Kagan Tumer
ABSTRACT
This paper addresses the evolution of control strategies for a collective: a set of entities that collectively strives to maximize a global evaluation function that rates the performance of the full system. Directly addressing such problems by having a population of collectives and applying the evolutionary algorithm to that population is appealing, but the search space is prohibitively large in most cases. Instead, we focus on evolving control policies for each member of the collective. The main difficulty with this approach is creating an evaluation function for each member of the collective that is both aligned with the global evaluation function and sensitive to the fitness changes of the member. We show how to construct evaluation functions in dynamic, noisy and communication-limited collective environments. On a rover coordination problem, a control policy evolved using aligned and member-sensitive evaluations outperforms global evaluation methods by up to 400%. More notably, in the presence of a larger number of rovers or rovers with noisy and communication limited sensors, the improvements due to the proposed method become significantly more pronounced.
- A. Agah and G. A. Bekey. A genetic algorithm-based controller for decentralized multi-agent robotic systems. In In Proc. of the IEEE Int. Conf. of Evolutionary Computing, Nagoya, Japan, 1996.Google Scholar
Cross Ref
- A. Agogino, K. Stanley, and R. Miikkulainen. Online interactive neuro-evolution. Neural Processing Letters, 11:29--38, 2000. Google ScholarDigital Library
- A. Agogino and K. Tumer. Efficient evaluation functions for multi-rover systems. In The Genetic and Evolutionary Computation Conference, pages 1--12, Seatle, WA, June 2004.Google ScholarCross Ref
- G. Baldassarre, S. Nolfi, and D. Parisi. Evolving mobile robots able to display collective behavior. Artificial Life, pages 9: 255--267, 2003. Google ScholarDigital Library
- M. Dorigo and L. M. Gambardella. Ant colony systems: A cooperative learning approach to the travelling salesman problem. IEEE Transactions on Evolutionary Computation, 1(1):53--66, 1997. Google ScholarDigital Library
- S. Farritor and S. Dubowsky. Planning methodology for planetary robotic exploration. In ASME Journal of Dynamic Systems, Measurement and Control, volume 124, pages 4: 698--701, 2002.Google Scholar
- D. Floreano and F. Mondada. Automatic creation of an autonomous agent: Genetic evolution of a neural-network driven robot. In Proc. of Conf. on Simulation of Adaptive Behavior, 1994. Google ScholarDigital Library
- F. Gomez and R. Miikkulainen. Active guidance for a finless rocket through neuroevolution. In Proceedings of the Genetic and Evolutionary Computation Conference, Chicago, Illinois, 2003. Google ScholarDigital Library
- F. Hoffmann, T.-J. Koo, and O. Shakernia. Evolutionary design of a helicopter autopilot. In Advances in Soft Computing - Engineering Design and Manufacturing, Part 3: Intelligent Control, pages 201--214, 1999.Google Scholar
- A. Martinoli, A. J. Ijspeert, and F. Mondala. Understanding collective aggregation mechanisms: From probabilistic modelling to experiments with real robots. Robotics and Autonomous Systems, 1999.Google ScholarCross Ref
- M. J. Mataric. Coordination and learning in multi-robot systems. In IEEE Intelligent Systems, pages 6--8, March 1998. Google ScholarDigital Library
- K. Stanley and R. Miikkulainen. Efficient reinforcement learning through evolving neural network topologies. In GECCO-2002, San Francisco, CA, 2002. Google ScholarDigital Library
- K. Tumer and A. Agogino. Overcoming communication restrictions in collectives. In Proceedings of the International Joint Conference on Neural Networks, Budapest, Hungary, July 2004.Google ScholarCross Ref
- K. Tumer and D. Wolpert, eds. Collectives and the Design of Complex Systems. Springer, New York, 2004. Google ScholarDigital Library
- K. Tumer and D. Wolpert. A survey of collectives. In Collectives and the Design of Complex Systems, pages 1,42. Springer, 2004.Google ScholarDigital Library
- K. Tumer and D. H. Wolpert. Collective intelligence and Braess' paradox. In Proceedings of the Seventeenth National Conference on Artificial Intelligence, pages 104--109, Austin, TX, 2000. Google ScholarDigital Library
- D. Whitley, F. Gruau, and L. Pyeatt. Cellular encoding applied to neurocontrol. In International Conference on Genetic Algorithms, 1995. Google ScholarDigital Library
- D. H. Wolpert and K. Tumer. Optimal payoff functions for members of collectives. Advances in Complex Systems, 4(2/3):265--279, 2001.Google ScholarCross Ref
Index Terms
- Coordinating multi-rover systems: evaluation functions for dynamic and noisy environments
Recommendations
Distributed evaluation functions for fault tolerant multi-rover systems
GECCO '06: Proceedings of the 8th annual conference on Genetic and evolutionary computationThe ability to evolve fault tolerant control strategies for large collections of agents is critical to the successful application of evolutionary strategies to domains where failures are common. Furthermore, while evolutionary algorithms have been ...
Selection of Radial Basis Functions via Genetic Algorithms in Pattern Recognition Problems
SBRN '08: Proceedings of the 2008 10th Brazilian Symposium on Neural NetworksThe mixed use of different shapes of radial basis functions (RBFs) in RBF Networks is investigated in this paper. For this purpose, we propose the use of the q-Gaussian function, which reproduces different RBFs by changing a real parameter q, in RBF ...
A surrogate-assisted selection scheme for genetic algorithms employing multi-layer neural networks
GECCO '18: Proceedings of the Genetic and Evolutionary Computation Conference CompanionIn this paper, we propose a simple yet effective approach in surrogate-assisted genetic algorithms employing a neural network to estimate survival probabilities of individuals in selections to reduce computational cost of their fitness evaluations. A ...
Comments