Skip to main content

Beginner AI Prompts for Chapter 3

Use these prompts with AI assistants (like Claude, ChatGPT, or your embedded RAG chatbot) to get help understanding perception, SLAM, and navigation concepts.

How to Use These Prompts​

  1. Copy the prompt that matches your question
  2. Paste it into your AI assistant
  3. Add context if needed (e.g., "I'm working on Lesson B2")
  4. Ask follow-ups to dive deeper

Understanding Perception​

What is Perception?​

Explain robotic perception as if I'm completely new to robotics.
What problem does it solve? Use simple analogies.

Perception Pipeline​

Break down the perception pipeline (Sensing → Preprocessing → Feature Extraction → Interpretation)
with a concrete example of a robot detecting a door.

Why is Perception Hard?​

What makes robotic perception challenging? Explain sensor noise,
real-time constraints, and environmental variations in simple terms.

Perception vs. Computer Vision​

What's the difference between computer vision and robotic perception?
When would I use each term?

Understanding Sensors​

Choosing Sensors​

I'm building a mobile robot. How do I decide between RGB cameras,
depth cameras, and LIDAR? What are the trade-offs?

RGB Cameras​

Explain how RGB cameras work in robotics. What can they detect?
What are their limitations? Give examples of use cases.

Depth Cameras​

What is a depth camera and how does it differ from a regular camera?
How do robots use depth information? Explain RGB-D data.

LIDAR Basics​

Explain LIDAR as if I've never heard of it. How does it work?
What does a LIDAR scan look like? Why is it used for navigation?

IMU Sensors​

What is an IMU and what does it measure? How do robots use IMU data
for balance and orientation? Explain in simple terms.

Sensor Fusion​

What is sensor fusion and why do robots need multiple sensors?
Give an example of combining camera and LIDAR data.

Understanding SLAM​

SLAM Basics​

Explain SLAM (Simultaneous Localization and Mapping) using a real-world analogy.
Why is it called "simultaneous"? What problem does it solve?

Localization vs. Mapping​

What's the difference between localization and mapping?
Why can't robots just do one or the other?

Loop Closure​

What is loop closure in SLAM? Why is it important?
Use an analogy to explain how it corrects drift.

Occupancy Grids​

What is an occupancy grid? How does it represent the environment?
Explain free space, occupied space, and unknown space.

SLAM Challenges​

What makes SLAM difficult? Explain the challenges of sensor noise,
dynamic environments, and computational constraints.

SLAM in Practice​

How do real robots use SLAM? Walk me through a robot building a map
of a room step by step.

Understanding Navigation​

Autonomous Navigation​

What does autonomous navigation mean? How is it different from
remote control? What capabilities does a robot need?

Global vs. Local Planning​

Explain the difference between global planning and local planning
in robot navigation. Why do we need both?

Costmaps​

What is a costmap in robot navigation? How does it represent
where the robot can safely go? Use simple terms.

Path Planning​

How do robots plan paths from point A to point B?
Explain the basic idea without complex math.

Obstacle Avoidance​

How do robots avoid obstacles while navigating?
What happens when a new obstacle appears in the path?

Recovery Behaviors​

What are recovery behaviors in navigation? Why do robots need them?
Give examples of when they're used.
What is Nav2? What components does it include?
Explain its role in ROS 2 navigation in simple terms.

Visualization & Tools​

RViz2 for Perception​

How do I use RViz2 to visualize sensor data?
What displays should I add to see camera images, depth data, and LIDAR scans?

Understanding TF Frames​

What are coordinate frames (TF) in robotics? Why do robots need multiple frames?
Explain map, odom, and base_link.

Debugging Perception​

My robot can't see obstacles. How do I debug perception issues?
What should I check first?

Conceptual Questions​

Perception Pipeline Stages​

Walk me through each stage of the perception pipeline with a specific example
(e.g., detecting a person in a camera image).

SLAM vs. GPS​

Why can't indoor robots just use GPS for localization?
What advantages does SLAM provide?

Simulation vs. Real World​

What's different between perception in simulation (Gazebo) and real robots?
What challenges appear in the real world?

Sensor Data Formats​

What do different sensor messages look like in ROS 2?
Explain Image, PointCloud2, and LaserScan message types.

Troubleshooting​

Can't See Sensor Data​

I launched my robot in Gazebo but don't see sensor data in RViz2.
What could be wrong? How do I troubleshoot?

SLAM Not Working​

I'm running SLAM Toolbox but the map looks wrong or doesn't update.
What are common issues and how do I fix them?
My robot won't navigate to the goal. It just sits there or spins.
What should I check?

Frame Errors​

I'm getting TF frame errors. What does "frame does not exist" mean?
How do I fix frame relationship issues?

Learning Strategies​

Study Approach​

I'm new to AI robotics. What's the best order to learn perception, SLAM, and navigation?
Should I focus on theory first or jump into practice?

Hands-On Practice​

What hands-on activities can I do to understand perception better?
I have access to Gazebo simulation.

Key Concepts to Master​

What are the most important concepts in this chapter that I absolutely must understand
before moving to intermediate implementation?

Analogies & Explanations​

Perception Analogy​

Explain robotic perception using the analogy of human vision and understanding.
How is it similar? How is it different?

SLAM Analogy​

Give me a non-robotics analogy for SLAM. Maybe something about exploring
a dark building with a flashlight?
Explain robot navigation using the analogy of driving a car with GPS.
What parts are similar? What's different?

Quick Reference​

Sensor Comparison​

Create a simple comparison table of RGB cameras, depth cameras, and LIDAR.
Include: what they measure, range, use cases, and limitations.

SLAM Components​

List and briefly explain the main components of a SLAM system.
Keep it simple for a beginner.
Describe the complete navigation pipeline from "I want to go there"
to "I arrived" in simple steps.

Tips for Using These Prompts​

  • Be Specific: Add details about what you're working on
  • Ask Follow-ups: Don't hesitate to ask "Can you explain that differently?"
  • Request Examples: Ask for concrete examples if explanations are too abstract
  • Visual Aids: Request diagrams or step-by-step breakdowns
  • Check Understanding: Ask the AI to quiz you on concepts

Remember: There are no stupid questions. If you don't understand something, ask for clarification. AI assistants are here to help you learn at your own pace.

Next: Intermediate AI Prompts (after completing Beginner tier)