BurnBot

For my graduate capstone project at UC Berkeley, my team and I were selected to design and build an obstacle detection system for BurnBot, Inc., a Bay Area startup. They are developing an autonomous, truck-sized mobile burn chamber to be driven through wildland to create firebreaks. Any potential fuel for wildfires is completely burned when the truck passes over it, efficiently creating a barrier that fire cannot cross.

​

Our team was tasked with creating a system to detect any obstacles in BurnBot's way that would damage the sensitive torches necessary for burning the brush. These torches can be raised and lowered, but burn more effectively when closer to the base of any brush. Consequently, our obstacle detection system needs to be able to detect the height of hard obstacles, while ignoring any soft brush. Since an optical system would not work for this purpose, a mechanical system is needed, and that is exactly what we built.

​

A series of blades is attached to a rotating shaft; these blades pass seamlessly through brush, but hard obstacles cause them to rotate. With some simple trigonometry, the height of the incoming obstacles is calculated then fed to BurnBot's central computer so that it can raise and lower the torches accordingly.

​

​This capstone project taught me to view our communication and deliverables through the lens of stakeholder management, which will prove invaluable throughout my career. The prototyping, particularly with various materials and geometries, served as good practice in figuring out the fastest way to reach the conclusion you need before moving on. Multiple times our questions and design standstills were solved by tweaking the current prototype and taking it out into the field for testing.