Choreorobotics: Teaching Robots How To Dance
summary
Argues that movement makes robots legible and emotionally engaging, and demonstrates a deployed system: six Everyday Robots mobile manipulators running a Python parent/client stack at ~20 Hz, a Boyd-style flocking base with tunable weight modes (gains) for cohesion separation alignment plus circle/line/follow behaviors, imitation learning that trains an agent to pick those weight modes from choreographer-labeled examples, and music mode maps joint velocities to sound; gesture-to-action mapping: left hand bell, right hand rotate, hands-together look up. Visitors, especially kids, respond positively in qualitative tests.
AI-written summary · about
chapters (10)
- 0:00Introduction to Robotics
- 1:08Robots as social agents
- 3:50The importance of movement
- 10:13Audience movement exercise
- 13:46Choreography and robotics
- 20:12Conveying intent and emotion
- 25:41System design for robots
- 29:40Programming robot behavior
- 33:59Performance and feedback
- 37:21Conclusion and Q&A
