Here is a description of the TiLAR interface as Alan Olsen left it in September 2009, and what is left to be done. First see TiLAR Interface for history and discussion of interface development.

Terminology

All icons in the actions panel can be dragged out to the work area. Each of these items (before or after dragging it into the work area) is called an <font color=“

  1. 22AA33”>action</font>.

Every grouping of one or more actions in the work area is called an <font color=“

  1. 22AA33”>action group</font>. If there is only one action in the group, then it has no border.

Functionality: Existing and To-be-implemented

  • Creating actions - Dragging an icon from the actions panel to the work area creates an action in the work area. Dropping on top of an existing action causes the actions to group together.
  • Moving actions - Dragging an existing action, moves it around in the work area. If it is part of an action group, then it is taken out of the group when the user starts moving it. If dropped on top of an existing action/group, then it groups with it/them.
  • Persistent Settings
  1. Which robot we are using (avatar, pleo, lego, troy) (done on 1/20/2010)
  2. What com port the robot is on (done on 1/20/2010)
  • Wii remote graphical interface element (almost finishing.)
  1. Push buttons on Wii remote to assign actions
  • (Check) Right now to display an emotion, must also do a sound or motion. Should be able to assign a Wii button to just display an emotion without also having to run a motion or play a sound
  • Create a special start node that cannot be deleted and is created whenever a new file is created.
  • Change toolstrip buttons so they don't stay checked after you click them, or so exactly one item is checked at any time (cannot have both selected, cannot have both unselected).
  • Check to see if power button can be used on wii remote.
  • Make it so sounds and emotions can be played any time, even while the robot is moving.
  1. Might need to make global wii remote actions.
  2. Should we create a separate thread for each of sounds, emotions, and motions?
    1. This is so we can, for example, play the sad sound and show the sad emotion for a few seconds, then go back to neutral, but still allow another action to happen before the animation is done.
  • Undo!
  1. Record history
  • Make deleting actions more consistent or intuitive. Maybe a trash icon they can drag nodes into? Make a trash cursor (like a toolstrip button mode), so they can click on nodes to delete them.
  • Make separate sound “action”
  • Make different shapes for different node types.
  • Take screenshots of happy face animation (done on 1/8/09)
  • Use Honda's MotionLearner to record new actions for the Avatar
  • Make new GUI portion to click and drag the avatar to new positions
  1. Change avatar positions with gui sliders (move slider to rotate arm)
  2. Save animation to use in the regular TiLAR interface
  3. Click and drag on hands, arms, and use inverse kinematics to calculate joint angles. Instead of using traditional keyframe management, call them poses, and have the option to “photograph” a pose (essentially save a keyframe) or to automatically capture a pose whenever the robot has not moved for several seconds.
  • use Honda's MotionLearner to record new actions for Troy and Trevor.
  1. Translate FUJ files into format that troy and trevor can understand
    1. Get vectors between shoulder and elbow, etc., and calculate angles between vectors.
  • Low-priority bug: First run after booting computer crashes. Second run is fine.
  • Automatically check for dead-end nodes, and tell the user that there are dead end nodes, maybe indicate which nodes are dead ends with a red squiggle line underneath like spell checkers.
  • GUI - Make new GUI portion to click and drag the avatar to new positions.
  1. Change avatar positions with gui sliders (move slider to rotate arm)
  2. Save animation to use in the regular TiLAR interface
  3. Click and drag on hands, arms, and use inverse kinematics to calculate joint angles. Instead of using traditional keyframe management, call them poses, and have the option to “photograph” a pose (essentially save a keyframe) or to automatically capture a pose whenever the robot has not moved for several seconds.

« Discussion Page

ar/tilar-alpha.txt · Last modified: 2014/08/13 20:11 by tmburdge
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0