An evolutionary algorithm for the creation of recurrent network structures is presented. The aim is to develop neural networks controlling the behaviour of miniature robots. Two neuro-modules are created separately using this evolutionary approach. The first neuromodule gives the agents the ability to move within an environment without colliding with obstacles. The second neuro-module provides the agents with a phototropic behaviour. The interaction of the neuromodules is then investigated evolving the necessary interface to provide the agents with a coherent obstacle avoidance and phototropic behaviour. The evolution process is carried out in a simulated environment and individuals with high performance are also tested on a physical environment with the use of Khepera robots.
[1]
Frank Pasemann.
Structure and dynamics of recurrent neuromodules
,
1997
.
[2]
F. Pasemann.
DYNAMICS OF A SINGLE MODEL NEURON
,
1993
.
[3]
Rolf Pfeifer,et al.
Understanding intelligence
,
2020,
Inequality by Design.
[4]
Francesco Mondada,et al.
Mobile Robot Miniaturisation: A Tool for Investigation in Control Algorithms
,
1993,
ISER.
[5]
F. Pasemann.
Evolving neurocontrollers for balancing an inverted pendulum.
,
1998,
Network.
[6]
Jeffrey L. Krichmar,et al.
Evolutionary robotics: The biology, intelligence, and technology of self-organizing machines
,
2001,
Complex..