A sensory information processing system using neural networks
A sensory information processing system using neural networks
复制标题
使用神经网络的感觉信息处理系统
DOI：
10.1109/icnn.1993.298632
复制
发表时间：
1993
期刊：
影响因子：
--
通讯作者：
S. Nagata
中科院分区：
文献类型：
--
作者：
D. Masumoto;T. Kimoto;S. Nagata
为了执行特定于目标的动作，机器人处理感觉信息，即，它将感知数据转换为内部表示。在某些情况下，机器人的内部表示不能从传感数据中唯一地确定。提出了一种克服这一不适定问题的感官信息处理系统的体系结构。该系统使用人工神经网络，该网络经过训练将内部表示转换为感官数据。将迭代方案应用于网络，可以确定唯一的内部表示。该方案将网络的输出(感测数据)与感测数据进行比较，并通过通过各层反向传播差异来更新输入(内部表示)，该输入本可以基于梯度下降方法创建应用的输出(感测数据)。通过根据系统自身运动的意图来预测结果状态，可以提高感官信息处理的准确性和速度。给出了三维目标识别的仿真结果。
In order to carry out actions particular to the goals, a robot processes sensory information, that is, it transforms sensed data to internal representation. In some cases, the robot's internal representation cannot be determined uniquely from the sensed data. An architecture is proposed for a sensory information processing system that overcomes this ill-posed problem. The system uses an artificial neural network which is trained to transform internal representation to sensory data. Applying an iterative scheme to the network, the unique internal representation can be determined. The scheme compares the network's output (sensory data) with the sensed data, and by backpropagating the difference through the layers updates an input (internal representation) which could have created the applied output (sensed data) based on the gradient descent method. By predicting the resulting state based on the intention of the system's own movement, the accuracy and speed of sensory information processing can be improved. Simulation results for three-dimensional object recognition are given.<<ETX>>