基于双神经网络结构的发展型机器人3D抓取

Developmental Robotic 3D Reaching Based on Double-Network Architecture

手眼协调属于感觉运动的协调,是人认知能力的基础。这篇论文的目标是要模仿人类婴儿的发育过程来建立一个机器人系统,可以接近或者抓取物体。该文旨在应用发展模型中的"增量过程",通过使用双神经网络在3D实验环境下来完成机器人抓取系统。这个机器人系统在全约束条件下开始发展,随着每个约束条件的成熟从而驱动双神经网络的训练,当所有的约束条件都释放,机器人系统就可以获得抓取能力。

Reaching ability is a kind of human sensory motor coordination and is the foundation of human cognitive abilities.The objec.tive of this work is to imitate the developmental progress of human infant to create a robotic system which can reach or capture objects.The work proposes to employ a developmental model"incremental process"to implement such a robotic reaching system by using a dou.ble neural network system within 3D experimental environment.The robotic system starts to develop under fully constrained conditions,and the training of the double neural network is driven by each constraint has been saturated,when the all contained conditions have been overcome,the robotic system is able to own reaching ability.In addition,this paper delicately describes and discuss the experimental results.