Dispatch

Inspiration:

First responders need rapid, realistic training that prepares them for high-stress, unpredictable environments. Traditional VR training systems often rely on treadmills, fixed setups, or limited forward-only movement. This restricts realism, especially in high-stakes scenarios like navigating burning buildings.

We built Dispatch — a VR training experience that uses motion sensors and natural locomotion to prepare first responders for high-stress, life-or-death scenarios inside a burning building. Dispatch enables cost-effective, at-home, or on-site VR training for firefighters, EMTs, and rescue teams improving accessibility.

Research shows the need:

• Arxiv 2301.13124 & 2410.14140: Highlight VR’s role in scalable first responder training.
  
• FIT 2023 Firefighter Study: Demonstrates that firefighters can use VR for drills outside of costly in-person sessions.

What it does:

Dispatch offers a VR training scenario inside a maze-like burning building where:

• Novel locomotion: instead of using a physical joystick, your body becomes the joystick. By walking/running in place the system estimates your speed and direction.
• Dynamic environment: Fires spread over time — starting with smoke, then flames, with sounds of cracking beams and obstacles.
• Sensory immersion: Players hear fire grow louder and victims cry for help as they get closer.
• Objectives: Rescue dummies trapped in rooms (randomized each round), make rapid situational decisions, and escape.
• HUD stats: Track maze progress, survival time, and decisions.
• Fail state: Screen fades to black with heartbeat and heavy breathing audio, mimicking collapse.

How we built it:

• Unity + C# for maze creation, particle effects, colliders, and VR setup.
• SteamVR/OpenVR Driver in C++ for integrating with external hardware
• PlatformIO in C++ for firmware to read IMU data and communicate wirelessly
• 3D printed tracker casings and velcro straps for customizable wearability (we made trackers for the chest and both knees but only used the chest one)

Challenges we ran into:

• Communication between microcontroller and SteamVR was difficult to implement
• Balancing realism with hackathon time constraints (e.g., scalable maze vs. simple proof-of-concept).

Accomplishments that we're proud of:

• Building a functional VR demo in just a hackathon sprint.
• Creating a locomotion system that can scale beyond first responder training — potentially into gaming, rehab, and VR fitness.
• Combining research-backed needs with a practical, lightweight prototype.

What we learned:

• Simple and working is better than complicated and not working
• Never give up

What's next for Dispatch:

• Better gait analysis with knee and foot trackers
• Physiological tracking (heart rate to gauge stress)
• Expand to adaptive training programs with different dififculty levels
• AI/LLM feedback on performance — e.g., highlighting weak spots like hesitation or poor route choice
• Pilot with first responder groups to validate effectiveness

view our pitch here: https://www.canva.com/design/DAG0O43r9ug/YQv7im-jHmJN4KTk9l8Tmw/view?utm_content=DAG0O43r9ug&utm_campaign=designshare&utm_medium=link2&utm_source=uniquelinks&utlId=hc58c1f20c6

Built With

• c#
• c++
• platformio
• python
• steamvr
• unity