FireWatch is low-cost device that is able to accurately identify and localize forest fires.
This image is the FireWatch Logo.
This image shows the FireWatch Device with the cost of each component.
This is an image of me installing one of the FireWatch Devices.
This image shows the specific details of the fire that was identified.
Eligibility: Please confirm you meet all the following criteria
You are aged between 14 - 18 as of August 1, 2020
You live in the United States or its territories
You are not employed by, or directly related (grand parents, parents or siblings) to a current General Motors (GM) or Ashoka employee
You have been working on this project for at least three months
You consent to us possibly featuring your work on social media
You confirm you have the rights to use and share any content uploaded on this entry form
Eligibility: Date of Birth
What gender do you identify with?
Help us stay in touch!
Phone Number: (703)-623-6700
Address: 3401 Waples Glen Ct., Oakton, VA 22124
Website or social media url(s) (optional):
Date You Started Your Project Started
Project Stage: Select the description below that best applies to your approach.
Scaling (expanding impact to many new places or in many new ways)
1. The Problem: What problem are you helping to solve?
Forest fires can destroy property, take lives, and wreak environmental havoc. Since wildfires grow exponentially, quickly alerting firefighters to the exact location of the fire is critical in preventing fires from becoming uncontrollable. Current detection systems either lack the ability to identify the location of the fire or are prohibitively expensive. In order to help firefighters it is necessary to create low-cost reliable fire detection.
2. Your Solution: How are you planning to solve this problem? Share your specific approach.
Current camera-based fire monitoring systems cost up to $70k and fail to detect the exact location of wildfires limiting their applicability. As a response to this problem, I designed, built and implemented a low-cost monitoring device that can accurately detect and localize a fire with hardware that costs only $103.
The device contains a 3d printed body that contains a Raspberry Pi Computer, a cellular modem, battery pack, and other electronic components. On top of the main body, I mounted a solar panel to power the device. Attached to the side of the main body is a camera hub that sits above the main device. The camera hub contains 4 cameras mounted at 90° angles in order to provide 360° coverage.
At one minute intervals, the device takes a picture with one of each of its for images and scans the image for potential wildfires. If a wildfire is detected, the exact Latitude and Longitude are sent to first responders in order to minimize the latency between detecting fires and first responders arriving to the fire.
Testing showed that the FireWatch device could operate for two days without sunlight and could withstand heavy rainfall making it usable in real-life conditions.
3. Using STEM as a force of good: Please specify how are you using STEM to solve for an environmental challenge you are passionate about.
In order to make the device low-cost and small, the onboard computer is not very powerful and has limited battery capacity. Existing camera detection algorithms would require more computation power and energy than possible.
Given these limitations, I had to adapt existing computer vision algorithms. The fundamental idea behind the data-sparsity approach is that most pixels in an image aren’t relevant for fire detection. For example, the parts of an image that show the sky without smoke or clouds aren’t relevant and an algorithm can detect a fire in the image without seeing that part of the sky. By focusing the neural network on the relevant parts of the image I reduce power consumption by 80% while maintaining equivalent accuracy.
I also created a technique called GeoMatching to precisely localize the fire. To do this, I extract keypoint features from the camera image and match these to the same features at that location in Google Earth. I then solve for the orientation of the camera and use ray tracing to determine the actual coordinates of the fire. This technique, is able to detect wildfires with a mean error of 43.2 m. Watch the attached at the top for more information.
4. Personal Journey: What’s the story behind why you decided to start this project?
I’ve always spent as much time as I could in the mountains. In summers, I hike and camp practically every weekend, while during the winters, I try to go skiing as often as I possibly can. There’s something about the crisp fresh air and freedom that nature provides, and over time, I’ve become passionate about preserving the pristine nature of the places, especially after witnessing the devastation of a wild fire in the summer of 2018.
I was hiking through the Sierra Nevada Mountains of California, when five miles in, I passed over a ridge and found hundreds of thousands of charred tree trunks littering the landscape. The mountainside looked like it was straight out of a post-apocalyptic novel. After seeing the utter devastation to the ecosystem and loss of habitat firsthand I decided that I wanted to use my knowledge of computer science to help first responder fight forest fires.
5. Video (Keep it simple, a video made on a hand-held phone is great): Please upload a 1-minute video to YouTube that answers the following “I am stepping up to be a Changemaker because...”
