DeepPicarMicro: Applying TinyML to Autonomous Cyber Physical Systems
DeepPicarMicro: Applying TinyML to Autonomous Cyber Physical Systems
复制标题
DOI：
10.1109/rtcsa55878.2022.00019
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
M. Bechtel;QiTao Weng;H. Yun
中科院分区：
文献类型：
--
作者：
M. Bechtel;QiTao Weng;H. Yun
在微型微控制器单元(MCU)上运行深度神经网络(DNN)是具有挑战性的，因为它们在计算、存储和存储容量方面受到限制。幸运的是，MCU硬件和机器学习软件框架的最新进展使在现代MCU上运行相当复杂的神经网络成为可能，从而产生了一个被广泛称为TinyML的新研究领域。然而，很少有研究表明TinyML在网络物理系统(CPS)中的应用潜力。本文介绍了一种小型无人驾驶RC汽车试验台DeepPicarMicro，它在Raspberry Pi Pico MCU上运行卷积神经网络(CNN)。我们应用了一种最先进的DNN优化方法，以成功地在MCU上适应著名的PilotNet CNN架构，该架构被用于驾驶NVIDIA的真正的自动驾驶汽车。我们应用最先进的网络架构搜索(NAS)方法来进一步优化网络，以端到端的方式有效地实时控制汽车。从广泛的系统实验评估研究中，我们观察到系统的准确性、延迟和控制性能之间存在有趣的关系。在此基础上，我们提出了一种联合优化策略，该策略在人工智能支持的CP的网络结构搜索过程中兼顾了模型的准确性和时延。
Running deep neural networks (DNNs) on tiny Micro-controller Units (MCUs) is challenging due to their limitations in computing, memory, and storage capacity. Fortunately, recent advances in both MCU hardware and machine learning software frameworks make it possible to run fairly complex neural networks on modern MCUs, resulting in a new field of study widely known as TinyML. However, there have been few studies to show the potential for TinyML applications in cyber physical systems (CPS).In this paper, we present DeepPicarMicro, a small self-driving RC car testbed, which runs a convolutional neural network (CNN) on a Raspberry Pi Pico MCU. We apply a state-of-the-art DNN optimization to successfully fit the well-known PilotNet CNN architecture, which was used to drive NVIDIA’s real self-driving car, on the MCU. We apply a state-of-art network architecture search (NAS) approach to find further optimized networks that can effectively control the car in real-time in an end-to-end manner. From an extensive systematic experimental evaluation study, we observe an interesting relationship between the accuracy, latency, and control performance of a system. From this, we propose a joint optimization strategy that takes both accuracy and latency of a model in the network architecture search process for AI enabled CPS.