Final Class Blog

This is my last blog entry as the robotics class finishes its journey of five months. Throughout the semester, I've learnt from the basic knowledge to the application of those knowledge by developing my ideas and connecting each of them in my unique ways. What are taught are mainly three: robot construction, programming the NXT and using each type of sensor. I want to talk about them, and review what I've gained from the class.

Robot construction may vary according to the purposes that the builder has on the robot. However, basically it is a general thought that a robot should be stable whatever its purpose is. Adding more parts doesn't help much in efficient ways, even it can harm its own robot. For a wheel, adding a beam between an axis and the motor helps it not to disturb the motion of the wheel much. The general shape of the robot should be built with pairs of parallel beams. The NXT which is the heaviest part of the robot should be placed lower, and also it is better if the robot is low like racing cars. Other than those, you can build in whatever shape you like so long as you've considered the efficiency.

Programming the NXT takes part in the robot like a brain. It orders and lets the robot know what to do and perform exactly. The only caution of programming is that if you don't fully understand what you've set and you don't consider all the factors hidden in the track, it might behave according to whatever you programmed. The other would-be caution is that you've got to match the ports.

The programming has mainly two sorts of blocks: action, sensor. The action may not vary much; there are only going(forward, backward or speed can be modified by you) and halting. The sensor blocks can be light, sound, ultrasonic, touch and rotation. The first four are used to help the motion of the robot by starting it or acting to obstacles. Rotation sensor just determines how far the robot goes. The programming looks simple, but it's really difficult thinking that it really follows EXACTLY what you've set. There might be some factors that you've not considered yet.

Throughout the course, I feel that I've learnt a lot in the ways of application of what I learnt like math and science, communication and cllaboration. The calculation of how far the robot should go was extremely confusing and a bit hard, but it was really helpful for me to understand uses of math. And other than that, how science affects even a small robot's behavior; friction of the carpet really confused the challenges and it really made itself be a real challenge. The other lesson that I've learnt from the course is the communication which helps reducing errors. Though I don't get an idea or I made a mistake, my partner would cover it since he knows. Like the previous case, for me it has been always helpful. Anyway, I don't regret my decision to take this course.

Tractor Pull

The tractor pull challenge is to construct a robot that will push or pull the most weight 50cm in a minute. Basically, for the tractor pull, high gear ratio is needed for more strength, and it is needed to add more weight on the robot so that the wheels have more grip which helps use whole energy transferred to the wheels. 

For the robot of our team, since we decided it to push the weight, I added some tricky device to keep the weights from moving aside. NXT was, of course, moved to the back, and one more motor is added at the back for more grip and for deco. By the way, the gear way is 40:8 which is reverse to the gear ratio used in the drag race. Until now, the maximum weight the robot could've pushed is about 3000g.

Challenge : Drag race

The challenge is to build a robot to be the fastest in the 3m track. It didn't need sensors except the sound sensor for fair start, but it required some techniques of construction and knowledge of gears. 

We first came out with the idea of the gear train. For more efficiency, we used the gear train of four gears with the same gear ratio, 1:40. Even I tried to build the gear train of six gears on each side, but Loren asked me not to do so, which was lucky. In the first trial, it couldn't move anyhow, which was because it couldn't bare its own weight. We had to take out the second parts of the gear trains; the robot functioned well. To prevent it from flopping, a back wheel was added. Of course, it was programmed to have a maximum speed.

Luckily or as a result of our brilliant techniques, we won the second place.

Gears and Speed investigation summary

In the project, "Gears and Speed", it had the purpose to find out what hypothesis is correct and how we can apply the hypothesis to the real speed of the robot. There were two hypotheses:
Hypothesis A : speed1/gear ratio1 = speed2/gear ratio2
Hypothesis B : speed1 * gear ratio1 = speed2 * gear ratio2
"A" was directly proportional, which means when one thing increases, the other thing increases as well, and "B" was indirectly proportional, which means when one thing increases, the other thing decreases.
I had experiments with the gear ratio from 16:16 to 20:12 and 12:20, and I tested each of the hypotheses by calculating based on the data from the experiments. As a result, the hypothesis B was correct since its assumed value was apparently quite close to the real one. It does mean that for fast speed, the gear ratio is supposed to decrease, meaning the size of the driven gear is smaller while the driving gear should be bigger. As the gear ratio is less, the speed is faster.