This past week, we’ve been navigating the transition to a design and simulation based project. We’re still working out a few of the details, but we’ve largely pinned down a final direction.
The three of us will be working in conjunction on receiving data from the IMUs on the user’s hand and on Kalman filters. As the IMUs are easily portable, we can all work together on this hardware aspect of the project remotely. As we still have over $300 left in the budget, Dan and Luca will also be ordering the Particle Photon (Hana already has one) to help with the transmission of data over WiFi. Kalman filters allow us to progressively get rid of noise data from the IMUs, and are especially well suited to retaining accurate data on any speed or directional changes. Together, we will also be working on building a software package for the simulation. There are currently existing open source solutions that we will be using to our advantage, but we will largely be building the simulation on our own. Utilizing Processing, the simulation will show the ball traveling through the air as predicted by the pre throw, and the robot moving to the predicted location. We will be displaying the ball’s speed, angle, and direction, and the motor controls for the robot’s movement. Testing and verification is a bit tricky, but we hope to solve this by looking at the predicted location from the pre throw, and comparing that to the actual landing location mapped back into the simulation, for example. We also plan to meet very often over Zoom/Google Hangouts to ensure we are on the same page at all time, as communication is key for a remote project like this.
0 Comments