Skip to main content

Assessments: Applying Your Knowledge

This chapter outlines the assessments for the "Physical AI & Humanoid Robotics" course. These projects are designed to be practical, hands-on experiences that allow you to apply the concepts you've learned in a tangible way.

1. Midterm Project (Week 7)​

Goal: To apply the concepts from the first half of the course in a small, integrated project.

Task: Create a ROS 2 application that involves a simulated mobile robot performing a simple task in Gazebo. The goal is to navigate to a specific point in a custom environment while avoiding at least one obstacle.

Requirements:​

  • Robot: You can use a simple differential drive robot model (e.g., a TurtleBot3 or a custom model).
  • Environment: You must create a custom Gazebo world (.sdf or .world file) that includes a ground plane, your robot, and at least one static obstacle (e.g., a wall, a cylinder, a box).
  • Control: The robot must be controlled by your own ROS 2 Python nodes. You should have at least:
    • A node that uses sensor data (e.g., from a laser scanner) to detect obstacles.
    • A node that implements a simple navigation logic (e.g., "go-to-goal" with obstacle avoidance).
  • Sensing: The robot must be equipped with at least one sensor for obstacle avoidance (a 360-degree laser scanner is recommended).

Deliverables:​

  1. GitHub Repository: A link to a public GitHub repository containing:
    • Your ROS 2 package(s).
    • Your Gazebo world file.
    • A README.md file with instructions on how to launch your simulation and run your code.
  2. Video Demonstration: A short (1-2 minute) video screen recording showing your robot successfully navigating to a goal while avoiding the obstacle.
  3. Brief Report: A short report (1-2 pages) in Markdown format within your repository that explains:
    • Your overall approach to the problem.
    • The structure of your ROS 2 nodes.
    • The logic behind your obstacle avoidance algorithm.
    • Any challenges you faced and how you overcame them.

2. Final Capstone Project (Weeks 9-13)​

Goal: To design, build, and demonstrate a more complex robotic system, culminating in the "Autonomous Humanoid" project.

Task: The detailed requirements for the capstone project are described in the Capstone Project chapter. This section provides a high-level overview of the grading and expectations.

Grading Criteria:​

The final project will be graded based on the following criteria:

  • Functionality (40%):
    • Does the robot successfully complete the required tasks as outlined in the capstone project description?
    • How robust and reliable is the system?
  • Technical Approach (30%):
    • The quality, clarity, and organization of your code.
    • Effective use of ROS 2 best practices (e.g., modular nodes, launch files, parameters).
    • The cleverness and creativity of your solution to the perception, control, and integration challenges.
  • Presentation (20%):
    • The quality of your final project presentation and live demonstration.
    • Your ability to clearly and concisely explain your work, your design decisions, and your results.
  • Report (10%):
    • The clarity, completeness, and professionalism of your final written report.