Course Overview: Robot Mapping

Introduction to the Course

• Short overview of topics covered in the course.
• Familiar concepts from the previous robotics course but with more detail.
• Emphasis on hands-on work and programming assignments.

Key Topics to Cover

1. Simultaneous Localization and Mapping (SLAM)

   • Introduction to SLAM principles.

2. Kalman Filters

   • Review of Kalman filter and extensions:
     • Extended Kalman Filter
     • Unscented Kalman Filter
     • Information Filters (inverse Gaussian distribution).

3. Particle Filtering

   • Focus on particle filters in robot mapping and SLAM, including:
     • Rao-Blackwellized Particle Filter.

4. Error Minimization Approaches

   • Concepts of least squares error minimization in mapping.
   • Hierarchical approaches and dealing with imperfect input data.

5. SLAM Front-end

   • How robots interpret sensor data and data association.

6. Appearance-Based Approaches

   • Focus on Kinect-based SLAM for high-resolution modeling.
   • Low-cost depth sensor advantages.

Learning Goals

• Understand key milestones and concepts in robot mapping over the last 20 years.
• Gain hands-on experience through programming assignments.
• Understand algorithms in-depth by implementing them independently.

Skills and Requirements

• Basic math skills:
  • Familiarity with linear algebra (vectors, matrices, Jacobians).
• Understanding of probabilistic concepts:
  • Probability distributions, Bayes rule.
• Basic programming skills:
  • Recommended language: Octave (similar to MATLAB).

Homework Assignments

• Participation in assignments is optional but highly encouraged for better exam performance.
• Assignments graded with feedback from teaching assistants.

Course Structure

• Lectures on Mondays (10 AM to 12 PM) and exercises on Wednesdays (2 PM to 4 PM).
• Bring laptops with Octave installed for hands-on work during exercises.

Communication

• Use Google Groups for questions & discussions (no solutions to exercises posted).
• Instructor and assistant available for office hours and feedback.

Textbooks and Resources

• Main textbook: "Probabilistic Robotics"
• Additional resources available as PDFs on course website.
• Errata list for textbook online to clarify inconsistencies.

Exam Structure

• Oral exam format.
• Understand material and transfer knowledge to new problems.
• Key concepts should be memorized but understanding is prioritized.

Interaction Encouraged

• Interactive course, welcome questions and feedback.
• Alter teaching methods based on student needs for better learning experience.