Robust Trajectory Generation for Robotic Control on the Neuromorphic Research Chip Loihi.
Robust Trajectory Generation for Robotic Control on the Neuromorphic Research Chip Loihi.
复制标题
DOI：
10.3389/fnbot.2020.589532
复制
发表时间：
2020
影响因子：
3.1
通讯作者：
Tetzlaff C
中科院分区：
文献类型：
--
作者：
Michaelis C;Lehr AB;Tetzlaff C
神经形态硬件有几个有前途的优势相比，冯诺依曼架构，是非常有趣的机器人控制。然而，尽管神经形态计算的速度和能量效率高，但在控制场景中利用这种硬件的算法仍然很少。一个问题是从硬件上的快速尖峰活动（其作用于几毫秒的时间尺度上）到数百毫秒量级的控制相关时间尺度的转变。另一个问题是复杂轨迹的执行，这需要尖峰活动包含足够的可变性，同时，为了可靠的性能，网络动态必须对噪声具有足够的鲁棒性。在这项研究中，我们利用最近开发的生物启发尖峰神经网络模型，所谓的各向异性网络。我们使用英特尔的神经形态研究芯片Loihi识别并将各向异性网络的核心原理转移到神经形态硬件，并在机器人手臂执行的电机控制任务的轨迹上验证了系统。我们开发了一种网络架构，包括各向异性网络和池化层，该网络架构允许从芯片快速读取尖峰信号并执行固有的正则化。通过这一点，我们证明了Loihi上的各向异性网络可靠地编码了神经活动的序列模式，每个模式代表一个机器人动作，并且这些模式允许在控制相关的时间尺度上生成多维轨迹。综上所述，我们的研究提出了一种新的算法，允许生成复杂的机器人运动，作为使用最先进的神经形态硬件进行机器人控制的构建块。
Neuromorphic hardware has several promising advantages compared to von Neumann architectures and is highly interesting for robot control. However, despite the high speed and energy efficiency of neuromorphic computing, algorithms utilizing this hardware in control scenarios are still rare. One problem is the transition from fast spiking activity on the hardware, which acts on a timescale of a few milliseconds, to a control-relevant timescale on the order of hundreds of milliseconds. Another problem is the execution of complex trajectories, which requires spiking activity to contain sufficient variability, while at the same time, for reliable performance, network dynamics must be adequately robust against noise. In this study we exploit a recently developed biologically-inspired spiking neural network model, the so-called anisotropic network. We identified and transferred the core principles of the anisotropic network to neuromorphic hardware using Intel's neuromorphic research chip Loihi and validated the system on trajectories from a motor-control task performed by a robot arm. We developed a network architecture including the anisotropic network and a pooling layer which allows fast spike read-out from the chip and performs an inherent regularization. With this, we show that the anisotropic network on Loihi reliably encodes sequential patterns of neural activity, each representing a robotic action, and that the patterns allow the generation of multidimensional trajectories on control-relevant timescales. Taken together, our study presents a new algorithm that allows the generation of complex robotic movements as a building block for robotic control using state of the art neuromorphic hardware.
登录
查看更多内容
影响因子：
4.3
作者：
Maes, Amadeus;Barahona, Mauricio;Clopath, Claudia
通讯作者：
Clopath, Claudia
影响因子：
25
作者：
Laje, Rodrigo;Buonomano, Dean V.
通讯作者：
Buonomano, Dean V.
影响因子：
20.6
作者：
Furber, Steve B.;Galluppi, Francesco;Plana, Luis A.
通讯作者：
Plana, Luis A.
影响因子：
7.7
作者：
Murray, James M.;Escola, G. Sean
通讯作者：
Escola, G. Sean
影响因子：
64.8
作者：
通讯作者：
--