The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics
Authors:
- 1 February 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 17 (1) , 122-145
Abstract
The reality gap, which often makes controllers evolved in simulation inefficient once transferred onto the physical robot, remains a critical issue in evolutionary robotics (ER). We hypothesize that this gap highlights a conflict between the efficiency of the solutions in simulation and their transferability from simulation to reality: the most efficient solutions in simulation often exploit badly modeled phenomena to achieve high fitness values with unrealistic behaviors. This hypothesis leads to the transferability approach, a multiobjective formulation of ER in which two main objectives are optimized via a Pareto-based multiobjective evolutionary algorithm: 1) the fitness; and 2) the transferability, estimated by a simulation-to-reality (STR) disparity measure. To evaluate this second objective, a surrogate model of the exact STR disparity is built during the optimization. This transferability approach has been compared to two reality-based optimization methods, a noise-based approach inspired from Jakobi's minimal simulation methodology and a local search approach. It has been validated on two robotic applications: 1) a navigation task with an e-puck robot; and 2) a walking task with a 8-DOF quadrupedal robot. For both experimental setups, our approach successfully finds efficient and well-transferable controllers only with about ten experiments on the physical robot.Keywords
This publication has 49 references indexed in Scilit:
- Artificial evolution of the morphology and kinematics in a flapping-wing mini-UAVBioinspiration & Biomimetics, 2007
- Resilient Machines Through Continuous Self-ModelingScience, 2006
- Visual odometry for ground vehicle applicationsJournal of Field Robotics, 2006
- Nonlinear System Identification Using Coevolution of Models and TestsIEEE Transactions on Evolutionary Computation, 2005
- Map Quality for Ordinary Kriging and Inverse Distance Weighted InterpolationSoil Science Society of America Journal, 2004
- A comprehensive survey of fitness approximation in evolutionary computationSoft Computing, 2003
- Structure optimization of neural networks for evolutionary design optimizationSoft Computing, 2003
- Explorations in design space: unconventional electronics design through artificial evolutionIEEE Transactions on Evolutionary Computation, 1999
- Evolutionary Robotics and the Radical Envelope-of-Noise HypothesisAdaptive Behavior, 1997
- Inertial navigation systems for mobile robotsIEEE Transactions on Robotics and Automation, 1995