In this thesis, we draw inspiration from both classical system identification and modern machine learning in order to solve estimation problems for real-world, physical systems. The main approach to estimation and learning adopted is optimization based. Concepts such as regularization will be utilized for encoding of prior knowledge and basis-function expansions will be used to add nonlinear modeling power while keeping data requirements practical. The thesis covers a wide range of applications, many inspired by applications within robotics, but also extending outside this already wide field. Usage of the proposed methods and algorithms are in many cases illustrated in the real-world applications that motivated the research. Topics covered include dynamics modeling and estimation, model-based reinforcement learning, spectral estimation, friction modeling and state estimation and calibration in robotic machining. In the work on modeling and identification of dynamics, we develop regularization strategies that allow us to incorporate prior domain knowledge into flexible, overparameterized models. We make use of classical control theory to gain insight into training and regularization while using flexible tools from modern deep learning. A particular focus of the work is to allow use of modern methods in scenarios where gathering data is associated with a high cost. In the robotics-inspired parts of the thesis, we develop methods that are practically motivated and ensure that they are implementable also outside the research setting. We demonstrate this by performing experiments in realistic settings and providing open-source implementations of all proposed methods and algorithms.
A Systematic Review of Machine Learning Techniques for Cattle Identification: Datasets, Methods and Future Directions
Charles Sturt University(查尔斯特大学);Food Agility CRC Ltd(Food Agility合作研究中心有限公司);TerraCipher Pty. Ltd.(TerraCipher私人有限公司)
生物安保和食品安全要求的提高可能会增加对供应链中牲畜有效追踪和识别系统的需求。机器学习和计算机视觉的先进技术已应用于精准畜牧管理,包括重大疾病检测、疫苗接种、生产管理、跟踪和健康监测。本文提供了一个系统的文献综述(SLR)基于视觉的牛识别。更具体地说,这个SLR是使用机器学习(ML)和深度学习(DL)识别和分析与牛识别相关的研究。对于牛的检测和牛的识别这两个主要应用,所有基于ML的论文都只解决牛的识别问题。然而,在基于DL的论文中研究了检测和识别问题。根据我们的调查报告,用于牛识别的最常用的ML模型是支持向量机(SVM),k-最近邻(KNN)和人工神经网络(ANN)。卷积神经网络(CNN)、残差网络(ResNet)、Inception、You Only Look Once(YOLO)和Faster R-CNN是所选论文中流行的DL模型。在这些纸张中,最显著的特征是牛的口印和皮毛图案。局部二值模式(LBP),加速鲁棒特征(SURF),尺度不变特征变换(SIFT),以及Inception或CNN被认为是最常用的特征提取方法。
State estimations and noise identifications with intermittent corrupted observations via Bayesian variational inference
School of Mathematical Sciences, Beihang University(北京航空航天大学数学科学学院);Key Laboratory of Mathematics, Informatics and Behavioral Semantics (LMIB), Beihang University(北京航空航天大学数学、信息与行为语义重点实验室(LMIB))
A design of human-like robust AI machines in object identification
Institute of Automation, Chinese Academy of Science(中国科学院自动化研究所);University of Chinese Academy of Sciences(中国科学院大学)
本文是从图灵(A. M. Turing,1912年6月23日-1954年6月7日)在1950年提出的图灵测试研究中得到的启发而撰写的。根据图灵测试的一个重要含义,使机器具有类似人类的行为或性能,我们为AI机器定义了类似人类的鲁棒性(robustness,简称robustness)。新定义的目标是强制AI机器执行可扩展性,包括根据可扩展性来评估它们。一个特定的任务只讨论对象识别,因为它是每个人在日常生活中最常见的任务。与图灵的观点或设计立场类似,我们提供了一个解决方案,即如何在不构建机器和进行真实实验的情况下实现人工智能机器。解决方案应包括机器的三个重要特征。推理机的第一个特点是利用人类的常识来实现因果推理。第二个特征是从语义空间做出决策,以便对决策进行解释。第三个特点是包括一个“人在回路”的设置,以推进机器人。我们展示了一个“识别游戏”,使用建议的设计的机器。本文试图从图灵测试中学习和进一步探索类人AI机器的设计。
Graph-Driven Models for Gas Mixture Identification and Concentration Estimation on Heterogeneous Sensor Array Signals
School of Electrical and Electronic Engineering, Nanyang Technological University(南洋理工大学电气与电子工程学院);College of Electronic and Information Engineering, Tongji University(同济大学电子与信息工程学院);School of Civil Engineering, Yancheng Institute of Technology(盐城工学院土木工程学院)