Completed
Quantitative results on tactile localization
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Towards Generalization of Precise Robot Skills - Accurate Pick-and-Place of Novel Objects
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Current progress toward generalization lacks precision
- 3 My goal: precise robotic generalization
- 4 What do we need to solve precise pick-and-place?
- 5 Test time: pick-and-place on the real system
- 6 What do we mean by tactile localization?
- 7 Pipeline for tactile pose estimation
- 8 What does it mean that 2 images are close?
- 9 We compute pose error using the object's pointcloud
- 10 Tactile localization using contrastive learning
- 11 Quantitative results on tactile localization
- 12 Pipeline for visual pose estimation
- 13 Benefits of matching images with encodings
- 14 Our solutions allows different goal configurations
- 15 Lessons learned and future directions
