As the bachelor’s thesis project for the graduation from Middle East Technical University EEE department, we have chosen to implement a robot that can climb vertical walls among a choice of 5 different projects which are intentionally defined vaguely so that the students are given the responsibility to shape the details of the project. The project is to be implemented by a group of 5, which is expected to be structured as a small startup company, and present and follow a strict workflow plan as professionally as possible. Also the final form of the project was required to be implemented in a budget of approximately 100 US dollars.
As a solution for this project, we have implemented a robot which uses its two servo-driven arms to climb up a wall. The chassis of the robot was constructed using copper-coated FR4 plates on which also the double-layered circuitry was realized. The wall was composed of easily configurable “holds” placed in a grid pattern, and the robot was only allowed to use these holding points on the grid to climb up or sideways. The robot was basically composed of three different building blocks. The first block wielded the image processing task through which the locations of the holds on the walls were determined and reduced into a matrix of ones and zeros. The second block was responsible for finding the shortest path up the wall taking the robots physical constraints into account. Once a valid path is found, this block would calculate each step of the robot through the climb, compact every state into one-byte data packages, and transmit these packages into the third block, which is basically the robot itself, through a serial port connection between the robot and the computer. The robot would write these packages onto the on-chip EEPROM and interpret them as it moved up the wall.
During the implementation of this project, I was mainly responsible for the hardware design and manufacturing; coding the software on the robot which interprets the moves, controls the operation flow, and drives the servo motors; mechanical design and manufacturing of the robot; design and construction of the climbing wall and the holds and several other apparatus used in the implementation. Another highlight, which I believe is a pretty important one, is that I also was able to arrange a sponsor for the project who helped us out with supplying the servo motors and given us access to a workshop.
In the end, even though our prototype could still use some quick improvements, we were honored with a design award by the professors who were supervising this design course. Note that in addition to the functionality of the final prototype, groups were evaluated taking every step in the implementation process into account.
The robot is still functional this day, and every now and then we take it off the shelf for demonstrations.
If you are interested to see more about this project, there are two reports on this and this link including details about the robot.
Yunus Terzioglu 