FBW-P02
Odor-Guided Source Search
Explore how local odor and directional cues can guide a virtual fly toward a source under changing and noisy conditions.
Training preparationApplications open
Statistics temporarily unavailable
Project background
Begin with a static concentration field, then study competing attractant and repellent sources and intermittent odor. The controller receives local cues rather than the source coordinates. The objective is to characterize the conditions in which a neural controller can search reliably.
This project connects a defined neural-computation question to a reproducible evaluation protocol and a community participation pathway. Results will be interpreted only within their recorded model, task and evidence scope.
Training goals
- Improve source-finding across starting positions.
- Reduce search duration and selection of the wrong source.
- Measure adaptation to concentration changes, wind cues and sensory noise.
Current foundation and milestones
Connectome and stimulation-response computation are reusable. Olfactory mapping, concentration fields, action readouts and a source-search benchmark need project-specific evidence.
Next milestone: Define sensory encoding and a static source-search benchmark
- PlannedDefine sensory encoding and a static source-search benchmark
- PlannedImplement the versioned task and measurement interface
- PlannedEvaluate frozen conditions and independent reproduction
Content updated:
Technical approach
Simulate concentration at left and right sensory positions, with wind and time variation added in later protocol versions. Encode local signals into declared pathways and read neural responses into forward motion, turning and search actions. Community evaluation can cover concentrations, starts and interference conditions. Ground-truth target coordinates remain restricted to the evaluator.
Method schematic
Task and evaluation design
Evaluation design
| Indicator | Definition | Direction | Units | Observation |
|---|---|---|---|---|
| Source arrival | Fraction of episodes reaching the protocol source region. | Higher is better | % | Awaiting evaluation |
| Search duration | Model time to reach the source, with failures reported separately. | Lower is better | model time | Awaiting evaluation |
| Wrong-source selection | Fraction of episodes choosing a competing source. | Lower is better | % | Awaiting evaluation |
| Signal recovery | Recovery time and failure conditions after an odor interruption. | Lower is better | model time | Awaiting evaluation |
No measured values have been published. Acceptance thresholds will be frozen in each task protocol before evaluation.
Expected deliverables
- Versioned source-search configurations.
- Trajectories with complete test-condition manifests.
- An interference and robustness evaluation report.
Ways to participate
- Environment parameter sweeps
- Robustness evaluation
- Input mapping research
- Independent reproduction
Device requirements and task schedules will be matched by the community after runnable profiles have been evaluated. Unknown specifications are welcome in the application.
Apply to join training
Applications are received for community review and matching. Selection does not mean training has started.
References and related work
External sources provide technical background. They do not establish equivalent FBW capabilities.
View published research records ↗