The students of That ONE Team #4967 had an awesome rookie season, and learned a lot. Here are their thoughts:
Always check all of your electrical connections to make sure they are plugged into the correct places. We learned that we need to do this more often because we were having power issues with the radio/cRIO for about half of the build season before we figured out that they were plugged into the wrong places.
We learned to leave as much time as possible at the end of build season for drive practice-the more the better. Just don't rush the robot build so much that you skip out on some abilities that you could have made it do in order to practice more. Plan your time well.
We didn't make solid prototype until the 3rd week. Build a robot prototype chassis as soon as possible. If we had done that, we would have had more time to improve our ideas and would have had the improvements we had for the State Championship for our first competition.
Solder as many connections as possible. We had many connections fail due to bad crimps and other reasons. If we had done it right the first time, by soldering it, it would have saved much time and stress.
Create a clear strategy and vision for what you want your robot to achieve before discussing how it should be designed. Having an ideal end goal in mind helps the design process and keeps it on task. Our robot design strayed far from what our initial strategy was, and it would have been better had we had a clear strategy first.
Document your work as you go, record important information such as robot IP address, and router and camera passwords. Label your electrical, and create a digital diagram to refer too when troubleshooting. We didn't write down our camera IP and password, so when we tried to get into it to change a setting we couldn't. Having the electrical diagram helped us when something was going wrong, and we needed to remember the location of particular wires.
Build season is when things start to get crazy, and it gets confusing between the members of the business team if you don't have a clear planned out strategy. We could have planned better.
Have a strong business team. We only had two people on our business team and that meant all the pressure was on us. Everything business-wise split between just two people became very stressful.
During Aerial Assist, we had a very abstract robot design. The robot itself was the prototype and it just sort of got put together over the six weeks. I feel it would have been better for us to build a prototype robot separately, and then build our competition bot during the fourth week of build season. This gives you enough time to tweak and get drive practice. Have a strong Strategy and Design/CAD team to start the first two/three weeks of build season. Also, build yourself a robot made of something other than metal, and test, test, test!
We learned a lot about scouting. Have a very strong scouting team, and multiple different methods of scouting, subjective, numerical, etc... While our scouting was helpful to the drive team, it would have been beneficial to have a more structured scouting plan. This way you can be more prepared for alliance selections, and make the best selection possible or promote your team to other teams better.
One of the things that I learned was to start prototyping early on. It took a while for us to actually start prototyping. Then once we did, we did a whole lot which took up time we have spent on other important things..
There were a couple times where we could have done something to win the match, but didn't think of it at the time. During a match, go through all possible scenarios in your head.
From one of our mentors:
Prints should be generated for all parts being manufactured by outside sources. We had some parts that were made with the wrong material. One time was the manufacturer's fault, one time was ours. Had we put together build prints along with the Solid Works model, the print would have likely went to the build floor with the material called out on it. Also in generating a print, it forces you to dimension the part which forces one to take a second look at the design. In the first instance, the material was mentioned in an email and did not make it out to the person building the part. In the second case, the model was incorrect.
We were late in ordering cylinders and other parts which delayed the initial build. Earlier build would have left us more time to find the flaws in the design early and redesign.
Our up to date design data was not readily available to all team members. This made it more difficult for more members of the team to contribute to the design. Access to the latest design data would have facilitated idea generation as well as reviews of the design. In the future, the use of a repository for code and maybe a dropbox or the like for CAD would make the design data available to everyone, anytime anywhere.
The lack of materials such as metals, tubes, cylinders, motors etc. at the build space slowed down idea generation and prototyping.
Lesson learned: Prototyping served us well. I firmly believe that prototyping, testing and tweaking made all of the difference. It is essential to a feel for what you are trying to do. Maybe after 5 years of exercise ball chucking, you could go from print to perfection. I was recently at Calvin College's senior engineering project day. The most popular lesson learned from these college engineering teams "Get a prototype put together earlier so you have something to work with."
We were limited by having one robot. When our robot was bagged, we didn't have a CRIO and other hardware to continue development and continue learning. Given the budget, having a second similar robot would have been very beneficial.
Some design decisions kind of just happened which may have resulted in a less than optimal solution. I think a few more structured "design only" meetings with all present to go over design options would have been beneficial. This would also have the effect of being more inclusive of all team members. Kind of like periodic "design reviews."
Not having enough hardware such as the CRIO made it less accessible for others interested in learning the ropes.
Mentors (I am referring to myself) need to put more effort in earlier to know the rules inside and out. There were several times where I went down some wrong paths because I didn't know the rules well enough - time wasted. Thankfully the students were on top of it.
Some of the learning we did was later than it should have been which increased the workload during the build season. There was quite a bit of learning - software and CAD that could have been more seriously started before the build season.
