floreano & Mattiussi, 2001).”
Towards evolving spiking networks with memristive synapses (2011, methodology): “Neuroevolution was first applied to IAF SNNs to evolve networks that produce temporally-dependant outputs [11] and SRM spiking networks were first evolved for a vision-based navigation task [12].”
Neuroevolution: from architectures to learning (2008, background): “From Mattiussi et al. [46] Reil [63, 64] studied the structural properties of such gene networks with a simple computational model.”
Evolving spiking neural network controllers for autonomous robots (2004, methodology): “The state of a spiking neuron is described by the voltage difference across its membrane, also known as membrane potential v [1].”
Evolutionary and Computational Advantages of Neuromodulated Plasticity (2008, methodology): “Several models of SNNs have been experimented on simulated and real robots (Floreano and Mattiussi, 2001; Floreano et al., 2004; Zufferey and Floreano, 2004; Srinivasan and Zhang, 2004; Chahl et al., 2004; Federici, 2005a).”