Prototype: CompressedImage camera dumb relay - #11
Open
rosterloh wants to merge 1 commit into
Open
Conversation
Forward sensor_msgs/msg/CompressedImage frames verbatim (header, format, opaque data bytes) without decoding, riding the existing topic pipeline so router, UDS, QUIC, CLI echo, and the TUI work unchanged. Inspired by Media over QUIC's "dumb relay" model. Documents the moq-inspired scaling path (dedicated binary frame type, per-keyframe-group streams with stale-frame dropping, codec-free agent, non-terminal viewer) in the roadmap as explicitly out of scope for this prototype. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Prototype "dumb relay" for camera streams: the agent forwards
sensor_msgs/msg/CompressedImageframes without decoding them, inspired by Media over QUIC's "keep the server dumb" model. It rides the existing topic pipeline, so the router, UDS, QUIC per-topic streams, CLIecho, and the TUI all work with no changes.talos echo /camera/image_raw/compressedprintsformat: "jpeg"/data: [N bytes], proving end-to-end forwarding without needing a renderer.Changes (+91 lines, 2 code files)
talos-agent/src/conversions.rs—compressed_image_to_dynvalue()mapping to{ header, format, data: Bytes }, copying the opaque payload verbatim, plus a passthrough unit test.talos-agent/src/bridge.rs—sensor_msgs/msg/CompressedImagesubscription arm (mirrors the LaserScan/Imu arms, honorsqos).docs/src/configuration/agent-config.md— camera subscription example + scope note.docs/src/future/roadmap.md— moq-inspired scaling path (dedicated binary frame type, per-keyframe-group streams with stale-frame dropping, codec-free agent, non-terminal viewer).CHANGELOG.md—[Unreleased]/Addedentry.Design decisions
DynValue::Bytesinstead of a dedicatedCameraFramevariant — a new variant would change theProtocolClient::recv_datatrait signature and ripple through both transports and both clients. For compressed frames the existing reliable pipeline is fine (within the 16 MiB frame cap). The dedicated-frame + group-rotation design is documented as the scaling step, not built.Out of scope (documented in roadmap)
Raw
sensor_msgs/msg/Image, in-agent transcoding, per-keyframe-group QUIC streams with stale-frame dropping, and a non-terminal viewer.Verification
rclrs_ws/install/ROS message crates). TheCompressedImagefield types (header: Header,format: String,data: Vec<u8>) were checked against the generated binding, and the code is mechanically identical to the proven LaserScan/Imu arms — type-correct by construction. Please confirm CI / localcargo test -p talos-agentis green before merge.🤖 Generated with Claude Code