Why I am stepping up to be a Changemaker
6. Please highlight the key activities you have carried out to bring your project to life.
The main activities that were required to develop the FireWatch prototype were developing the fire detection algorithm and designing/fabricating the physical components of the FireWatch device.
After developing the prototype, I decided to recruit a team of 12 volunteers from local high schools and universities who help me with hardware, software, and public outreach. Without the help of my team, the device could never have advanced from a prototype to a fully deployable end-to-end product.
7. The X Factor: What is different about your project compared to other programs or solutions already out there?
There exists two modern widely-used fire detection systems: Camera based systems and CO2 Detectors. Camera based systems cost upwards of $70k and can't provide the exact location of the fire. CO2 detectors can provide a location, but have highly limited range making them impractical in most circumstances.
The FireWatch device combines the best of both worlds with the high range and accuracy of the camera systems with the low-cost decentralized localization of the CO2 detectors. Additionally, the cost of the FireWatch device is only $103.22, less than 1% the cost of state of the art systems.
8. Impact: In the last three months, please detail the impact your project has made?
In February of 2020 I partnered with several local land owners in the Shenandoah Mountains in Virginia to set up 8 FireWatch devices as a beta program. The devices cover a combined area of greater than 600,000 acres of mainly wooded land. The purpose of the beta program was to test device durability, battery length, and waterproofing all in real-world circumstances.
However, the beta test took an unexpected turn when on the morning of March 10th, one of the devices was able to accurately detect and localize a real forest fire in the Tanners Ridge area. The accurate detection and localization of a real forest fire in real-world conditions providing convincing proof for the reliability and accuracy of the system.
Starting in late March, in partnership with local organizations in Allegheny County in VA, I have set up three more devices that are actively monitoring for forest fires.
9. What’s Next: What are your ideas for taking your project to the next level?
In December 2019, I laid out my plan for 2020 which focused on local testing and partnerships until June. I planned over the summer (2020) to transition to find partners on the West Coast where fires are more prevalent.
Unfortunately, due to the Coronavirus Pandemic it will be very difficult for me to establish local connections in the region without being able to travel, demo or install devices Instead, I am taking the Pandemic as an opportunity to focus on furthering the software and hardware components of the FireWatch device and continuing to recruit a larger team. Additionally, I am looking into how I can transition FireWatch from being entirely reliant on donations into a sustainable non-profit with a revenue model for the future.
10. Please share how you have influenced other young people to get involved in your project and/or care about environmental sustainability.
In my effort to bring FireWatch from a prototype to a deployment-ready product. I recruited young people from High Schools in Northern Virginia along with students from GMU and GWU in DC.
The FireWatch volunteers were split into 3 teams: the software, hardware, and public outreach teams. Many of the volunteers were hesitant at first about helping an environmental non-profit. However, over time it was evident that all of our volunteers became as passionate as I am about protecting forests.
11. Please share ideas of how you can partner with other changemakers to make a difference?
Some Changemakers may have contacts with local governments that would prove invaluable for finding new local partners. Additionally, they may be able to volunteer their time or technical expertise to FireWatch. I'm always looking for volunteers with engineering and software experience. Finally, I'd like to collaborate with anyone with social media marketing experience because I am looking to increase our online presence. I look forward to the opportunity to collaborate with other Changemakers!
12. How would you engage others who have never heard about your project to get their buy-in?
Most people already understand that forest fires are bad for the environment and that helping firefighters is a good thing. I think with FireWatch, the numbers speak for themselves. The 97% fire detection accuracy at 1% the cost of existing systems is simply staggering. When people see what we're building here they want to volunteer for the kind of organization that is committed to protecting against forest fires around the world and is already making significant strides.
13. Finances: If applicable, have you mobilized any of the following resources so far?
Donations between $1k-$5k
Name: Musad Haque
Name: Ravin Suri
Name: Scott Ober
Are you employed, or directly related (grand-parents, parents, sibling) to a GM or Ashoka employee?
Help Us Support Diversity! Part 1 [optional] Which of the following categories do you identify with?
Asian (for example: Chinese, Filipino, Indian, Vietnamese, Korean, Japanese, Pakistani) (9)
Help Us Support Diversity! Part 2 [optional] Do you identify as part of any of the following underrepresented communities?
Communities of color
Religious minority (non-Christian)
How did you hear about this challenge?
Referral: If you discovered the Challenge thanks to an organization or person other than Ashoka or General Motors, who was it?
I was referred to the challenge by Scott Rechler of LearnServe International.