ME 210: Mechatronics
S.K.K.P
Structure
We used a tiered design made from laser-cut Duron, threaded rods, and hex nuts. The bottom layer was used for the drive train (motors, batteries, ball bearings). The first and second layers held our electronic components. The topmost layer was used for reloading and depositing the balls into the shooting goal. We used heat shrink and tape to keep our circuit tidy and secure.
When designing our bases, we started prototyping very early with foam core–creating several iterations to account for new components as we integrated limit switches and IR sensors. In retrospect, this consumed much time, as we had to remount circuits onto a new chassis every time.
While testing, we found the structure unstable, as the nuts loosened due to the motors' vibrations. To alleviate this issue, we implemented hex nuts and Loc-tite.
Flag Arm
To meet the second component of the challenge (making contact with the Contact Zone), we incorporated a flag arm attached to a servo motor. Doing so allowed us to make contact by using the arm to sweep thread along the top of the Contact Zone, as opposed to having the main frame of the robot collide with the block.
Secondly, the flag arm served as our celebratory action which was required to signal the end of our 2-minute-and-10-second run of the obstacle course.
Placing the components of the robot as well as the key landmarks of the course into a Fusion360 CAD assembly was helpful in making sure the hardware components of the robot would interact with its environment as expected — that the bottom layer would be the point of contact with the wall and that the flag arm was properly positioned above the Contact Zone — as well as being able to verify the robot stayed within the size constraint.
Flag Arm - Positioning Relative to Contact Zone with Embroidery Floss (drawn in yellow)
Preliminary Flag Arm Sketches - The flag arm was initially designed to spin vertically, but switching to horizontal spinning allowed us to more easily meet the 12" x 12" x 12" overall dimension constraints.
Ball Release & Dropper
Addressing the third component of the challenge (dropping balls into the Shooting Goal), we chose to house the balls in spinning flywheel driven by a servo motor attached to its center.
With the servo motor's range of motion being limited to 180 degrees, five spaces for the balls were arranged in a semicircle with openings at the ends so that the balls could be deposited on either side of the robot.
Shoot Goal Scenario - Robot Positioned with Slides Above the Shooting Goal
Flywheel - Top View with Cut Outs for Balls (highlighted in blue)
If we run into drive issues as the robot turns away from the Contact Zone and starts making it's way to the Shooting Goal (see "Inconsistent Drive" video below), it might not be able to make as tight of a turn as we'd like.
To account for this, we wanted to make sure that the balls could traverse the potential gap that'd separate the side of the robot and the goal. To do so, our goal was to maximize the horizontal distance the balls could cover, so we maximized the height of the slides (while leaving enough room for the wires on the bottom layer), set the exit slant to horizontal, and extended one slide to the robot's max dimension.
Having this asymmetrical setup (see "Side View" image below) gave us the option of using the extended slide if, worst-case, the gap was too large for the shorter slide to work. For check-off purposes, we could choose which side (A or B) of the obstacle course we ran our robot on, so we could choose from which side the robot approached the Shooting Goal and, thus, which slide was used.
Slides - Side View
Slides - Profile (highlighted in blue) with Shooting Goal
Drive Train
Issues Encountered
Inconsistent Drive & "Jumping" Off Ground
We used the Greartisan A6587 DC motors with 60 RPM, 100 mm, and Polulu scooter wheels. Our motor did not give us the fastest speed during the competition. However, after multiple iterations of testing, we found that a slower speed was better for accurate line following and easier to control turns.
Initially, we ran into multiple drive issues: motor stalling and the robot "jumping" off the ground (see videos on the right). One of the solutions we used was to add two casters with ball bearings for balance issues; it was important that the ball bearings were at the correct height. When our ball-bearing casters were too high, the drive would stall shortly after a run.
Robot Stalling
Foam-Core Prototype
For our approach to prototyping, we chose to prioritize conducting tests as early as possible and throughout the entire process. We incrementally incorporated the mechanical, electrical, and software aspects to our robot (rather than all at once) to give us the best chance at success, especially given this project's short time frame.
