Inspiration
We realized it was difficult to inspect some pipes, and it would be an interesting project to make a robot that could follow pipes and navigate around obstacles to inspect the pipe remotely
What it does
A robot travels along a 3" pipe with a sensor package including sonar and an air quality monitor
How I built it
Moving through several iterations of the robot throughout the night, we settled on our design due to time and material restrictions. The robot was mainly built with hand tools and scraps from Harvard's machine shop.
Challenges I ran into
A shortage of parts drove us to change our design a few times. The available sensors did not enable us to detect the size of the leaks and integrating a web page into Arduino code was also a challenge. Trying to get live numeric data from the robot was difficult. Building a drive train to maneuver the robot was taxing. Prototyping robotic arms was challenging too. Determining tensile strength for robot parts expended a lot of fabrication time.
Accomplishments that I'm proud of
At this hackathon, we learned how to use SparkFun's ESP32 Thing using WiFi. We solved the problem of detection of holes using SONAR and Air Quality Index(CCS811) Sensor. We managed to connect HTML to Arduino code without using microSD card. We used a meta tag to constantly get live feed to detect pipes and cracks. We used a gear reduction to use a high speed motor to power our low speed drive wheel therefore, building a drive train. We prototyped the arms with different low risk materials. After prototyping, we slimmed down the design using mathematical analysis, therefore, creating functional arms. There was intense testing to determine the final robotic parts.
What I learned
Techniques learned for future fabrication: Using Computer Aided Design to lay out the construction before hand could have revealed issues before time was spent building. We could also utilize 3D printing and laser cutting to quickly manufacture parts. The micro controller on the Arduino runs into issues with latency when uploading data over WiFi to a web server. New techniques we learned during this challenge: Learned how to program a web server on a microcontroller. Soldered pin headers to breakout boards. Learned how to design gear ratios and drive trains. Learned manufacturing techniques for different materials. Learned to use SparkFun's ESP32 micro controller.
What's next for Pipe Crawler
Computer aided analysis to determine the center of gravity and design future iterations to balance properly. Tighter part tolerances would also aid in this. 3D printing and laser cutting would be integral in this process. Moving the sensor package off of a breadboard would clean up our data. It would be good to find a way to store the data, as currently it is displayed and then overwritten. Eventually, the Pipe Crawler should be able to climb on pipes, avoid obstacles and sense holes and cracks by sending data over WiFi to the web/phone application.
Log in or sign up for Devpost to join the conversation.