Thesis
Unifying planning and learning for contact-rich manipulation
- Abstract:
-
Contact-rich manipulation tasks in semi-structured and unstructured environments pose significant challenges for robotic systems, such as in small-batch manufacturing and open-world scenarios where robots are required to generalise or rapidly adapt to novel objects, tasks, or scenes with minimal setup or reconfiguration. In such settings, manually engineered solutions are inherently unscalable, as they necessitate substantial task-specific design effort and are unable to accommodate the diver...
Expand abstract
Actions
Access Document
- Files:
-
-
(Preview, Dissemination version, pdf, 23.9MB, Terms of use)
-
Authors
Contributors
+ Posner, I
- Institution:
- University of Oxford
- Division:
- MPLS
- Department:
- Engineering Science
- Role:
- Supervisor
- ORCID:
- 0000-0001-6270-700X
+ Havoutis, I
- Institution:
- University of Oxford
- Division:
- MPLS
- Department:
- Engineering Science
- Role:
- Examiner
- ORCID:
- 0000-0002-4371-4623
+ Ramamoorthy, S
- Role:
- Examiner
+ University of Oxford (Department of Engineering Science)
More from this funder
- Funding agency for:
- Yamada, J
- Programme:
- Research Studentship: Learning Real-World Collaborative Assembly Tasks Through Observation, Action and Interaction
- DOI:
- Type of award:
- DPhil
- Level of award:
- Doctoral
- Awarding institution:
- University of Oxford
- Language:
-
English
- Keywords:
- Subjects:
- Deposit date:
-
2026-04-15
- ARK identifier:
Terms of use
- Copyright holder:
- Jun Yamada
- Copyright date:
- 2025
- Notes:
- Leveraging scene embeddings for gradient-based motion planning in latent space, COMBO-Grasp: learning constraint-based manipulation for bimanual occluded grasping, and Efficient skill acquisition for insertion tasks in obstructed environments are derived from this thesis.
If you are the owner of this record, you can report an update to it here: Report update to this record