[#13] Add sensor diagnostics demo #14
Merged
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
Add lightweight sensor diagnostics demo (no Gazebo required): - CMakeLists.txt and package.xml for ROS 2 package - Simulated sensor base class with fault injection support - LiDAR simulator node with configurable parameters: - scan_rate, range_min/max, noise_stddev - Runtime fault injection via parameters - Diagnostics publisher for sensor status This is the foundation for the sensor diagnostics demo that demonstrates ros2_medkit's data monitoring and configuration management features.
Add remaining simulated sensor nodes with fault injection: - IMU simulator: acceleration, angular velocity, drift simulation - GPS simulator: NavSatFix with position noise and signal loss - Camera simulator: RGB image generation with noise/black frames All sensors support runtime parameter changes for fault injection: - failure_probability for timeouts - inject_nan for invalid values - drift_rate for gradual sensor drift - noise parameters for degraded operation
Add anomaly detector that monitors sensor streams and reports faults: - Subscribes to LiDAR, IMU, GPS topics - Detects NaN values, out-of-range readings - Monitors message rates and timeouts - Publishes DiagnosticArray for detected faults Fault detection thresholds are configurable via ROS parameters: - rate_timeout_sec - max_nan_ratio - sensor-specific rate minimums
Add ROS 2 launch file and configuration: - demo.launch.py: launches all nodes with proper namespacing - /sensors: lidar_sim, imu_sim, gps_sim, camera_sim - /processing: anomaly_detector - /diagnostics: ros2_medkit_gateway - sensor_params.yaml: default sensor parameters - medkit_params.yaml: gateway configuration - sensor_manifest.yaml: SOVD entity hierarchy definition
Add lightweight Dockerfile and docker-compose.yml: - Based on ros:jazzy-ros-base (no Gazebo/GUI packages) - Image size: ~500MB vs ~4GB for TurtleBot3 demo - Startup time: ~5s vs ~60s - Includes CI profile for headless testing - Optional sovd-web-ui service for visualization
Add interactive demo scripts: - run-demo.sh: walkthrough of ros2_medkit API features - inject-noise.sh: increase sensor noise levels - inject-failure.sh: cause sensor timeouts - inject-nan.sh: inject NaN values - inject-drift.sh: enable sensor drift - restore-normal.sh: clear all faults Add comprehensive README.md with: - Quick start guide (Docker and source) - API examples for all sensor operations - Fault scenario documentation - Parameter reference tables - Comparison with TurtleBot3 demo
- Add sensor diagnostics demo to demos table (marked as Ready) - Add Quick Start section with docker compose instructions
Build sensor diagnostics demo image in Docker build job. Runs before TurtleBot3 build (faster, validates basic ros2_medkit).
The gateway requires additional packages: - ros2_medkit_serialization (JSON/ROS message conversion) - ros2_medkit_msgs (custom messages) - dynmsg (dynamic message introspection) Also add ros-jazzy-yaml-cpp-vendor to apt dependencies.
bburda
force-pushed
the
feat/sensor-diagnostics-demo
branch
from
9c090fb to
2c05967
Compare
There was a problem hiding this comment.
Pull request overview
This PR adds a lightweight, Gazebo‑free sensor diagnostics demo for ros2_medkit, including simulated sensors, an anomaly detector, Docker/docker‑compose integration, and CI wiring. It provides REST‑driven fault injection and a walkthrough script to showcase monitoring, configuration, and fault detection.
Changes:
- Introduces simulated LiDAR, IMU, GPS, and camera nodes with configurable fault injection and diagnostic publishing.
- Adds an anomaly detector node, parameter/config/manifest files, Dockerfile, and docker‑compose setup for running the demo plus an optional web UI.
- Updates documentation, helper shell scripts, and CI to build the new demo image and advertise it as the primary quick‑start path.
Reviewed changes
Copilot reviewed 23 out of 23 changed files in this pull request and generated 8 comments.
