EE · Fall 2025 · Graduate
Introduction to Robot Learning
A practical bridge between robotics, machine learning, perception, planning, and intelligent decision-making.
Course Overview
Robot learning sits at the intersection of robotics and artificial intelligence. This course introduces the essential concepts needed to understand modern robot-learning literature, from coordinate transforms and kinematics to planning, simultaneous localization and mapping, reinforcement learning, and 3D perception. Programming assignments and a team project connect these ideas to working robot systems.
- Instructor
- Prof. Nuri Kim
- Term
- Fall 2025
- Meetings
- Tuesday
18:00–20:30 - Classroom
- Engineering Building 7
Room 329
Learning Objectives
Robotics foundations
- Represent rigid-body motion with coordinate transformations.
- Analyze forward and inverse kinematics for robot manipulators.
- Understand simulation, path planning, and robot perception pipelines.
- Connect sensor-based and visual SLAM to autonomous operation.
Learning and implementation
- Formulate sequential decision problems using reinforcement learning.
- Implement and evaluate deep reinforcement-learning agents.
- Read recent robot-learning papers and identify open limitations.
- Integrate perception, planning, and control in a team robot project.
Recommended background: Python programming, linear algebra, probability, and introductory machine learning.
Schedule
Robot Foundations
Research & Project
Autonomy & Learning
This page records the Fall 2025 offering. Course files remain available to enrolled students through the JBNU LMS.
Coursework
Assignment 1
Ubuntu & ROS 2 Setup
Prepare the Ubuntu 24.04 and ROS 2 Jazzy development environment.
Assignment 2
Simulation & Planning
Work with a simulated manipulator and implement motion-planning components.
Assignment 3
Deep Reinforcement Learning
Train a DQN agent for CartPole and an A3C agent for Atari Pong.
Team Project · Robot Drawing
Midterm presentation
Review three recent papers and present a focused project proposal.
Final deliverables
Demonstrate the robot system, present the methodology and results, and submit an academic-style report of at least four pages.
- Midterm Presentation
- 30%
- Final Presentation
- 30%
- Assignments
- 30%
- Attendance
- 10%
Resources
- Course platformJBNU LMS
- DevelopmentUbuntu 24.04 · ROS 2 Jazzy · Python · PyTorch
- ReferenceShuuji Kajita et al., Introduction to Humanoid Robotics
- ReferenceRichard Hartley and Andrew Zisserman, Multiple View Geometry in Computer Vision
- Contactnuri.kim@jbnu.ac.kr