Rubies
- Don't be afraid to try new things, new strategies, new sensors, new equipment, etc.
- If you are going to try something new, make sure you fully characterize it's performance before moving on to integration (e.g., ultrasonic sensors are very picky about having a short triggering pulse).
- Remember to reevaluate your gameplay strategy throughout the course of the project, especially if rules and/or requirements change.
- Pay a few extra dollars and buy an off-the-shelf sensors that contains signal processing circuitry rather than forcing yourself to solder unnecessary op-amps and comparators. Jameco has IR proximity sensors that will give analog or digital outputs. There is a huge difference in time between breadboarding and protoboarding (soldering) your standard IR detection circuit. Ultrasonic range finders are also awesome from this perspective.
- Make your drivetrain modular so that you can remove it from the robot - or at least make it easy to service.
Diamonds
- Start planning early - the full functional decomposition, concept generation/selection, and prototyping process really aided our project
- Agree on a set of conventions before divvying coding assignments
- Put in long work days before the last week before the project is due
- Always aim to accomplish more than necessarily required for the check-offs. This will ensure that you are ahead of schedule.
- Decide early if your goal is to win the competition. This affects the overall design from the beginning. If the goal is to win, allow sufficient time for coding your game strategy. Also, do not make changes before the competition without testing.
- It can be hard to reserve time to test your robot on the playing field. One solution is to find other means of testing. We did ample testing at our workstation and even tested in a hallway across from the SPDL.
- Have fun! The overall goal is to learn mechatronics, and it is not the end of the world if your robot does not win.
Sapphires
- Simple and reliable is better than fancy and broken. One means check-off, the other doesn’t.
- Remember to always keep in mind the grading checkpoints. Getting ahead is useful, but hitting the check-off is more important.
- Making the ‘bot larger than necessary pays off in ease of access, especially when you’re fixing your *insert problem part here* for the fifth time.
- Be sure to design your drive train structure well. Ask TAs and coaches about what works. If your ‘bot’s drive train is constantly being fixed, it’s difficult to get a drive control system working on time— this is the essential part of the B project.
- Turn off your robot after you’re done running under wireless SPI communication. You don’t want your ‘bot sprinting off the table (bruises and sadness ensue).
- Don’t be scared to try new things: sensors, ICs, etc. We lost a lot of time clinging to tape sensors for navigation sensing before we finally looked into ultrasonic sensors and realized they were vastly simpler from a circuit standpoint, and only a little more difficult in coding.
- Don't use old batteries or batteries from the Cabinet of Freedom, we thought it was a score to find them only to discover that they couldn't hold their charge
- A laser engraved acrylic piece lit with a tri-color LED will make any ‘bot beautiful.
Jaspers
- The right protoboard can make a difference
- Use the protoboards that are more than just a matrix of dots, make sure they are connected (acorss columns) in addition to having power rails. This will save you a TON of time when soldering. Adafruit has some good ones
- Adhere to the Gospel of Simplicity: Start Simple, Stay Simple, Finish Simple
- Don’t worry about making the best bot ever - you’re bot has to work before it can be considered the best. Don’t be ashamed to keep it simple – truth is, it’s not trivial to do something that’s simple but well done
- (contradiction alert) Don’t be afraid to try new things
- If you’ve adhered to the Gospel of Simplicity then spend some time thinking about where/how you want to stretch yourself as engineers and explore those new tools / circuits. The ultrasonic sensors were new to us but came in clutch
- No man (or team) is an island unto himself, get help from outside your team
- Our team owes a ton of thanks to Scott Purpura (from another team), it was info from him that helped us debug something we literally spent the entire night on. Don’t be afraid to ask others what they are doing or if they saw a similar problem
- Frontload the project
- Simple things aren’t easy – anyone in industry will tell you that. Spend time at the beginning of the project thinking of mechanisms and then thinking how they can be easier – it may save you some time near the end of the project