NeuRonICS Lab, Department of Electronic Systems Engineering, Indian Institute of Science, Bangalore 560012, India.
Cellular Neurophysiology Laboratory, Molecular Biophysics Unit, Indian Institute of Science, Bangalore 560012, India.
Neural Netw. 2021 Jul;139:45-63. doi: 10.1016/j.neunet.2021.01.028. Epub 2021 Feb 13.
The mammalian spatial navigation system is characterized by an initial divergence of internal representations, with disparate classes of neurons responding to distinct features including location, speed, borders and head direction; an ensuing convergence finally enables navigation and path integration. Here, we report the algorithmic and hardware implementation of biomimetic neural structures encompassing a feed-forward trimodular, multi-layer architecture representing grid-cell, place-cell and decoding modules for navigation. The grid-cell module comprised of neurons that fired in a grid-like pattern, and was built of distinct layers that constituted the dorsoventral span of the medial entorhinal cortex. Each layer was built as an independent continuous attractor network with distinct grid-field spatial scales. The place-cell module comprised of neurons that fired at one or few spatial locations, organized into different clusters based on convergent modular inputs from different grid-cell layers, replicating the gradient in place-field size along the hippocampal dorso-ventral axis. The decoding module, a two-layer neural network that constitutes the convergence of the divergent representations in preceding modules, received inputs from the place-cell module and provided specific coordinates of the navigating object. After vital design optimizations involving all modules, we implemented the tri-modular structure on Zynq Ultrascale+ field-programmable gate array silicon chip, and demonstrated its capacity in precisely estimating the navigational trajectory with minimal overall resource consumption involving a mere 2.92% Look Up Table utilization. Our implementation of a biomimetic, digital spatial navigation system is stable, reliable, reconfigurable, real-time with execution time of about 32 s for 100k input samples (in contrast to 40 minutes on Intel Core i7-7700 CPU with 8 cores clocking at 3.60 GHz) and thus can be deployed for autonomous-robotic navigation without requiring additional sensors.
哺乳动物的空间导航系统的特征是内部表示的初始发散，不同类别的神经元对包括位置、速度、边界和头部方向在内的不同特征做出反应；随后的收敛最终实现了导航和路径整合。在这里，我们报告了仿生神经结构的算法和硬件实现，该结构包含一个前馈三模块、多层架构，代表网格细胞、位置细胞和解码模块，用于导航。网格细胞模块由以网格状模式发射的神经元组成，由构成内侧缰状核背腹跨度的不同层构成。每个层都是作为一个独立的连续吸引器网络构建的，具有不同的网格场空间尺度。位置细胞模块由在一个或少数几个空间位置发射的神经元组成，根据来自不同网格细胞层的模块化输入的收敛，组织成不同的簇，复制了沿着海马体背腹轴的位置场大小梯度。解码模块是一个由两个层组成的神经网络，构成了前向模块发散表示的收敛，接收来自位置细胞模块的输入，并提供导航物体的特定坐标。在涉及所有模块的重要设计优化之后，我们在 Zynq Ultrascale+现场可编程门阵列硅芯片上实现了三模块结构，并展示了其在最小化整体资源消耗的情况下精确估计导航轨迹的能力，仅使用了 2.92%的查找表利用率。我们实现的仿生数字空间导航系统稳定、可靠、可重构、实时，对于 10 万个输入样本的执行时间约为 32 秒（相比之下，在 Intel Core i7-7700 CPU 上使用 8 核时钟频率为 3.60 GHz 的情况下需要 40 分钟），因此可以部署用于自主机器人导航，而无需额外的传感器。
