Authr: A Task Authoring Environment for Human-Robot Teams
summary
Authr is a ROS-based authoring and simulation tool with a web front end that encodes tasks as simplified HTA trees of therbligs (six physical therbligs) plus high-level parameters Agents, Things, and Destinations, verifies plans with Z3, and executes robot motion via MoveIT. It runs an optimization that automatically allocates agents to minimize time and effort with adjustable weights. Two user studies with engineering students show automated allocation produces more streamlined workflows and fewer allocation errors versus manual assignment, where users invent inconsistent heuristics and generate suboptimal plans. Authoring, verification, and simulation are integrated and functional.
AI-written summary · about
chapters (5)
related talks
20mAuthoring and Verifying Human-Robot Interactions
14mTowards Unobtrusive Physical AI: Augmenting Everyday Objects with Intelligence and Robotic Moveme...
21mThe Ultimate Vibe Coding Guide for Hackathons
14mAgent-Driven Software Development
31mEngineering the harness
18mCollaborative AI Engineering: One Dev, Two Dozen Agents, Zero Alignment
08mGesturAR: An Authoring System for Creating Freehand Interactive Augmented Reality Applications
10mCognitively Assistive Robots at Home: HRI Design Patterns for Translational Science