Show a summary per file
| File | Description |
|---|---|
| demos/sensor_diagnostics/src/lidar_sim_node.cpp | LiDAR simulator node publishing LaserScan with noise, NaN, drift, and failure injection plus diagnostic status. |
| demos/sensor_diagnostics/src/imu_sim_node.cpp | IMU simulator node publishing Imu with configurable noise, NaN, drift, and failure injection plus diagnostics. |
| demos/sensor_diagnostics/src/gps_sim_node.cpp | GPS simulator node publishing NavSatFix with noise, drift, NaN, and loss‑of‑fix behavior plus diagnostics. |
| demos/sensor_diagnostics/src/camera_sim_node.cpp | Camera simulator node publishing Image and CameraInfo with noise, black frames, and brightness faults plus diagnostics. |
| demos/sensor_diagnostics/src/anomaly_detector_node.cpp | Anomaly detector node subscribing to sensor topics, detecting anomalies/timeouts/rate degradation, and publishing fault events. |
| demos/sensor_diagnostics/run-demo.sh | Interactive shell script driving the REST API to show gateway health, sensor data, configs, and current faults. |
| demos/sensor_diagnostics/restore-normal.sh | Helper script to reset all sensor fault‑related parameters back to normal values via the REST API. |
| demos/sensor_diagnostics/inject-noise.sh | Helper script to inject high‑noise faults into all sensors via configuration updates. |
| demos/sensor_diagnostics/inject-nan.sh | Helper script to enable NaN injection on LiDAR, IMU, and GPS via configuration updates. |
| demos/sensor_diagnostics/inject-failure.sh | Helper script to force LiDAR sensor timeouts by setting failure probability to 1.0. |
| demos/sensor_diagnostics/inject-drift.sh | Helper script to configure drift faults on LiDAR, IMU, and GPS sensors. |
| demos/sensor_diagnostics/restore-normal.sh | Script to clear all configured faults and restore baseline sensor settings. |
| demos/sensor_diagnostics/package.xml | Defines the sensor_diagnostics_demo ROS 2 package and its dependencies/exec deps. |
| demos/sensor_diagnostics/launch/demo.launch.py | Launch description that starts all simulated sensors, the anomaly detector, and the ros2_medkit gateway with appropriate namespaces and configs. |
| demos/sensor_diagnostics/include/sensor_diagnostics/simulated_sensor_base.hpp | Adds a (currently unused) base class and config struct for reusable fault‑injection behavior in simulated sensors. |
| demos/sensor_diagnostics/docker-compose.yml | docker‑compose stack to run the sensor demo container and optional SOVD web UI, plus a CI profile variant. |
| demos/sensor_diagnostics/config/sensor_params.yaml | Default ROS 2 parameters for each simulated sensor and the anomaly detector. |
| demos/sensor_diagnostics/config/sensor_manifest.yaml | SOVD manifest describing areas, components, and apps for the sensor diagnostics demo. |
| demos/sensor_diagnostics/config/medkit_params.yaml | ros2_medkit gateway configuration (HTTP server, CORS, discovery via manifest) for this demo. |
| demos/sensor_diagnostics/README.md | Dedicated README explaining the demo, architecture, REST API examples, fault scenarios, scripts, and parameter reference. |
| demos/sensor_diagnostics/Dockerfile | Dockerfile building a lightweight Jazzy‑based image containing ros2_medkit_gateway and the sensor diagnostics demo. |
| demos/sensor_diagnostics/CMakeLists.txt | CMake configuration for building and installing the new simulator and anomaly detector nodes plus config/launch assets. |
| README.md (root) | Top‑level README updated to add the Sensor Diagnostics demo and quick‑start instructions. |
| .github/workflows/ci.yml | CI workflow updated to build the sensor diagnostics Docker image alongside the TurtleBot3 demo image. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
demos/sensor_diagnostics/include/sensor_diagnostics/simulated_sensor_base.hpp
Outdated
Show resolved
Hide resolved
- Add --recurse-submodules to clone dynmsg submodule - Add ros-jazzy-example-interfaces to apt dependencies - Skip test dependencies not in ros-base image - Disable BUILD_TESTING for faster builds - Use ghcr.io/selfpatch/sovd_web_ui:latest image
