Evolution of Controllers from a High-Level Simulator to a High DOF Robot

Authors:
Greg Hornby;S. Takamura;O. Hanagata;M. Fujita;Jordan B. Pollack
Affiliations:
-;-;-;-;-
Venue:
ICES '00 Proceedings of the Third International Conference on Evolvable Systems: From Biology to Hardware
Year:
2000

Citing 4
Cited 8

A qualitative dynamical analysis of evolved locomotion controllers

Proceedings of the second international conference on From animals to animats 2 : simulation of adaptive behavior: simulation of adaptive behavior
Development of an Autonomous Quadruped Robot for Robot Entertainment

Autonomous Robots - Special issue on autonomous agents
An Artificial Life Approach for the Synthesis of Autonomous Agents

AE '95 Selected Papers from the European conference on Artificial Evolution
Evolving mobile robots in simulated and real environments

Artificial Life

Evolving neural networks

Proceedings of the 12th annual conference companion on Genetic and evolutionary computation
Evolving neural networks

Proceedings of the 13th annual conference companion on Genetic and evolutionary computation
Pruning neural networks for a two-link robot control system

IWANN'05 Proceedings of the 8th international conference on Artificial Neural Networks: computational Intelligence and Bioinspired Systems
Highly modular architecture for the general control of autonomous robots

IWANN'05 Proceedings of the 8th international conference on Artificial Neural Networks: computational Intelligence and Bioinspired Systems
Evolving controllers for high-level applications on a service robot: a case study with exhibition visitor flow control

Genetic Programming and Evolvable Machines
Evolving neural networks

Proceedings of the 14th annual conference companion on Genetic and evolutionary computation
Single-unit pattern generators for quadruped locomotion

Proceedings of the 15th annual conference on Genetic and evolutionary computation
Evolving neural networks

Proceedings of the 15th annual conference companion on Genetic and evolutionary computation

Quantified Score

Hi-index	0.02

Visualization

Abstract

Building a simulator for a robot with many degrees of freedom and various sensors, such as Sony's AIBO, is a daunting task. Our implementation does not simulate raw sensor values or actuator commands, rather we model an intermediate software layer which passes processed sensor data to the controller and receives high-level control commands. This allows us to construct a simulator that runs at over 11000 times faster than real time. Using our simulator we evolve a ball-chasing behavior that successfully transfers to an actual AIBO.