Georgia Institute of Technology
Partial Adversarial Motion Priors for Robust Legged Loco-Manipulation
Abstract
dc:description.abstractThere has been much interest in developing legged robots capable of performing practical manipulation tasks in real-world environments, such as pushing and transporting objects. However, learning robust loco-manipulation policies remains a significant challenge, as maintaining stable locomotion while simultaneously performing precise manipulation behaviors requires complex coordination that is difficult to achieve through traditional reward-based learning alone. This work aims to address some of these challenges by proposing a partial imitation learning framework that transfers stable locomotion behaviors learned from a simpler task to the more complex setting of legged loco-manipulation.
Degree
thesis:*- Name thesis:degree_name
- Computer Science, MS
- Department thesis:degree_department
- Interactive Computing
- Grantor
- Georgia Institute of Technology
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Das, Mili
- Advisor dc:contributor.advisor
-
- Ha, Sehoon
- Committee members dc:contributor.committeemember
-
- Chernova, Sonia
- Garg, Animesh
- Kira, Zsolt
Rights
- Language dc:language.iso
- English
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/81667