Pi Scout

Pi Scout: A teleoperated rescue rover that detects invisible phone signals to locate trapped victims and maps their position.

Comment

Inspiration

In search and rescue operations - whether after an earthquake, a building collapse, or a forest fire - the first 72 hours are critical. This is known as the "Golden Window" for survival. However, sending human rescuers into these zones is incredibly dangerous due to unstable structures, toxic gas leaks, and low visibility.

We realized that while victims might be unconscious or buried under rubble, they almost always have one thing with them: a smartphone. Even without cellular service, phones constantly emit "Probe Requests" looking for known Wi-Fi networks.

We were inspired to build a robot that could "hear" these invisible cries for help. We wanted to create a tool that not only locates victims through digital signals but also acts as a "canary in the coal mine," sensing environmental hazards like toxic gas to keep human responders safe.

What it does

Pi Scout is teleoperated tactical rover designed for hazardous environments. It performs three critical functions:

  1. Digital Search & Rescue: It uses a Raspberry Pi in "Monitor Mode" to sniff for Wi-Fi packets. It detects the unique MAC addresses of smartphones in the area and uses RSSI (Signal Strength) to estimate the victim's distance.

  2. Environmental Hazard Analysis: Equipped with an Air Quality Sensor (MQ-135), it streams real-time atmospheric data to the dashboard. If it detects smoke, CO2, or toxic gas, it instantly warns the operator, preventing rescuers from walking into a lethal trap.

  3. Real-Time Mapping: As the operator drives the bot via a web interface, the robot tracks its own path using dead reckoning, creating a 2D map of the safe path to the victim.

How we built it

We approached this challenge by integrating hardware engineering with complex network analysis.

The Hardware: We designed a custom chassis (Fusion 360) and 3D printed it to house a Raspberry Pi 4. We integrated standard DC motors with an TB6612FNG driver for locomotion, an Ultrasonic Sensor for collision avoidance, and an Air Quality sensor for environmental data.

The Software: The core logic is written in Python.

The Interface: We built a low-latency web dashboard using HTML Canvas and HTML/CSS for GUI, FlaskAPI. This allows the user to drive the robot from any laptop using just a browser, viewing live data graphs and the map in real-time.

The Math: To map the robot's path without expensive GPS (which fails indoors), we implemented a dead reckoning algorithm based on motor timing: $$x_{new} = x_{old} + d \cdot \cos(\theta)$$ $$y_{new} = y_{old} + d \cdot \sin(\theta)$$

Challenges we ran into

The "Monitor Mode" Hurdle: Configuring the Raspberry Pi's internal Wi-Fi chip to enter "Monitor Mode" (promiscuous mode) was difficult, as standard drivers often block this. We had to dig deep into Linux networking configurations to bypass these restrictions.

Power Management & Brownouts: Early in testing, the high-torque servo motors drew massive current spikes, causing the Raspberry Pi to suffer undervoltage warnings and reboot. We solved this by electrically isolating the power rails. We used a dedicated power bank to drive the servo motors separately, ensuring the Raspberry Pi remained stable on its own independent power source.

Accomplishments that we're proud of

Seeing the Invisible: The first time our terminal populated with the MAC address of a "hidden" phone in the room was a massive victory. It proved our concept that Wi-Fi can be a rescue beacon.

Sensor Fusion: Successfully getting the Air Quality data, motor controls, and packet sniffing to all run simultaneously on one dashboard without crashing the Python script.

Sustainability: We built a functional prototype that promotes "data-driven environmental safety," proving that we can monitor disaster zones without risking human lives or using heavy, fuel-burning machinery.

What we learned

Network Forensics: We learned a huge amount about the 802.11 Wi-Fi protocol - specifically how devices "shout" for connection even when idle.

Hardware Limitations: We learned the hard way about current draw and the importance of common grounds when mixing 5V motors with 3.3V logic boards.

The Importance of UX: A rescue tool is useless if it's hard to control. Building a clean, graphical interface was just as important as the backend code.

What's next for Pi Scout

LoRaWAN Integration: Currently, we rely on Wi-Fi for control. We plan to integrate LoRaWAN modules to extend our telemetry range from 100 meters to 5+ kilometers, allowing deep-forest deployment.

: We want to upgrade from simple dead reckoning to full SLAM (Simultaneous Localization and Mapping) using LiDAR for precise indoor mapping.

Thermal Imaging: Adding a thermal camera would allow us to fuse "heat signatures" with "Wi-Fi signatures" for 100% confirmation of a survivor's location.

Built With

Share this project:

Updates