Always check all of your electrical connections to make sure they are plugged into the correct places. We learned that we need to do this more often because we were having power issues with the radio/cRIO for about half of the build season before we figured out that they were plugged into the wrong places.
We learned to leave as much time as possible at the end of build season for drive practice-the more the better. Just don't rush the robot build so much that you skip out on some abilities that you could have made it do in order to practice more. Plan your time well.
We didn't make solid prototype until the 3rd week. Build a robot prototype chassis as soon as possible. If we had done that, we would have had more time to improve our ideas and would have had the improvements we had for the State Championship for our first competition.
Solder as many connections as possible. We had many connections fail due to bad crimps and other reasons. If we had done it right the first time, by soldering it, it would have saved much time and stress.
Create a clear strategy and vision for what you want your robot to achieve before discussing how it should be designed. Having an ideal end goal in mind helps the design process and keeps it on task. Our robot design strayed far from what our initial strategy was, and it would have been better had we had a clear strategy first.
Document your work as you go, record important information such as robot IP address, and router and camera passwords. Label your electrical, and create a digital diagram to refer too when troubleshooting. We didn't write down our camera IP and password, so when we tried to get into it to change a setting we couldn't. Having the electrical diagram helped us when something was going wrong, and we needed to remember the location of particular wires.
Build season is when things start to get crazy, and it gets confusing between the members of the business team if you don't have a clear planned out strategy. We could have planned better.
Have a strong business team. We only had two people on our business team and that meant all the pressure was on us. Everything business-wise split between just two people became very stressful.
During Aerial Assist, we had a very abstract robot design. The robot itself was the prototype and it just sort of got put together over the six weeks. I feel it would have been better for us to build a prototype robot separately, and then build our competition bot during the fourth week of build season. This gives you enough time to tweak and get drive practice. Have a strong Strategy and Design/CAD team to start the first two/three weeks of build season. Also, build yourself a robot made of something other than metal, and test, test, test!
We learned a lot about scouting. Have a very strong scouting team, and multiple different methods of scouting, subjective, numerical, etc... While our scouting was helpful to the drive team, it would have been beneficial to have a more structured scouting plan. This way you can be more prepared for alliance selections, and make the best selection possible or promote your team to other teams better.
One of the things that I learned was to start prototyping early on. It took a while for us to actually start prototyping. Then once we did, we did a whole lot which took up time we have spent on other important things..
There were a couple times where we could have done something to win the match, but didn't think of it at the time. During a match, go through all possible scenarios in your head.
From one of our mentors:
Prints should be generated for all parts being manufactured by outside sources. We had some parts that were made with the wrong material. One time was the manufacturer's fault, one time was ours. Had we put together build prints along with the Solid Works model, the print would have likely went to the build floor with the material called out on it. Also in generating a print, it forces you to dimension the part which forces one to take a second look at the design. In the first instance, the material was mentioned in an email and did not make it out to the person building the part. In the second case, the model was incorrect.
We were late in ordering cylinders and other parts which delayed the initial build. Earlier build would have left us more time to find the flaws in the design early and redesign.
Our up to date design data was not readily available to all team members. This made it more difficult for more members of the team to contribute to the design. Access to the latest design data would have facilitated idea generation as well as reviews of the design. In the future, the use of a repository for code and maybe a dropbox or the like for CAD would make the design data available to everyone, anytime anywhere.
The lack of materials such as metals, tubes, cylinders, motors etc. at the build space slowed down idea generation and prototyping.
Lesson learned: Prototyping served us well. I firmly believe that prototyping, testing and tweaking made all of the difference. It is essential to a feel for what you are trying to do. Maybe after 5 years of exercise ball chucking, you could go from print to perfection. I was recently at Calvin College's senior engineering project day. The most popular lesson learned from these college engineering teams "Get a prototype put together earlier so you have something to work with."
We were limited by having one robot. When our robot was bagged, we didn't have a CRIO and other hardware to continue development and continue learning. Given the budget, having a second similar robot would have been very beneficial.
Some design decisions kind of just happened which may have resulted in a less than optimal solution. I think a few more structured "design only" meetings with all present to go over design options would have been beneficial. This would also have the effect of being more inclusive of all team members. Kind of like periodic "design reviews."
Not having enough hardware such as the CRIO made it less accessible for others interested in learning the ropes.
Mentors (I am referring to myself) need to put more effort in earlier to know the rules inside and out. There were several times where I went down some wrong paths because I didn't know the rules well enough - time wasted. Thankfully the students were on top of it.
Some of the learning we did was later than it should have been which increased the workload during the build season. There was quite a bit of learning - software and CAD that could have been more seriously started before the build season.
