Outdoor Advertising Robot (Immersion Project)

Overview

I improved an outdoor autonomous robot stack by adding a prediction-based dynamic obstacle avoidance module and building a GUI-linked behavior tree for flexible navigation control. :contentReference[oaicite:22]{index=22}

Key Contributions

• Dynamic obstacle motion prediction
  • Enhanced avoidance by predicting obstacle motion using ByteTrack + Kalman filter. :contentReference[oaicite:23]{index=23}
• GUI-linked behavior tree navigation
  • Developed a GUI-linked behavior tree to enable flexible, user-driven control over navigation behavior. :contentReference[oaicite:24]{index=24}

Context / Role

In my role as an autonomous driving developer, I contributed to outdoor autonomy development and improved dynamic obstacle handling through prediction-based modules. :contentReference[oaicite:25]{index=25}

Tech Stack

• Tracking & Prediction: ByteTrack, Kalman filter
• Decision & Control: Behavior Tree (GUI-linked)
• Outdoor autonomy stack integration (perception + navigation oriented) :contentReference[oaicite:26]{index=26}:contentReference[oaicite:27]{index=27}

Media

Replace placeholders with:

• tracking visualization (ID + predicted trajectories)
• BT graph screenshots
• before/after avoidance behavior clips or images

Takeaway

This project highlights my strengths in:

• integrating perception outputs into planning/control decisions,
• improving real-world robustness with motion prediction,
• and building human-in-the-loop tools (GUI + BT) for reliable autonomy. :contentReference[oaicite:28]{index=28}