用脉冲神经网络生成极坐标轨迹，节能高效且可调控方向速度半径。
A Spiking Sequence Generator for Polar Trajectories on Neuromorphic Hardware
- 采用竞争机制与辅助神经群实现轨迹动态控制
- 在SpiNNaker2上实现步时减少2-3个数量级，能耗降低3-4个数量级
- 适合对能效和系统可解释性要求高的嵌入式神经控制场景
面向尺寸、重量和功耗受限系统的类脑控制器需要兼具能效与系统动态可解释性的神经架构。现有方法或依赖端到端训练的脉冲网络导致可解释性差，或采用转换的古典控制器无法充分利用类脑硬件特性。本文提出一种用于生成极坐标轨迹的脉冲神经网络（SNN）架构，采用竞争选一（WTA）结构并引入辅助神经群体以诱导神经活动的可控转换。我们设计了调节这些群体动态的规则，并利用一种分流抑制机制，实现对轨迹方向、速度和半径的独立控制。该网络在SpiNNaker2类脑处理器上实现，与传统计算平台相比，壁钟步时减少2至3个数量级，能耗降低3至4个数量级。
原文摘要 · Abstract (English)
Neuromorphic controllers for size, weight, and power-constrained systems require neural architectures that are both energy-efficient and interpretable at the level of system dynamics. However, existing approaches either rely on end-to-end trained spiking networks with limited interpretability, or on converted classical controllers that fail to fully exploit neuromorphic dynamics. We present a spiking neural network (SNN) architecture for generating polar trajectories, using a winner-take-all (WTA) architecture with accessory populations that induce controlled transitions in neural activity. We demonstrate tuning rules for these population dynamics, and utilize a form of shunting inhibition to enable independent control of direction, speed, and radius of the resulting polar trajectories. We implement the network on the SpiNNaker2 neuromorphic processor, and demonstrate a two to three orders of magnitude reduction in wall-clock step time and three to four orders of magnitude reduction in energy expenditure when compared to conventional computing platforms.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。