Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Serving Drinks Scenario #109

Closed
1 of 15 tasks
henrikschnor opened this issue Mar 5, 2019 · 1 comment
Closed
1 of 15 tasks

Serving Drinks Scenario #109

henrikschnor opened this issue Mar 5, 2019 · 1 comment
Assignees

Comments

@henrikschnor
Copy link
Contributor

henrikschnor commented Mar 5, 2019

This is kind of a meta issue for the "Serving Drinks" scenario of RoboCup 2019 to keep track of what needs to be done. I will open more issues for every point listed below.

In "Serving Drinks" the robot has to find all party guests without a drink, take orders and deliver the drinks.

Things to clarify

  1. Decide how to deliver drinks: The robot can either hand over drinks or deliver them with an attached tray. If we use the tray, we need an extra action for detecting if the guest takes the wrong drink. Do we have a tray at all? If we hand over the drink (from a tray, or directly from the bar? -> clarify), we need to implement an action for that as well.
  2. Implementation design choice: Should people be actively detected (i.e. should we implement an action for that) or should we passively always try to detect people and store information about them (location, attributes) in the knowledge base to be retrieved when needed? This might be related: Continuous scene perception for accurate world modelling #73

To do

  • Create a state machine for the scenario (see Define Scenario "Serving Drinks". #106)
  • Perception (todo: clarification, create issues)
    • Create action "find_people" for finding all persons in a room / in front of the robot, also remember their faces and possibly other attributes like clothing or what they are holding (identify if holding a drink)
    • Train object recognition with drinks we need to identify
  • Create action "dialogue" for having simple conversations with people. Integrate NLP/Rasa for this (todo: create issue)
  • Manipulation
    • Create a behavior for picking up specific objects from a table
    • Create an action (or behavior) for handing over objects. Maybe we can just use the "place_action" for this.
  • Create action "choose_action" for serving drinks, to decide what the robot is going to do next (get order / get drink / is done)
@henrikschnor henrikschnor changed the title Serving Drinks (RoboCup 2019) Serving Drinks (German Open 2019) Mar 5, 2019
@argenos argenos added this to the German Open 2019 milestone Mar 5, 2019
@argenos
Copy link
Contributor

argenos commented May 23, 2019

Can you please update the to-dos for this issue?

@argenos argenos changed the title Serving Drinks (German Open 2019) Serving Drinks Scenario May 23, 2019
@argenos argenos removed this from the German Open 2019 milestone May 23, 2019
@argenos argenos closed this as completed Jun 14, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants