Neural Information Processing
Neural Information Processing
复制标题
神经信息处理
DOI：
10.1007/978-3-319-12643-2_68
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
Adams S
中科院分区：
文献类型：
--
作者：
Adams S
神经形态硬件和认知机器人似乎是一个明显的适合，但迄今为止的进展一直受到挫折，因为在实现有用的现实世界行为方面缺乏切实的进展。系统限制：神经形态和机器人平台的简单和通常专有的性质，往往是根本的障碍。在这里，我们提出了一个集成的成熟的“神经模拟”芯片，SpiNNaker，与人形iCub机器人使用直接AER -地址事件表示-接口，克服了需要复杂的专有协议通过发送信息UDP编码的尖峰通过以太网链路。使用现有的视觉对象选择设计的神经模型，我们使机器人能够执行一个现实世界的任务：固定在选定的刺激的注意力。结果表明，接口和模型的有效性，能够控制机器人对刺激特定的对象选择。使用SpiNNaker作为可嵌入的神经形态设备说明了两个设计特征在未来神经机器人中的重要性：通用可配置性，允许芯片符合机器人的要求，而不是相反，以及标准接口，消除了连接器，电缆，信号电压和协议的低层次问题。虽然这项研究只是实现这一目标的一个基石，但iCub-SpiNNaker系统展示了一条由神经网络芯片控制的机器人实现有意义行为的道路。
Neuromorphic hardware and cognitive robots seem like an obvious fit, yet progress to date has been frustrated by a lack of tangible progress in achieving useful real-world behaviour. System limitations: the simple and usually proprietary nature of neuromorphic and robotic platforms, have often been the fundamental barrier. Here we present an integration of a mature “neuromimetic” chip, SpiNNaker, with the humanoid iCub robot using a direct AER - address-event representation - interface that overcomes the need for complex proprietary protocols by sending information as UDP-encoded spikes over an Ethernet link. Using an existing neural model devised for visual object selection, we enable the robot to perform a real-world task: fixating attention upon a selected stimulus. Results demonstrate the effectiveness of interface and model in being able to control the robot towards stimulus-specific object selection. Using SpiNNaker as an embeddable neuromorphic device illustrates the importance of two design features in a prospective neurorobot:universal configurabilitythat allows the chip to be conformed to the requirements of the robot rather than the other way ’round, andstandard interfacesthat eliminate difficult low-level issues of connectors, cabling, signal voltages, and protocols. While this study is only a building block towards that goal, the iCub-SpiNNaker system demonstrates a path towards meaningful behaviour in robots controlled by neural network chips.