CPS: Small: Formally Correct Deep Perception For Cyber-Physical Systems

CPS：小：形式上正确的网络物理系统深度感知

• 批准号: 2211146
• 负责人: Paulo Tabuada
• 金额: $ 50万
• 依托单位: University of California-Los Angeles
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-06-15 至 2025-05-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2211146&HistoricalAwards=false
• 关键词: CPS; Small; Formally; Correct; Deep

Light Detection and Ranging (LiDARs) and cameras are an indispensable part of the sensor suite used in autonomous cyber-physical systems such as self-driving cars and unmanned aerial vehicles. The data generated by these sensors is often processed by a deep neural network that transforms it into state estimates used in control loops. Although one can analyze the impact that erroneous state estimates have on control loops, less is known about how to characterize the errors produced by deep neural networks. The objective of this project is to develop analysis and design techniques that provide formally guaranteed bounds on how large these errors can be. Formally establishing error bounds will enable the verification of existing systems as well as the design of new autonomous systems for which formal guarantees of safety and performance can be given.This project addresses the challenge of using deep neural networks in the perception pipeline of autonomous cyber-physical systems by following two different approaches, termed correctness-by-training and correctness-by-supervision. The first approach, correctness-by-training, is based on the use of monotone neural networks for which deterministic generalization bounds can be established. The challenge of using monotone neural networks is that their training is more challenging and several novel training techniques will be investigated. The second approach, correctness-by-supervision, consists of attaching a supervisor to the neural network that overrides the network output so as to enforce guaranteed error bounds. A supervisor will be developed in the context of localization using LiDAR measurements using novel point-set registration techniques based on moments. Both approaches aim to provide guaranteed error bounds on the state estimates computed by deep neural networks. The ultimate contribution is to use these error bounds in the formal analysis of safety and performance of control loops using deep neural networks in the perception pipeline.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

光探测和测距 (LiDAR) 和摄像头是自动驾驶汽车和无人机等自主网络物理系统中使用的传感器套件中不可或缺的一部分。这些传感器生成的数据通常由深度神经网络处理，将其转换为控制回路中使用的状态估计。尽管人们可以分析错误状态估计对控制环路的影响，但人们对如何表征深度神经网络产生的错误知之甚少。该项目的目标是开发分析和设计技术，为这些误差的大小提供正式的保证界限。正式建立错误界限将能够验证现有系统以及新的自主系统的设计，从而可以正式保证安全和性能。该项目解决了在自主网络感知管道中使用深度神经网络的挑战。物理系统通过遵循两种不同的方法，称为训练正确性和监督正确性。第一种方法是训练正确性，基于单调神经网络的使用，可以为其建立确定性泛化边界。使用单调神经网络的挑战在于它们的训练更具挑战性，并且将研究几种新颖的训练技术。第二种方法是监督正确性，包括将监督者附加到神经网络，该监督者会覆盖网络输出，以强制执行有保证的错误界限。将使用基于矩的新颖点集配准技术，在使用 LiDAR 测量进行定位的背景下开发监督程序。这两种方法都旨在为深度神经网络计算的状态估计提供有保证的误差范围。最终的贡献是在感知管道中使用深度神经网络对控制环路的安全性和性能进行正式分析时使用这些误差界限。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力优点和技术进行评估，被认为值得支持。更广泛的影响审查标准。

项目成果

Sharp Performance Bounds for PASTA

PASTA 的急剧性能限制

• DOI: 10.1109/lcsys.2023.3285514
• 发表时间: 2023-01
• 期刊: IEEE Control Systems Letters
• 影响因子: 3
• 作者: Marchi, Matteo;Bunton, Jonathan;Gas, Yskandar;Gharesifard, Bahman;Tabuada, Paulo
• 通讯作者: Tabuada, Paulo

LiDAR Point Cloud Registration with Formal Guarantees

激光雷达点云注册有正式保证

• DOI: 10.1109/cdc51059.2022.9992625
• 发表时间: 2022-12
• 期刊: 2022 IEEE 61st Conference on Decision and Control (CDC
• 作者: Marchi, Matteo;Bunton, Jonathan;Gharesifard, Bahman;Tabuada, Paulo
• 通讯作者: Tabuada, Paulo