Bootstrapped Neuro-Simulation as a method of concurrent neuro-evolution and damage recovery

Abstract Bootstrapped Neuro-Simulation (BNS) is a method of concurrent simulator and robot controller evolution. The algorithm requires little domain knowledge and no pre-investigation data gathering. Additionally, it bridges the reality gap effectively, rapidly evolves functional controllers, and recovers from damage automatically. In this paper, the first evidence of the ability of BNS to evolve closed-loop controllers is shown; in this case to solve a light-following problem. The algorithm is then evaluated for its damage recovery ability for these closed-loop controllers and shown to be very effective, with only minor adaptations.

[1]  Magdalena D. Bugajska,et al.  Challenges and Opportunities of Evolutionary Robotics , 2007, ArXiv.

[2]  Kalyanmoy Deb,et al.  Simulated Binary Crossover for Continuous Search Space , 1995, Complex Syst..

[3]  Mathys C. du Plessis,et al.  Evolving Snake Robot Controllers Using Artificial Neural Networks as an Alternative to a Physics-Based Simulator , 2015, 2015 IEEE Symposium Series on Computational Intelligence.

[4]  Mathys C. du Plessis,et al.  A comparison of neural networks and physics models as motion simulators for simple robotic evolution , 2014, 2014 IEEE Congress on Evolutionary Computation (CEC).

[5]  Charmain Cilliers,et al.  Simulating Robots Without Conventional Physics: A Neural Network Approach , 2013, J. Intell. Robotic Syst..

[6]  Francesco Mondada,et al.  Automatic creation of an autonomous agent: genetic evolution of a neural-network driven robot , 1994 .

[7]  Charmain Cilliers,et al.  A Neural Network-based kinematic and light-perception simulator for simple robotic evolution , 2010, IEEE Congress on Evolutionary Computation.

[8]  Hod Lipson,et al.  Resilient Machines Through Continuous Self-Modeling , 2006, Science.

[9]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[10]  David Beasley,et al.  An overview of genetic algorithms: Part 1 , 1993 .

[11]  Mathys C. du Plessis,et al.  Concurrent controller and Simulator Neural Network development for a differentially-steered robot in Evolutionary Robotics , 2016, Robotics Auton. Syst..

[12]  Mathys C. du Plessis,et al.  Concurrent controller and Simulator Neural Network development for a snake-like robot in Evolutionary Robotics , 2017, Robotics Auton. Syst..

[13]  Jean-Baptiste Mouret,et al.  Reset-free Trial-and-Error Learning for Robot Damage Recovery , 2016, Robotics Auton. Syst..

[14]  Gary B. Parker,et al.  Co-evolving model parameters for anytime learning in evolutionary robotics , 2000, Robotics Auton. Syst..

[15]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[16]  Josh C. Bongard,et al.  Evolutionary robotics , 2013, CACM.

[17]  Hod Lipson,et al.  Automatic Design and Manufacture of Soft Robots , 2012, IEEE Transactions on Robotics.

[18]  Elliot Meyerson,et al.  Evolving Deep Neural Networks , 2017, Artificial Intelligence in the Age of Neural Networks and Brain Computing.

[19]  Betty H. C. Cheng,et al.  An evolutionary approach to discovering execution mode boundaries for adaptive controllers , 2016, 2016 IEEE Symposium Series on Computational Intelligence (SSCI).

[20]  James G Bellingham,et al.  Robotics in Remote and Hostile Environments , 2007, Science.

[21]  Antoine Cully,et al.  Robots that can adapt like animals , 2014, Nature.

[22]  Charmain Cilliers,et al.  Towards an artificial neural network-based simulator for behavioural evolution in evolutionary robotics , 2009, SAICSIT '09.

[23]  Javier Ruiz-del-Solar,et al.  Combining Simulation and Reality in Evolutionary Robotics , 2007, J. Intell. Robotic Syst..

[24]  Katharine Sanderson,et al.  Mars rover Spirit (2003–10) , 2010, Nature.

[25]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[26]  Jürgen Schmidhuber,et al.  LSTM can Solve Hard Long Time Lag Problems , 1996, NIPS.

[27]  Mathys C. du Plessis,et al.  Neuroevolution of Inverted Pendulum Control: A Comparative Study of Simulation Techniques , 2017, J. Intell. Robotic Syst..

[28]  Stéphane Doncieux,et al.  The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics , 2013, IEEE Transactions on Evolutionary Computation.

[29]  Jean-Baptiste Mouret,et al.  Safety-Aware Robot Damage Recovery Using Constrained Bayesian Optimization and Simulated Priors , 2016, NIPS 2016.

[30]  Luis Perez,et al.  The Effectiveness of Data Augmentation in Image Classification using Deep Learning , 2017, ArXiv.