機器人視覺

This course aims at training students to participate world-class competition, and we will target the RobotX Competition (https://www.robotx.org/) in 2018. This course will cover the fundamental and advanced domains in vision for mobility tasks, including the challenges combining object recognition, pose estimation, motion planning, and SLAM. We will systematically study each components from previous winning teams in 2014 and 2016, as well as cutting-edge methods that may better improve the performance. Students will form teams to develop term projects, and we encourage in-class discussions and presentation skills with a white board for class participations. This course is a "learning-by-doing," and include in-class lab/tutorial in each teaching module. We will use "Duckietown" (an open course "MIT 2.166 Autonomous Vehicles) as platform, and focus advanced topics using Gazebo, 3D Perception, and deep learning.

This course involve a fair amount of probability, linear algebra, and programming. Students are required to take image processing or computer vision. Being familiar with Unix-like system, Git, C++, Robot Operation System (ROS) and Python is needed. Due to the tight competition schedule, we wish to cover more advanced topics directly relevant to the competition. *We will assume that you have been familiar with Duckietown A to H (http://duckietown.nctu.edu.tw/materials.html), which is an open-source, self-learning material. 20 Duckiebots will be available after 1/20. Please feel free to get started early during winter break! *Due to the limited robots/computing resource for Gazebo and Deep Learning, we will ask all attended students to fill in a short self-evaluation and motivation statement. We apologise that we can only enroll a limited number of students according to students' 1) motivation, and knowledge of 2) of Unix/Ubuntu, Vim, Git, 3) Duckietown and ROS, 4) computer vision and image processing, and 5) your own computing resource (such as a native Ubuntu machine with GTX 1060 or above).

* Task-oriented: We aim at the RobotX competition (https://www.robotx.org/) in 2018. * Learning by Doing: We have lecture and lab sections, and include hand-on tutorial/lab materials each week. * In-class Discussion: We encourage discussions and presentation skills using a white board each team for class participation. We also wish to train students' presentation skills. * Project-based learning: each team will work on a term project related to RobotX Competition.

1.學期作業: We have in-class tutorial/lab each week, and the lab materials typically include 3 specific tasks (programming/algorithm/system work). and students are expected to finish the lab materials during the class. 2. 考試狀況: There will be no midterm/final exams. 3.評量方法: In-Class Lab/Tutorials (40%), Midterm Project Proposal (20%) Final Presentation and Term Paper/Report (40%)

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.