基本素養 Basic Literacy

畢業生應具備科技人文素養、資訊工程倫理與終身學習的態度
graduates should equip with the attitude of technological/cultural literacy, information engineering ethics, and life-long learning
畢業生應具備專業外語能力及良好國際觀
graduates should equip with both the professional foreign language proficiency and excellent global view

核心能力 Competence

畢業生應具備資訊專業理論知識
graduates should equip with professional theoretical knowledge in informatics
畢業生應具備資訊專業理論推導、分析、歸納之能力
graduates should equip with the capability of professional theory derivation, analysis, and induction in informatics
畢業生應具備資訊領域獨立發掘問題、策劃實驗、解決問題之能力
graduates should equip with the informatics ability to identify problems independently, to implement experiments, and to solve problems
畢業生應具備資訊領域設計、驗證及實作整合之能力
graduates should equip with the informatics ability in designing, verification, and integrating engineering practices
畢業生應具備資訊領域創新思考之能力
graduates should equip with the informatics capability in innovative planning
畢業生應具備專業簡報及論文撰寫之能力
graduates should equip with the ability in professional presentation and thesis writing
畢業生應具備良好溝通協調與團隊合作之能力
graduates should equip with fair ability in communication, coordination, and team-work collaboration

課程概述 Course Description

本課程模組分為三個主要的部分,分別為即時追蹤與地圖建置(SLAM)、基於機器學習之場景理解(Scene Understanding)與探索導航的動作控制(Action Control)。即時追蹤與地圖建置部分包含機率模型與相機模型等理論基礎,再搭配2D場景追蹤建圖的實作並介紹RGB-based的3D SLAM。場景理解的部分包含機器學習的基本概念,再帶到深度學習的技術與目前的物件偵測與語意切割技術。動作控制的部分則包含路徑規劃與導航演算法,並帶入強化學習的概念來引導行進的路徑。
本課程模組分為三個主要的部分,分別為即時追蹤與地圖建置(SLAM)、基於機器學習之場景理解(Scene Understanding)與探索導航的動作控制(Action Control)。即時追蹤與地圖建置部分包含機率模型與相機模型等理論基礎,再搭配2D場景追蹤建圖的實作並介紹RGB-based的3D SLAM。場景理解的部分包含機器學習的基本概念,再帶到深度學習的技術與目前的物件偵測與語意切割技術。動作控制的部分則包含路徑規劃與導航演算法,並帶入強化學習的概念來引導行進的路徑。

課程學習目標 Course Objectives

  • 熟悉自走 車導航與自動駕駛所需的各種基礎理論
  • 深入學習實作底層的演 算法
  • 將理論應用在模擬環境
  • 深度學習技術的使用
  • 課程進度 Progress Description

    進度說明 Progress Description
    1Introduction of Robotic Navigation and Exploration Algorithm-based Methodology Learning-based Methodology w/ Map vs. w/o Map
    2Sensor Technologies for Simultaneous Localization and Mapping (SLAM) Probability Basics and Bayesian Statistics Grid Map Algorithm HW1 : Grid Map
    3Online Bayesian Learning Kalman Filter Extended Kalman Filter Sampling Method CDF Sampling Rejection MCMC
    4Monte-Carlo Sequential Learning Importance (particle filter) SIS (Sequential Importance Sampling) SIR (Sampling Importance Resampling) Particle Filter Fast-SLAM / Grid Fast-SLAM
    5Implementation Details of Monte-Carlo Sequential Learning Line Tracing Laser Beam Model Motion Model Adaptive Resampling Method HW2 : Grid Fast-SLAM
    63D SLAM Multi-View Geometry Structure from Motion Indirect Method ORB SLAM (Project Bonus) Direct Method Depth-based Method
    73D SLAM Multi-View Geometry Structure from Motion Indirect Method ORB SLAM (Project Bonus) Direct Method Depth-based Method
    8Machine Learning Basics What is Machine Learning Categories of Machine Learning Methods Optimization Basics
    9Deep Learning Artificial Neural Network and Back Propagation Computational Graph and Tensor Differential Data Simulation and Feedforward Implementation with NumPy Backward Propagation Implementation with NumPy Modern Deep Learning and Structures Tensorflow Basics and Visualization
    10Deep Learning Artificial Neural Network and Back Propagation Computational Graph and Tensor Differential Data Simulation and Feedforward Implementation with NumPy Backward Propagation Implementation with NumPy Modern Deep Learning and Structures Tensorflow Basics and Visualization
    11Deep Neural Network for Computer Vision Object Detection Semantic Segmentation HW3 : YOLO
    12Path planning algorithms Optimization-based control
    13Markov Decision Process Value Iteration Policy Iteration
    14Reinforcement Learning Q-Learning / Sarsa / DQN Policy Gradient / Actor-Critic HW4 : A2C (Car Example)
    15Reinforcement Learning Q-Learning / Sarsa / DQN Policy Gradient / Actor-Critic HW4 : A2C (Car Example)
    16Project Development Talk by Dr. Trista Chen
    17Project Development Talk by Dr. Trista Chen
    18Final Project
     以上每週進度教師可依上課情況做適度調整。The schedule may be subject to change.

    有關課程其他調查 Other Surveys of Courses

    1.本課程是否規劃業界教師參與教學或演講? 否
    Is there any industry specialist invited in this course? How many times? No
    2.本課程是否規劃含校外實習(並非參訪)?
    Are there any internships involved in the course? How many hours?
    3.本課程是否可歸認為學術倫理課程? 否
    Is this course recognized as an academic ethics course? In the course how many hours are regarding academic ethics topics? No
    4.本課程是否屬進入社區實踐課程? 否
    Is this course recognized as a Community engagement and Service learning course? Which community will be engaged? No

    教師上傳大綱內容
    本資訊僅提供本校師生參考。有著作權,非本校人員若欲使用本資訊,請洽本校取得授權。
    © (2008-2023) National Cheng Kung University ALL RIGHTS RESERVED.

    課程大綱及進度表資訊系所-機器導航與探索_108下.pdf