The three-class prototyping project requires us to get rid of the wheels on our robot cat and implement a locomotion behavior like a real animal.
At first the jumping behavior in the video shown by professor intrigued us and we intended to implement that. But soon we found we were short of materials.
I expected this material could be used as a leg for jumping because it seemed elastic and indeed when I bent it, it would restore to the original shape. However soon I found it could not even support the body (chips battery etc.). The body would slide backward and fell down.
Then we decided to change another focus. After researching online Jacky found a video on YouTube that tells the mechanism to change the rotating motion of a motor to a swinging motion of a stick to simulate legs. And we decided to focus on it.
When the motor rotates, the stick will be in a swinging motion.
We fist drew a graph (shown above) to better understand the principle behind it, then we tested it on cardboards. It worked really well.
Then we laser cut those components out to assemble real legs. By then we knew that there is a long way from succeeding in lab experiment to implement the result into real world. We faced huge amount of problems that were beyond our expectation from the beginning: The laser cut machine didn’t work; the first models were so narrow and thin that they broke; the holes for screws were zero point some millimeter smaller and the screws could not pass through; the spinning motion got stuck by the screws …
The motor got stuck by a screw.We had to solve them one by one, for hours. We rebooted the machine several times (and the order of operations really matters!); we measured and redesigned the scales of those parts for several times on Illustrator; we used drills to make the holes wider but then they became too wide that we had to use some material to fill in them… Finally we made it!
The first version.We could have stopped there. But as the body was made by cardboard, we decided to laser cut a prettier body for our robot. When we made that decision, we had never realized it would cause us the greatest trouble ever in this project.
The robot kept felling over when it moved; the motors were more easily to get stuck; a leg fell off during presentation.
(I’m so grateful to Jacky that he video recorded the successful first version.)
After class I thought through and realized the reason behind the unsuccessful second version. I found both I and Jacky ignored one key factor for our first model: the card boards are soft. Because of that, when the robot was about to tilt over and one side firstly touched the ground. the soft material alleviated most of the force so that it wouldn’t fell over; and when the head of a screw got slightly stuck under the body and prevented the movement, it could press against the cardboard and force it out.
Then there are other improvement we could achieve. For example, as professor suggested, we could use bearings in stead of screws for more smooth movement.