A robot vision location based on developmental algorithm of monoamine neurotransmitters modulation is proposed to solve the problem that a large number of neurons need to be allocated in vision location task based on general developmental algorithms. Firstly, the monoamine neurotransmitter theory of dopamine and serotonin controlling a variety of physiological functions in the brain is introduced to realize neural modulation. Then, the developmental algorithm of monoamine neurotransmitters modulation is established based on general developmental algorithms. The robot uses autonomous trial and error strategies to complete the process of reinforcement learning, store "memory", and dynamically change the learning rate, and ultimately it realizes vision location task. Experimental results show that the number of neurons to be allocated in advance in the proposed method is as few as the number of required knowledge concepts, which can significantly reduce the required number of neurons and increase algorithm efficiency.
[1] 陈洋,张道辉,赵新刚,等.基于IHDR自主学习框架的无人机3维路径规划[J].机器人,2012,34(5):513-518. Chen Y, Zhang D H, Zhao X G, et al. UAV 3D path planning based on IHDR autonomous-learning-framework[J]. Robot, 2012, 34(5): 513-518.[2] Gomi T. Evolutionary robotics: From intelligent robots to artificial life[M]. Berlin, Germany: Springer, 1998.[3] Weng J, McClelland J, Pentland A, et al. Artificial intelligence--Autonomous mental development by robots and animals[J]. Science, 2001, 291(5504): 599-600. [4] Baldassarre G, Mannella F, Fiore V G, et al. Intrinsically motivated action-outcome learning and goal-based action recall: A system-level bio-constrained computational model[J]. Neural Networks, 2013, 41(S1): 168-187.[5] Berke J D, Hyman S E. Addiction, dopamine, and the molecular mechanisms of memory[J]. Neuron, 2000, 25(3): 515-532. [6] Cavallaro S. Genomic analysis of serotonin receptors in learning and memory[J]. Behavioural Brain Research, 2008, 195(1): 2-6. [7] Montague P R, Hyman S E, Cohen J D. Computational roles for dopamine in behavioural control[J]. Nature, 2004, 431: 760-767.[8] Cox B R, Krichmar J L. Neuromodulation as a robot controller: A brain-inspired strategy for controlling autonomous robots[J]. IEEE Robotics and Automation Magazine, 2009, 16(3): 72-80. [9] Paslaski S, Vandam C, Weng J. Modeling dopamine and serotonin systems in a visual recognition network[C]//International Joint Conference on Neural Networks. Piscataway, USA: IEEE, 2011: 3016-3023.[10] Ji Z P, Weng J. WWN-2: A biologically inspired neural network for concurrent visual attention and recognition[C]//International Joint Conference on Neural Networks. Piscataway, USA: IEEE, 2010: 1-8.[11] Luciw M, Weng J. Where-what network 3: Developmental top-down attention for multiple foregrounds and complex backgrounds[C]//International Joint Conference on Neural Net-works. Piscataway, USA: IEEE, 2010: 4233-4240.[12] Weng J. Three theorems: Brain-like networks logically reason and optimally generalize[C]//International Joint Conference on Neural Networks. Piscataway, USA: IEEE, 2011: 2983-2990.[13] Weng J. A general purpose brain model for developmental robots: The spatial brain for any temporal lengths[C]//Work-shop on Bio-Inspired Self-Organizing Robotic Systems. Piscataway, USA: IEEE, 2010: 1-6.[14] Weng J, Luciw M. Dually optimal neuronal layers: Lobe component analysis[J]. IEEE Transactions on Autonomous Mental Development, 2009, 1(1): 68-85. [15] Qian K, Song A G, Bao J T, et al. Small teleoperated robot for nuclear radiation and chemical leak detection[J]. International Journal of Advanced Robotic Systems, 2012, 9: No.70.
