From reading other CD threads it seems as if our team is taking the long and hard road by using a propulsion based launching system using springs instead of using a friction based design.
We are planning on being able to shoot from any angle and having the “turret” of the robot rotate to track a selected vision target the entire time unless another command is given through the dashboard.
What we are trying to figure out is since we are loading out of the feeder slot and going across the field how to do we recreate the same launching conditions each time. We have talked about using the ultrasonic sensor and the photo-sensors combined to detect where it is on the field or if the driver could get the robot close to the same point each time then initiate “auto-park” mode where it goes to the same position right outside the key each time. So that is one of the problems that we are trying to work through right now.
Another problem is calculating the trajectory of the ball and accordingly adjusting that to the robots position on the field or away from the fender.
We are kind of stuck so any influence at all would be great.
Our team has a similar idea. We are using the ultrasonic sensor and camera to find out how far away we are from the hoops, then you can calculate trajectory with the equations of motion. The position vs. time equation can be modified to include the drag coefficient of the ball when you launch it, and the angle and range of the trajectory can be calculated using either:
The nature of trajectory is release angle and velocity. Check out “Pumpkin Chunkers” and you will see a lot of different launching mechanisims. There they are not going for accuracy, but just distance. In order to gain accurracy, you need tight control of angle and velocity. Not knowing your mechanism is difficult to advise on this. Old war engines (trebuchets, catapults, giant cross-bows) often had mechanisms to tune in distance. On catapults and trebuchets, angle of release was usually a bit more constant, and velocity was adjusted by controlling energy input (balance weight in trebuchet, or amount or draw on catapults and cross-bows).
Catapults worked well as seige machines, because the targets and launch position were stationary, and many projectiles could be launched and adjustments made afterwards. In this game, the target is stationary, but the launch position will likely be quite dynamic. You will need to practice a lot with your launcher and understand the parameters to adjust to meet the range your sensors find. I believe once you have repeatable, and adjustable launch energy, then you can use a look-up table in labview to adjust your mechanism.
Another thing to keep in mind, many “catapult” or spring loaded launchers have latching mechanisms that lock in a particular direction. This works great to increase range of shot, but often is very difficult to adjust from a longer to short shot. Keep that in mind while designing your launcher.
Last and certainly not least, make sure your device is safe to “dry fire”. A “dry fire” is releasing the launcher without its payload. Often the additional mass of the payload keeps mechanisms from detroying themselves.
Good luck, and have fun. Please post your stuff up after season as I think this will be a fairly unique machine.