Accomplishments

Back

IMU Turn

Jumping off from the programming efforts of last year, we have developed an algorithm that utilizes the IMU sensor in the Control Hub to assist in the robot’s ability to turn to a specified angle with a given error of half a degree.

Power Ramp

With the help of our sister team’s coach, we have integrated an algorithm that slowly increases and decreases the speed a motor runs when going to a set position to reduce the error inertia would otherwise have.

Camera

With autonomous being a significant factor in ranking, we have taken on the challenge of using a camera for object detection and april tag detection. We figured using a camera would increase the speed and accuracy of the robot during autonomous. Opening up new possibilities like getting more pixels and allowing for our alliance partner also to complete their autonomous.

Object Detection

We developed a Tensorflow Object Detection model using the FIRST Machine Learning Toolchain. We use it to quickly identify where our team prop is and determine the appropriate spike mark. We also aid its accuracy by using a color and distance sensor.

April Tag Detection

With some research into the Software Development Kit that FIRST provides and the use of April Tags, we have developed an algorithm that utilizes the values returned from the pose estimates to increase our accuracy in scoring the yellow pixel during autonomous.

Odometry

To start with Odometry, we first wanted to test it on a separate robot. This would also allow others to practice coding on a separate robot so that their robot could be worked on. We decided that since we were new to odometry, we would order the goBilda odometry pods because we heard from multiple sources online that they were great and worked very well.

Road Runner

Using roadrunner has been a goal for our team. We knew that using it would take a lot of time and effort to get it up and running, so we started off by doing research. In this research, we found that the longest process would be tuning. What we decided to do was to let our mentor, Alex, lead the development for our team by first applying it to the separate test robot. With the help of the makers of learnroadrunner.com, we were able to slowly understand and perform the tuning and path building required for Road Runner. The use of Road Runner has proven to be quite beneficial as it has increased our speed and accuracy in autonomous.

Page updated

Report abuse