mirror of
https://github.com/azaion/autopilot.git
synced 2026-06-21 05:41:09 +00:00
Oleksandr Bezdieniezhnykh
e077d3bd15
ci/woodpecker/push/build-arm Pipeline failed
Stand up a production-target test runner on jetson-e2e and run the deferred cargo test --workspace for batch 19. Infra: - Dockerfile.test: ubuntu:22.04 + libopencv-dev + libav*-dev + libclang-dev + protobuf-compiler + rust 1.82.0 (rustfmt, clippy). Sets LIBCLANG_PATH so clang-sys can dlopen libclang under the opencv-rust clang-runtime path. - scripts/jetson-test.sh: rsync source to jetson-e2e, docker build, docker run cargo test --workspace --no-fail-fast. Workspace fix exposed by the gate: - Cargo.toml: enable opencv "clang-runtime" feature. Without it the workspace fails to build because clang-sys is shared between opencv-binding-generator and bindgen (via ffmpeg-sys-next) and the opencv generator panics with "a `libclang` shared library is not loaded on this thread" (opencv-rust GH issue #635). Batch-19 code bugs exposed by the gate (6 compile errors + 1 algo bug): - movement_detector::optical_flow: min_max_loc signature (opencv 0.98 expects Option<&mut f64> / Option<&mut Point>); data_mut() returns *mut u8 directly, not Result. RANSAC residual now filters by the inlier mask returned by find_homography (matches the docstring; was systematically over-reporting motion magnitude on synthetic pure-pan input). - semantic_analyzer::scoring::freshness: same data_mut() fix; stddev_f32 now takes &impl core::ToInputArray so it accepts the BoxedRef<Mat> that Mat::roi returns in opencv 0.98. Result: 391 tests passed across 58 binaries, 0 in-scope failures. Two pre-existing failures in frame_ingest (batch 16-18 scope) are NOT addressed here and are recorded as leftovers: - frame_ingest_cuvid_segv: HIGH severity production bug; libavcodec58 advertises h264_cuvid but libnvcuvid.so.1 is missing at runtime, the software fallback never fires, first send_packet SEGVs. - frame_ingest_publisher_timing_flake: LOW severity; Jetson-specific timing budget too tight for ac1_three_consumers_at_rate_lose_no_frames. Neither blocks batch 20 (movement_detector / semantic_analyzer next). Co-authored-by: Cursor <cursoragent@cursor.com>
autopilot
Onboard mission executor for the AZAION reconnaissance UAV. Single Rust binary; runs on
NVIDIA Jetson Orin Nano Super (aarch64). See _docs/02_document/architecture.md for the
authoritative system design.
Layout
crates/
shared/ # canonical DTOs, config, error, health, observability, clock, contracts
autopilot/ # binary crate — runtime composition root + /health endpoint
mavlink_layer/ # hand-rolled MAVLink v2 transport
mission_client/ # missions API REST client + MapObjects sync
frame_ingest/ # RTSP pull + decode
detection_client/ # bi-directional gRPC to ../detections
movement_detector/ # ego-motion-compensated residual-motion clustering
semantic_analyzer/ # Tier 2 — primitive graph + ROI CNN
vlm_client/ # Tier 3 — optional NanoLLM/VILA local IPC
mapobjects_store/ # H3-indexed on-device map + ignored items
gimbal_controller/ # ViewPro A40 UDP control
scan_controller/ # central typed state machine (ZoomedOut/ZoomedIn/TargetFollow)
operator_bridge/ # POI surface + operator command authentication
mission_executor/ # multirotor + fixed-wing FSMs + geofence + failsafe
telemetry_stream/ # always-on uplink to Ground Station
config/ # TOML config per environment (dev / staging / prod)
deploy/systemd/ # on-airframe native systemd unit (Option A)
fixtures/ # replay clips (RTSP, MAVLink, missions, detections)
tests/e2e/ # workspace-level blackbox scenarios
benches/ # NFR benchmark-gate harness
Build
# Host-arch build + tests
cargo build --workspace
cargo test --workspace --locked
# Optional VLM feature path
cargo build --workspace --features vlm
# No-default-features path (enforces the VLM optionality contract)
cargo build --workspace --no-default-features
cargo test --workspace --no-default-features
# aarch64 cross-build (CI uses cargo-zigbuild; locally `cross` also works)
cargo install --locked cargo-zigbuild
rustup target add aarch64-unknown-linux-gnu
cargo zigbuild --release --target aarch64-unknown-linux-gnu --workspace
Run (dev)
cp .env.example .env
docker compose up -d
# Then inspect:
curl -s http://127.0.0.1:8080/health | jq
Documentation
The full document tree lives under _docs/. Start with:
_docs/00_problem/problem.md— the problem statement_docs/02_document/architecture.md— system architecture_docs/02_document/system-flows.md— sequence diagrams_docs/02_document/components/<name>/description.md— per-component specs_docs/02_document/deployment/{containerization,ci_cd_pipeline,observability}.md
CI
.woodpecker.yml drives the pipeline. Stages: fetch → lint → unit-test → build-arm64 → build-no-vlm → integration-test → sitl-conformance → security-scan → package → sign → publish → benchmark-gate (opt-in).