Research program 02
Embodied Intelligence
Learning systems that connect perception and reasoning to physical action, beginning with agricultural machines operating in unstructured outdoor environments.
Research programInternal
Research objective
Study how physical agents can acquire useful behaviors from demonstrations, interventions, failures and repeated operation while remaining subject to deterministic safety controls.
01 / Core scientific questions
Questions under investigation.
- How can teleoperation data support gradual autonomy?
- How should a system learn from human takeover and recovery?
- Which representations transfer across machines and operating conditions?
02 / Current projects
Research topics.
- Vision-language-action models
- Teleoperation-to-autonomy
- Learning from human intervention
- Outdoor navigation
- Cross-machine adaptation
- Failure recovery
- Multi-robot learning
03 / Methods
Intended methods.
- Imitation learning
- Intervention-aware data collection
- Offline policy evaluation
- Hardware-in-the-loop testing
04 / Evaluation
How the program will be evaluated.
- Mission completion
- Human interventions
- Recovery from failure
- Behavior under safety constraints
Evaluation protocols and results will be published only after review. No performance result is claimed on this page.
05 / Research infrastructure
Systems and related models.
06 / Verification pending
Researchers and related publications.
Researchers involved
Verified researcher profiles have not yet been published.
Related publications
No public publication is currently associated with this program.