### Make a fork or copy of this repo and fill in your team submission details! ###

Team: Team #2, viva la vida, Aiden, Vi, and Vincent

Summary: Robot built with the goal of automatically aiming at targets.

• Beat your friends in a Nerf war even though your aim is garbage.

• Firing mechanism and mounting system custom designed and 3D printed. Servo breakout board was hand designed and soldered on protoboard.

• 3D design done in Onshape, and 3D printed with onsite printers.

• Teleoperation / Dataset capture

  • 

• Training

  • We trained more than 3 models. We took over 200 episodes of data total. We are waiting on our model to train to 1k steps.

• Inference

  • We are waiting on our model to train to 1k steps.

• Should be usable on any toy gun with a sutiable mount. ''
• In a real world situation this could be modified to shoot water from hoses, create"dead zones", etc.
• Flexibility and adaptability of the solution
• Types of commands or interfaces needed to control the robot We created many custom bash shell scripst to interface with our robot. these include calieration scripts, teleoperation scripts, data collection scripts, and inference scripts..

For example, you can provide links to:

• Link to a video of your robot performing the task
• URL of your dataset in Hugging Face huggingface.co
• We have several datasetrs. These are available on https://huggingface.co/squeakmouse They have names such as aim, etc https://huggingface.co/squeakmouse/smolvla_aim_v3_1ksteps
• URL of your model in Hugging Face
• Link to a blog post describing your work N/A

This is the directory tree of this repo, you need to fill in the mission directory with your submission details.

AMD_Robotics_Hackathon_2025_ProjectTemplate-main/
├── README.md
└── mission
    ├── code
    │   └── 
    └── wandb
        └── 
ur training job>
b>

$ tree outputs/train/smolvla_so101_2cube_30k_steps/wandb/
outputs/train/smolvla_so101_2cube_30k_steps/wandb/
├── debug-internal.log -> run-20251029_063411-tz1cpo59/logs/debug-internal.log
├── debug.log -> run-20251029_063411-tz1cpo59/logs/debug.log
├── latest-run -> run-20251029_063411-tz1cpo59
└── run-20251029_063411-tz1cpo59
    ├── files
    │   ├── config.yaml
    │   ├── output.log
    │   ├── requirements.txt
    │   ├── wandb-metadata.json
    │   └── wandb-summary.json
    ├── logs
    │   ├── debug-core.log -> /dataset/.cache/wandb/logs/core-debug-20251029_063411.log
    │   ├── debug-internal.log
    │   └── debug.log
    ├── run-tz1cpo59.wandb
    └── tmp
        └── code
g
    ├── run-tz1cpo59.wandb
    └── tmp
        └── code
   └── code