Tutorial 4 - Follow the Gap

• Follow the Gap Method
  • Core Concept
  • Step-by-Step Process
    • 1️⃣ Data Acquisition
    • 2️⃣ Preprocessing
    • 3️⃣ Obstacle Masking (Safety Bubble)
    • 4️⃣ Gap Detection
    • 5️⃣ Best Point Selection Within the Gap
    • 6️⃣ Navigation Command
  • Practical Considerations
  • Advantages
  • Limitations
• Create Gap Following Package
  • 1️⃣ Create the Gap Follow Package
  • 2️⃣ Modify package.xml
  • 3️⃣ Modify CMakeLists.txt (If Using C++)
  • 4️⃣ Install Dependencies Using rosdep
  • 5️⃣ Implement the Gap Follow Node
  • 6️⃣ Make It Executable
  • 7️⃣ Build & Run
    • C++ Alternative Implementation
• Preprocessing LiDAR Data
  • Why Preprocess?
  • Detailed Implementation
  • Usage
  • (Optional) Smoothing
    • Why Apply Smoothing?
    • Moving Average Smoothing (Simple Concept)
    • Python Code: Smoothing with Convolution
    • How Smoothing Fits into Preprocessing
    • Why Convolution?
    • Tip: Choosing Window Size
    • Before vs After Example
• Obstacle Masking: Creating a Safety Bubble
  • 🛠 Why Create a Safety Bubble?
  • 📈 What the Data Looks Like
  • ✏️ Safety Bubble Method (Skeleton Code)
  • ⚙️ How It Works Step-by-Step
  • 🧠 Why Mask Points to Zero?
  • 🚀 Key Parameters
  • 📚 Summary
• Gap Detection
  • Follow The Gap - Step-by-Step Breakdown
    • Step 1: Find the Closest Obstacle and Create a Safety Bubble
    • Step 2: Find the Maximum Gap
    • Step 3: Pick the Best Point
  • Example Code for Step 1 and Step 2
  • Key Points to Remember
  • Next Step
• Finding the Best Point: Naive Selection
  • Why This Approach?
  • Step-by-Step Breakdown
  • Python Implementation
  • Converting Index to Steering Angle
  • Strengths of Naive Selection
  • Weaknesses and Limitations
  • When to Use This Method
  • Next Steps
• Finding the Best Point: Disparity Selection
  • 🛠 What is a Disparity?
  • 🧠 Why Use Disparity?
  • 🛠 How It Works
    • Step 1: Focus on the Largest Gap
    • Step 2: Detect Disparities
    • Step 3: Find the Open Region
    • Step 4: Aim for the Middle
  • ✏️ Python Example (Disparity-Based Selection)
  • 📊 Visual Diagram
  • ⚡ Advantages Over Farthest Point
  • 🚀 Summary