Inspiration
We both have strong passions and backgrounds in math and physics, so we have plenty of experience with deriving relationships. We decided that we could train our AI to derive the relationships for us for projectile motion in order to create a graph of its trajectory.
What it does
The program takes the input of velocity, and creates all possible pairs of angles and launch distances based on that velocity. Since the velocity is constant, it is always equal to the ratio of the launch distance to the angle, and our AI knows how to maintain that ratio at all times.
How we built it
We started with a program to solve for the distances based on the input velocities using kinematic equations, creating data points based on the horizontal and vertical distances, and plotting the points on a scatterPlot (which all formed a parabolic curve).
Challenges we ran into
Most of our issues were with the size of our variables, because in some cases, the size of our arrayList for our distance coordinates and our iteration variables were zero. Therefore, the plots were not displayed on screen.
Accomplishments that we're proud of
We are proud of the way that our graph changes as the input changes, knowing that our AI is able to maintain the correct ratio of distance to angle for a constant velocity.
What we learned
We learned the importance of tracing, because for our training, we realized that each of the training angles would correspond to the single input velocity, and the distance had to be calculated from each angle (since the velocity is always known).
What's next for Projectile Motion - Predicting Launch Trajectory
We will try to improve our training method such that it can pick a random angle from our pre-existing data set of test angles (NOT training angles) and correctly graph it based on the velocity.
Log in or sign up for Devpost to join the conversation.