Project Time

I have finally finished building and programming my Puma robot. It was really rough. I first had to learn inverse kinematics, which I thought i understood until Inho gave me a test. On the test were two robots and I had to show the denavit-hartenberg parameters, tranformation matricies, and inverse kinematic solutions for them. I got through the first one fine and the second one was the puma robot. I did the forward kinematics and then started the inverse kinematics but none of my expressions were making any sense. I spent about a day working through it on my own because I wasn't allowed to ask any questions (it was a test, picture on the right). But it really just got to the point where I had no idea what to do and so I started asking other people in the lab questions whenever Inho wasn't in the room because Inho kept telling me "this is easy! this is easy!" He had checked my forward kinematics just to be sure I was on the right path. I must have talked to every single person in the lab asking little tid-bit questions asking for help until finally Minguk realized that the forward kinematics were wrong. Inho was really surprised it was wrong. When he made the table he made it the same way I did so that made me feel less stupid. So now that I had the revised parameters I went to work on the inverse kinematics again.

I worked on it all night and all morning and still couldn't figure it out though. I was struggling so hard and felt so bad about myself because Inho kept laughing and telling me how easy it was. I just got so frustrated I took a nap and when I woke up Choel offered to go through the whole problem with me, from the beginning. I was so grateful. He sat down with me and we made it about five minutes before he informed me the robot was impossible. I'm not really sure what that means but he said there was no way to do the inverse kinematics. So I was back to square one and more confused than ever.

I finally went to Inho for help who just kept telling me i was getting confused because I was talking to lots of different people and getting conflicting information, which was true, and that I should just listen to him. The big controversy was one parameter in the denavit-hartenberg table, which describes the structure and movement of the robot. But one parameter changes all of the transformation matrices and the elements of the matrices are essential in figuring out the inverse kinematics (determining the joint angles for a robot to be in a certain position). I was beginning to just feel really bad about myself and that I didn't deserve to be in Hubo Lab. I am surrounded by math and engineering masters and PhD students, why would a political science major be here? I don't even know the trig identites, crucial to discovering the solution. I kept trying to tell myself that I am trying, I learn quickly, and that I am doing the best I can but i couldn't help thinking I just wasn't good enough.

I was fianlly just so beat down I went up to Inho and said "if this is so easy could you please just walk me through it." He had the most annoyed look on his face, he was clearly getting sick of having to deal with me. He sat down and started at the beginning. After twenty minutes I couldn't tell if it was difficult for him or not so I asked him if it was still easy for him and he said no, it was very hard. We sat there for over an hour. He approached in in three different ways, moving on to the next method when he couldn't solve it with the current one. He eventually used some method I hadn't read about in any of the books and was more advanced he said. And even then we didn't get all of the angles. Inho said he accidentally gave me a robot arm that has an offset (or something...) instead of what he had intended. So I think he understood why I was running around asking so many questions. I felt so much better after he had struggled so long. I had no confidence in my math abilities when I went to ask him for help but by the end I was proud that I had put so much effort into it and gave it my all because even Inho really struggled to solve the problem.

So with my spirits renewed I set off on building the Puma in Robotics Lab, the original simulation program I was using instead of Webots. I was going to work on the finger pointing for SimonSays at Colby but Inho was adamant I do the Puma robot to get experience. Now that I am writing about it I wonder why he cared so much. I bet he regrets it now (wait till you hear about programming the robot...). There were many senior students in the lab who told me not to do the puma robot and to work on Webots but Inho would get angry whenever I suggested doing otherwise and said to only listen to him, which is really uncharactersitic of him because he is a huge clown, he never gets angry so i figured it must be serious. He was my mentor so I listened to him.

I built the Puma robot in Robotics Lab and set out on programming it. Inho gave me some sample code he used to program Hubo and I examined it to learn how to control mine. I started off ok, asking intelligent questions and got control of two out of three of the joints. No matter what I did though the last joint was constantly moving. It looked like Inho had a lot of work so I spent hours trying to figure it out on my own but I couldn't get it. I also don't know C, the programming language of Robotics Lab, so I spent most of time reading the code over and over again, stepping through it to make sure I didn't do anything stupid when I inevitably would have to ask for help. It all looked sound to me so I took a deep breath and asked Inho.

He came into the room and when I showed him the problem he said it was a very small problem and not to bother him with little problems. I felt so bad. I know he is working really hard and has his own work and taking care of the americans is not high on his to do list. I told him how long i had been working on it and he replied that lots of people have been asking him questions and it is really hard to understand someone else's code, and he is right. I kept feeling worse and worse. My guilt was magnified because Inho was totally in the right. I was being such a burden and even though it may not be my fault I was the problem, i still was the problem. Something I really appreciate about the guys in the lab is that they are very straight forward. Maybe it is because they don't know english that well so they can't beat around the bush or maybe it is cultural but they just come out and say whatever. I mean it sucks to be on the negative receiving end of this bluntness but I would rather that than not know what is really going on. Inho left the room and I was back to zero confidence and on top of that felt so guilty about irratating him so much.

Peter and I went to the conveinence store (i didn't even get ice cream i was feeling so terrible...) and when I came back Inho was sitting at my computer looking at my code. He spent an hour on it to find the problem and to make things even worse he turned to me half way through and said "i am not your teacher." I didn't even know how to respond, i was just so emotionally dead, lost, confused, feeling guilty. He figure out that there are different sets of axes in Robotics Lab and that I changed one set to make the motors oriented appropriately but it had to be a different set that we hadn't even heard of. I had no energy so I just went to sleep again so I could wake up and tackle the problem (left is the sleeping room in the lab)

I woke up feeling a lot better and went to work on fixing the axes. I ran the program and IT WORKED!! It was probably the most exctied I have ever been. I was so happy. I worked really hard and even though I got tons of help from everyone I felt like I had actually done something. Now that I can get the robot to go to a certain point I am going to try and link together movements. For example getting it to spell out my name in the air, or something like that. Who knows what emotional roller coaster this task will bring but I think I know a lot more of what I am doing. And as much as I used to get frustrated when Inho told me I needed to do this for experience, he was right. I am really happy I struggled through because it was definitely worth it.