Chen Zhanfei, Wang Xiaoping, Wang Zilu, Yang Chao, Huang Tingwen, Lai Jingang, Zeng Zhigang
IEEE Trans Biomed Circuits Syst. 2025 Jun;19(3):686-698. doi: 10.1109/TBCAS.2024.3480272.
Cognitive navigation, a high-level and crucial function for organisms' survival in nature, enables autonomous exploration and navigation within the environment. However, most existing works for bio-inspired navigation are implemented with non-neuromorphic computing. This work proposes a bio-inspired memristive spiking neural network (SNN) circuit for goal-oriented navigation, capable of online decision-making through reward-based learning. The circuit comprises three primary modules. The place cell module encodes the agent's spatial position in real-time through Poisson spiking; the action cell module determines the direction of subsequent movement; and the reward-based learning module provides a bio-inspired learning method adaptive to delayed and sparse rewards. To facilitate practical application, the entire SNN is quantized and deployed on a real memristive hardware platform, achieving about a 21$\times$ reduction in energy consumption compared to a typical digital acceleration system in the forward computing phase. This work offers an implementation idea of neuromorphic solution for robotic navigation application in low-power scenarios.
认知导航是生物体在自然环境中生存的一项高级且关键的功能，它能够使生物体在环境中进行自主探索和导航。然而，大多数现有的受生物启发的导航研究工作都是通过非神经形态计算来实现的。本文提出了一种用于目标导向导航的受生物启发的忆阻脉冲神经网络（SNN）电路，该电路能够通过基于奖励的学习进行在线决策。该电路由三个主要模块组成。位置细胞模块通过泊松脉冲实时编码智能体的空间位置；动作细胞模块确定后续运动的方向；基于奖励的学习模块提供一种适应延迟和稀疏奖励的受生物启发的学习方法。为便于实际应用，整个SNN被量化并部署在一个实际的忆阻硬件平台上，在前向计算阶段，与典型的数字加速系统相比，能耗降低了约21倍。这项工作为低功耗场景下的机器人导航应用提供了一种神经形态解决方案的实现思路。
