- adjust climb arms to be same by default
- fix issues with manual
- time delay on arm lift so turret safe
- reformat Shuffleboard (require driver-station)
- augmented loop (pg 109 of textbook)
- detla U: add voltage into state model
- https://www.chiefdelphi.com/t/what-impressive-things-did-you-do-in-software-this-year/382245/48?u=tatestaples
- https://github.com/Team254/FRC-2020-Public/blob/master/src/main/java/com/team254/frc2020/planners/DriveMotionPlanner.java#L225
- current management
- test other turret controls (it they want to redo tether, than recharacterize) (2 hours) **
- velocity
- state space (hard with bad ff)
- test shot while moving (6 hour)
- test why auto shot slow (3 hour)
- generally make auto better
- flash other limelight (try) (4 hour)
- tune ball detection
- climb prepare (30 min) [Wednesday]
- closed loop climb testing
- remove drivetrain swap
- climb stablization (30 min)
- Chris driver practice (whatever time remains) [Thursday]
- Auto Drive to Climb
- Auto Climb
- goal system (shouldn't be too hard - should update) -> moderate difficulty
- limelight
- obstacles (can work on this) -> easy, just time intensive
- ball detection in hopper
- figure out colors senrsors (talk to bots on wheels) -> with resourses, not that hard
- pypico + 2 rev color sensors
- ability to reaccesss if lost for too long -> simple
- robot avoidance -> hard (try installing photon on picam or something)
- climb -> unknown, probably not bad but will need empirics
- should add endgameInit or something (idk)
- Make sure things are isolatable (specifically autoDrive and Navigation)
- think about roborio2 (they said 30% faster)
- probably get falcons
- Auto
- pathweaver option
- simulation testing (how it works)
- playback feature
- Auto pathing
- obstacle avoidance
- mention semester independent project (automation as my passion)
- auto turret
- auto climb
- early prototypes of auto shot and intake
- pathweaver option
- Driver Simplicity
- custom driver controls (maybe cut)
- intake retraction (be breif)
- no penalties
- shoot from anywhere
- auto turret
- limelight
- pose estimator
- control
- wire management
- side spin
- back spin
- auto turret
- Additional features (things that really put us apart)
- driving
- driver camera
- continually updating cameras
- shooting
- shooting while moving
- disposal
- climbing
- prepare climb
- auto climb
- bonus
- manual overrides
- every motor
- simplicity if errors
- manual overrides
- driving
- Kyber (8.5k lines of code - 75% of all code can be immediately reused)
- motivation
- personal journey in understanding
- documentation for future year
- I am a senior
- motor control systems integration
- purpose
- rapid iteration
- can write a bunch of different control schemes and swap them out to see what worked
- had 6 different control schemes for flywheel at one point
- single line of code change to switch motors
- allow for simple simulation support
- can write a bunch of different control schemes and swap them out to see what worked
- personal learning
- usability for new coders
- rapid iteration
- modes
- percent
- position
- velocity
- follow
- current
- torque
- music!!
- control types - each able to be setup in a single line
- native controls
- pid
- ff (intuitive to use)
- statespace
- positon, velocity, dual, arm, elevator, drivetrain
- custom loops with latency compensation
- bang bang
- take half back
- custom
- purpose
- Simulation support
- why important for our robot (time)
- 3 days of testing
- easy way to write code for subsystems not there
- simulated ESC
- fake solenoids
- halucinating climb past 2 events
- why important for our robot (time)
- motivation
- unit libraries **
- greater understanding and readiblity
- custom polar coordinates
- nav updates
- turret adjustments
- Characterization
- use our own code to characterize systems and fit the data
- allows for more control of the data we get and ensures safety on constrained systems
- Expanded Trajectory generaion
- obstacles
- Informed Rapidly Expanding Random Trees
- illustraion
- optimizations
- Traveling Salesman
- brute
- nearest neighbors
- branch and bound
- general graph class
- Informed Rapidly Expanding Random Trees
- goals
- CommandManager
- obstacles
- Playback System
- recording inputs
- overriding the controller to play inputs
- Navigation / Dynamics
- integrated and simple to use
- built in for all drivetrain types (diff, mecanum, swerve)
- drive chassis
- drive auto
- update auto
- simulate motors
- calibration
- debug system
- allows for graphing and understanding of what everything is doing
- fancy controller modifications
- drone modification
- LED animations
- 11 distinct animations
- sim support and simplicity
- math
- multiple coordinate systems
- robot rel
- hub rel
- polar
- threeD geometry
- multiple coordinate systems
- builtin mechanism (personal learning and example building)
- drivetrains
- elevator
- arm
- flywheel
- motor control
- one line fix to swap out Neos (before musics)
- experimental statespace controls
- latency compensation
- delta U control
- simulating motors
- Feedforwards
- linear systems
- simulations
- What needs simulation?
- actuators
- KSolenoid
- KMotor
- LEDs
- sensors
- PhotonCamera
- KGyro
- actuators
- Practical uses
- coding in 3 days
- make sure nothing hits
- make sure everything moves at the speeds they should
- linear systems are really important for this because they allow for guessed simulations before getting to touch the actual motor
- What needs simulation?
- tutorials
- I am a senior
- setup by step
- what is code -> Control Theory
- UcoSLAM
- the research process
- talk about other approaches and their limitations
- the website
- relation to our auto Intake prototyping
- how it works
- mapping the roof
- the research process
/**
- Definetly
-
- east
-
- hedgehogs
-
- triple strange
- Good:
- sequence
- omegabytes
- pitt pirates
- zebracorns
- platypi :(
- hawtimus prime?
- yeti
- bots on wheels
- green hope falcons (5190)
- gear cats
- I want
- collect all the data from NCFIRST matches
- teach me all of the programming required for FRC
- predict all the matches for States
- predict potential from top teams
- graph trends for all the teams over matches
- predict our potential */