感測與智慧系統
Sensing and Intelligent Systems
| 節 | 週一 |
|---|---|
5 13:20–14:10 | 感測與智慧系統 EE635 3 節連堂 |
6 14:20–15:10 | |
7 15:30–16:20 |
* 根據陽明交大上課時間表所列
This course aims at training students to participate world-class competition, and we will target the Amazon Robotics Challenge (https://www.amazonrobotics.com/#/roboticschallenge; previously known as Amazon Picking Challenge) in 2017. This course will cover the sensing and intelligence aspects of robots, including the challenges combining object recognition, pose recognition, grasp planning, compliant manipulation, motion planning, task planning, task execution, and error detection and recovery. We will systematically study each components from previous winning teams in 2015 and 2016, as well as cutting-edge methods that may better improve the performance. We will prepare virtual environment (Gazebo) for a virtual challenge in class.
This course involve a fair amount of probability, linear algebra, and programming. Students who took image processing, computer vision, and creative software project are encouraged to join. Being familiar with Unix-like system, Robot Operation System (ROS), C++, Python, and basic knowledge of Gazebo and Deep Learning are required. We recommend the Duckietown course in Creative Software Project and Robotic Vision OCW to learn essential technical skills.
1.學期作業: The assignments include problem sets, and the process of writing and reviewing a research paper. Students form teams, with 2-3 people. Each team will prepare a in-class tutorial with a demo of running code. Each team can choose a component (such as pose estimation or others) with systematic evaluation or improvement as term project. 2. 考試狀況: There will be midterm/final exams and in-class quiz. 3.評量方法: Exam (20%), Class Participation, In Class Quiz, Problem Sets (10%), Final presentation (10%) Project Research Paper (50%)
| 週次 | 主題 |
|---|---|
| 第 1 週 | Introduction to Sensing and Intelligent System: Robotic Competition |
| 第 2 週 | Lecture: Robotics Tools & Technical Writing Lab: VirtualBox, Git, Docker |
| 第 3 週 | Lecture: Spatial Representations Lab: ROS Tutorial 1 (Command line, topic, tf) |
| 第 4 週 | Lecture: AprilTag 1, 2 Lab: ROS Tutorial 2 (Service. Launch, Apriltag) |
| 第 5 週 | Lecture: Invited Speaker: Peter Yu (CTO at XYZ Robotics) Lab: OpenCV and Depth Sensing (Realsense) |
| 第 6 週 | 校際活動週 停課一週 |
| 第 7 週 | Lecture: 2D/3D Perception Lab: Calibration ChAruCo |
| 第 8 週 | Lecture: Motion Planning of Robot Arms Lab: Arm 1 (Dynamixal Pyrobot & Control Locobot, IK, FK) |
| 第 9 週 | Lecture: System Integration/Task Planning Lab: Arm 2 (MoveIt!, Simulate in RViz, Yumi) |
| 第 10 週 | Lecture: Problem Formulation & Experiment Design Lab: PCL and ICP |
| 第 11 週 | Lecture: Deep Learning & Semantic Segmentation (1) Lab: PyTorch Intro |
| 第 12 週 | Lecture: Deep Learning & Semantic Segmentation (2) Lab: CNN Classification |
| 第 13 週 | Lecture: Deep Learning & Semantic Segmentation (3) Lab: Training on Workstation Tutorial / Dataset / Labelme |
| 第 14 週 | Lecture: Deep Learning & Semantic Segmentation (4) Lab: Semantic Segmentation: FCN / Unet |
| 第 15 週 | Lecture: Deep Learning & Semantic Segmentation (5) Lab: Semantic Segmentation: Mask RCNN |
| 第 16 週 | Lecture: Navigation Lab: PyRobot Navigation (CMP) & Wheel control |
| 第 17 週 | Final Exams (Open Book) |
| 第 18 週 | Mini Competition |
1. Computer Vision: Algorithms and Applications, Richard Szeliski, Springer, 2010. 2. Robotics, Vision, and Control, Peter Croke, Springer, 2011. 3. Introduction to Autonomous Robots, Nikolaus Correll, 2015.