Thesis Defense: Optimistic Active Learning of Action Models for Robotic Manipulation
, EECS, CSAIL
Date: Friday, April 22, 2022
Time: 3:00 PM to 4:00 PM Note: all times are in the Eastern Time Zone
Location: 32-G449 (Kiva)
Event Type: Thesis Defense
Host: Leslie Pack Kaelbling, Tomás Lozano-Pérez, Nicholas Roy
Contact: Caris Moses, email@example.com
Speaker URL: None
TALK: Thesis Defense: Optimistic Active Learning of Action Models for Robotic Manipulation
Manipulation tasks such as construction and assembly require reasoning over complex object interactions. In order for a robot to successfully plan for, execute, and achieve a given task these interactions must be modeled accurately. Existing methods for engineering these dynamics models fail to accurately capture underlying complexities such as friction or non-uniform mass distribution. Therefore, in this work we leverage a data-driven approach to acquiring action models. We propose active learning strategies which aid the robot in learning action models efficiently, with the ultimate goal of using them with a planner. Additionally, we supply the robot with initial optimistic action models which are a relaxation of the true unknown transition model and are easier to specify that fully accurate action models. We are generally interested in the scenario in which a robot is given an initial (optimistic) action model, an active learning strategy, and a space of domain-specific problems to generalize over. In this talk I will present several active learning strategies for manipulation tasks which leverage optimism. I will also give results in several domains involving complex object interactions such as block stacking and tool use.
Created by Caris Moses at Thursday, April 14, 2022 at 5:22 PM.