{"references": ["S. Fujii and T. Nakamura, \"Development of an amphibious hexapod robot\nbased on a water strider,\" in Proc. 10th International Conference on\nClimbing and Walking Robots, pp. 135-143, 2007.", "A. J. Ijspeert, A. Crespi, D. Ryczko, and J. M. Cabelguen, \"From\nswimming to walking with a salamander robot driven by a spinal cord\nmodel,\" Science, vol. 315, pp. 1416-1420, 2007.", "R. A. Brooks, \"A robot that walks: emergent behaviors from a carefully\nevolved network,\" Neural Computation, vol. 12, pp. 253-262, 1989.", "H. Cruse, T. Kindermann, M. Schumm, J. Dean, and J. Schmitz,\n\"Walknet-A biologically inspired network to control six-legged walking,\"\nNeural Networks, vol. 11, pp. 1435-1447, 1998.", "H. Kimura, Y. Fukuoka, and A. H. Cohen, \"Adaptive dynamic walking\nof a quadruped robot on natural ground based on biological concepts,\"\nInternational Journal of Robotics Research, vol. 26, pp. 475-490, 2007.", "R. C. Arkin, K. Ali, A. Weitzenfeld, and F. Cervantes-Perez, \"Behavior\nmodels of the praying matis as a basis for robotic behavior,\" Robotics\nand Autonomous Systems, vol. 32, pp. 39-60, 2000.", "W. G. Walter, The Living Brain, New York: Norton, 1953.", "P. Manoonpong, Neural Preprocessing and Control of Reactive Walking\nMachines: Towards Versatile Artificial Perception-Action Systems, Cognitive\nTechnologies, Springer, 2007.", "P. Manoonpong, F. Pasemann, and F. Woergoetter, \"Sensor-driven neural\ncontrol for omnidirectional locomotion and versatile reactive behaviors\nof walking machines,\" Robotics and Autonomous Systems,\ndoi:10.1016/j.robot.2007.07.004, 2007, in press.\n[10] F. Pasemann, M. Huelse, and K. Zahedi, \"Evolved neurodynamics\nfor robot control,\" in Proc. European Symposium on Artificial Neural\nNetworks, vol. 2, pp. 439-444, 2003.\n[11] F. Pasemann, \"Discrete dynamics of two neuron networks,\" Open\nSystems and Information Dynamics, vol. 2, pp. 49-66, 1993.\n[12] M. Huelse, S. Wischmann, and F. Pasemann, \"The role of non-linearity\nfor evolved multifunctional robot behavior,\" in Proc. 6th International\nConference on Evolvable Systems-ICES 2005, LNCS vol. 3637, pp. 108-\n118, 2005.\n[13] S. L. Hooper, \"Central pattern generators,\" Current Biology, vol. 10, pp.\nR176-R177, 2000.\n[14] P. Manoonpong, F. Pasemann, J. Fischer, and H. Roth, \"Neural processing\nof auditory signals and modular neural control for sound tropism of\nwalking machines,\" International Journal of Advanced Robotic Systems\n(ARS), vol. 2, no. 3, pp. 223-234."]}

This paper describes reactive neural control used to generate phototaxis and obstacle avoidance behavior of walking machines. It utilizes discrete-time neurodynamics and consists of two main neural modules: neural preprocessing and modular neural control. The neural preprocessing network acts as a sensory fusion unit. It filters sensory noise and shapes sensory data to drive the corresponding reactive behavior. On the other hand, modular neural control based on a central pattern generator is applied for locomotion of walking machines. It coordinates leg movements and can generate omnidirectional walking. As a result, through a sensorimotor loop this reactive neural controller enables the machines to explore a dynamic environment by avoiding obstacles, turn toward a light source, and then stop near to it.