Behind the Bumpers | 6328 Mechanical Advantage | Archimedes Division Winners

Подписаться 26 тыс.

Просмотров 4,6 тыс.

50% 1

6328 Mechanical Advantage is a leader when it comes to open build blogs like the Open Alliance. Check out the full robot overview of this Archimedes Division Winner along with their awesome dashboard, simulator and all the other software including 4 co-processors that lead to their second consecutive division win on Behind the Bumpers.
This video is supported by SOLIDWORKS. All FIRST teams can get SOLIDWORKS suite of apps for free at solidworks.com... and make sure you utilize the 3DEXPERIENCE platform to maximize your teams SOLIDWORKS experience.
Up your game by becoming a member of FUN Nation! Join this channel to get access to perks:
/ @funroboticsnetwork or subscribe on Twitch at / firstupdatesnow
Join the FUN Discord to keep the conversation going at / discord

Опубликовано:

14 окт 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 12

@kaldraganov9703 Год назад

Brilliantly crafted and ingenious software, as per usual with Jonah and 6328!

@mmry5 Год назад

Most underrated robot in all of the world championship

@sanjithudupa Год назад

Facts

@kaldraganov9703 Год назад

Totally

@sanjithudupa Год назад

6328 inspirational as always

@coopermacarthur652 Год назад

Can anybody thoroughly explain the software to track the location of the robot on the field, i have messed around with limelight 3 but not with field positions and movement from them

@oliverstivers5967 Год назад

Essentially, there are ways you can have your drivebase track where it is on the field through odometry. For example, on our robot this year, there was a SwerveOdometry object that tracked the location from the encoders in our falcons. Then, using the apriltags, you can use some camera to detect where your robot is in relation to the april tags, which will give you a field location, which can then be used to update the odometry to ensure your robot knows where it is as accurately as possible at all times.

@suryathoppae6567 Год назад

They don’t use the odometry object from wpilib exactly but make their own thing which uses a continuous kalman filter to weight their vision pose estimates and combine it with the odometry pose. Also they coded their own cv module for each camera and are sending it over NT themselves and it is set up so they configure certain constants through the Java side which is pretty convenient.

@Nity6000 Год назад

@@suryathoppae6567 Not quite. An odometry object is wrapped in the corresponding pose estimator which utilizes kalman filtering for weighting sensor readings like you said, however this is built into wpilib, they just have a slightly altered version with a few changes to aid in using multiple streams of vision data. They also did not code their own cv module, they use the existing ArUco module which performs april tag detection and pose estimation, they have a custom python program that runs on the coprocessor to execute it and handle io/config.