Repository navigation
Use measured robot sensors for receptor input and recover live streams - #272
Draft
superposition wants to merge 3 commits into
Draft
superposition wants to merge 3 commits into
superposition wants to merge 3 commits into
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Camera timeouts terminated the live producer, small firing frames stayed buffered, and brightness was injected directly into 95,494 visual-stage neurons without body sensors. Add bounded camera/IMU/odometry/lidar acquisition, source timestamps, explicit engineering feature conditioning and paired input/output JSONL. Drive only annotated receptors, skip neural steps on failed acquisition, clear temporal history, and emit a hold. Flush every QLSP frame, including zero-firing frames. A bounded observation helper exposes fresh frames and current input provenance to the console with wheel transport disconnected.
Freshness is checked after input evidence writes and immediately before wheel serialization. Expiry during trace/spike I/O cancels the nonzero candidate, emits zero, annotates its cancellation and fails. Graph signs and QLSP relative timestamps are preserved. The 6,063 receptors without positions receive uniform measured features instead of invented spatial assignments.
The receptor outgoing graph is inhibitory and zero-resting LIF does not establish downstream motor firing. This PR does not solve that model/calibration gap or complete driving. No synthetic spikes, graph-sign flips or tonic activity are added to manufacture output. Camera capture time and calibrated pixel/range alignment remain unavailable.
Validation: CUDA-feature build; 3 focused CPU fusion tests; zero-firing flush regression; provenance gate; bounded host GPU run on live robot camera and compact telemetry with max wheel speed zero and no transport. Console evidence under #270 records 1,151 measured bundles/outputs and zero nonzero decoded commands. Three static reviews covered the final fixes, not physical acceptance.
Depends on #267. Refs #270, #262. No issue is closed by this partial recovery.