This week, our team updated our plans and Gantt chart for the remainder of the semester and began implementing the Kalman filter head on. Immediately running into issues related to bias drift, we are currently trying to resolve these issues with calibration and adjusting the Kalman filter accordingly. We are also currently ordering relevant parts such as hacky sacks and Particle Photons. Next week, we will be focusing on the gyroscope portion of the Kalman filter, performing 3D calibration, and beginning the specifics of the Processing simulation. The updated Gantt chart is included below.
Risk Management
In case of insurmountable obstacles related to obtaining relevant data, we will be creating our simulation as soon as this coming week with dummy data that we hope will model our predicted throws.
We are also aware that we may not receive additional parts (i.e. Particle Photons) on time. If this is the case, group members that already have Arduinos will use that software in order to process the data.
If we are unable to reliably solve the issues related to 3D calibration and sensor drift, we will find ways to simplify our problem in a 2D sense and record data for the shortest amount of time possible. This may mean the elimination of planned items such as the pre-throw. I have included an image of the existing drift issues we’re trying to solve (for an IMU sitting still)
0 Comments