Learned From Past Years

This year, we had the lofty goal of making it to the Einstein field. Here's what we learned from past years in order to help us get there:

Weidmuller Ferrules:

We decided to standardize the use of ferrules for all Weidmuller connections.

Pit Layout:

This year, we standardized our pit layout to make setup easier.

Robot Cart Improvements:

Our robot cart was so big and bulky that at times we could hardly fit through doors! To combat this problem, we shrunk it down to about four inches high.

Strategy Improvements:

After some not well reasoned strategic decisions in 2018, we decided to make a concerted effort to increase the emphasis on data driven decisions in our strategy. This included doing qualitative as well as quantitative scouting, using scouting data to inform match strategy, and following up with any teams that had problems on the field.

Tethering Improvements:

We had issues with viewing a camera feed in 2018 while tethered. The issue is that the radio only has 2 ports, and the Rio needs one and the camera needs one. We solved this by adding a switch to the network and connecting the camera to the switch. This allowed us to just plug into the switch to tether — meaning we were less likely to forget to put a cable back because we unplugged it and also allowing us to access the Rio and camera simultaneously.

Limelight:

This year was our first year using the Limelight for vision tracking. The Limelight is easily mountable and also easy to tune, so we were able to track vision targets on the field without much extra effort.

Instant Match Replay:

During playoffs, we assign a student to each of our alliance partners. Their job is to record their assigned robot so that during strategy meetings, we can review and instantly replay a robot's every move during a match.

Software and Controls

Sensors:

NavX (gyro and accelerometer), SRX Mag Encoders, Limelight for vision tracking

Motor Controllers:

Talon SRXs and Victor SPXs controlling the 775s and BAG motors, SPARK MAXs controlling the NEOs, via CAN

Autonomous:

Bezier splines to plot paths, followed with feedback from encoders and gyroscope (pure pursuit). This was also our first year using the Limelight.

Controls:

Automatic movement to default positions; manual adjustments (within limits) and overrides if needed.

Automated Climb:

Our climb required a coordinated movement of our intake and elevator. In order to maximize the reliability, we automated this part of the climb. Our operator would engage the climb with a 2 button combination, and our robot would raise itself, and drive forward. Our operator would then let go of the buttons to lift the elevator and complete the climb.

Automated Hatch Panel Alignment:

Computer vision (via the Limelight) was used to centre the robot on a hatch panel. Our hatch mechanism used preset heights, and our auto-alignment set the horizontal position. Our driver manually controlled the speed of the drive in. The auto-alignment was used for both picking up and deploying hatches, and greatly improved our cycle time.

Performance

• Can deliver Cargo and Hatch Panels to Cargo ship and every level of the rocket
• Can retrieve Cargo and Hatch Panels from the loading station
• Can climb to all HAB Platform levels (two foot vertical climb in four seconds)
• Ten seconds per cycle
• Up to fourteen cycles per match

New Initiatives

Pre-Season Workshops:

Students built safety bots as an introduction to robotics and to build skills

Scouting App:

Custom Android app developed for gathering robot performance data

Off-Season Design and Programming Workshops:

Over the summer, we built a robot that was entirely student designed. Originally slated to compete as 9406 at Rah Cha Cha Ruckus, the robot (named Woodie, after Dr. Woodie Flowers) never quite made it there — however, it was still a great learning experience