- Add ros2_medkit_fault_manager to Docker image
- Add fault_manager node to launch file
- Integrate anomaly_detector with ReportFault service
- Fix topic subscriptions (/sensors/scan instead of /sensors/lidar_sim/scan)
- Fix run-demo.sh jq queries for {items: []} response format
- Add libsqlite3-dev for fault_manager storage
Fault injection now properly reports to FaultManager:
- inject-nan.sh -> SENSOR_NAN faults
- inject-failure.sh -> SENSOR_TIMEOUT faults
- inject-noise.sh -> RATE_DEGRADED faults
Demonstrates two fault reporting mechanisms in ros2_medkit: LEGACY PATH (diagnostics → bridge): - LiDAR and Camera publish DiagnosticArray to /diagnostics topic - ros2_medkit_diagnostic_bridge converts and reports to fault_manager - Auto-generates fault codes from diagnostic names (e.g., LIDAR_SIM) MODERN PATH (direct service call): - IMU and GPS monitored by anomaly_detector - Anomaly detector calls /fault_manager/report_fault directly - Uses specific fault codes (SENSOR_NAN, SENSOR_TIMEOUT, etc.) Changes: - Add diagnostic_bridge and fault_reporter to Dockerfile - Update lidar_sim and camera_sim to publish DiagnosticArray - Add diagnostic_bridge node to launch file - Remove LiDAR monitoring from anomaly_detector - Update all inject scripts with path documentation - Add comprehensive dual-path documentation to README
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 23 out of 23 changed files in this pull request and generated 6 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- Add missing <algorithm> include for std::clamp (lidar, camera) - Validate rate parameters to prevent division by zero (all sensors) - Remove unused max_nan_ratio parameter from anomaly_detector - Remove unused lidar_rate_min from config (LiDAR uses legacy path now) - Remove empty clear_passed_faults function and clear_timer_ - Fix DRIFT_DETECTED → DRIFTING in README fault table - Add documentation to simulated_sensor_base.hpp explaining future use
- Add response callback to async_send_request() in anomaly_detector
to ensure service requests are properly processed by the executor
- Fix restore-normal.sh to use correct entity paths for fault clearing:
- Legacy path faults: /apps/diagnostic_bridge/faults/{fault_code}
- Modern path faults: /apps/anomaly_detector/faults/{fault_code}
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 23 out of 23 changed files in this pull request and generated 7 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
mfaferek93
reviewed
Jan 26, 2026
demos/sensor_diagnostics/include/sensor_diagnostics/simulated_sensor_base.hpp
Outdated
Show resolved
Hide resolved
Fixes from @mfaferek93: - Update copyright year 2025 → 2026 in all source files - Add early return in report_fault() to avoid spamming FaultManager - Remove unused normal_dist_ member from camera_sim_node - Unify diagnostic topic to /diagnostics for all sensors (IMU, GPS, LiDAR, Camera) Fixes from @Copilot: - Add rcl_interfaces to ament_target_dependencies for lidar/camera nodes - Fix docker-compose jq filter: .[] → .items[] - Add exec_depend for ros2_medkit_diagnostic_bridge and fault_manager - Fix anomaly_detector topic subscriptions to match sensor namespaces - Add runtime scan_rate validation in lidar parameter callback - Add num_readings validation with fallback to 360 Additional fixes: - Update restore-normal.sh to clear IMU_SIM and GPS_SIM faults
Per @mfaferek93's review: removed speculative base class that was never used. Can be re-added when/if sensors actually need shared fault injection logic.
- Fix anomaly_detector topic subscriptions to match actual sensor topics (/sensors/imu instead of /sensors/imu_sim/imu) - Change NaN injection rate from 5% to 100% for clear fault demonstration - Simplify diagnostic levels: all non-OK statuses now ERROR for reliable fault reporting through diagnostic_bridge - Reset drift start_time when drift_rate parameter changes for accurate drift accumulation - Remove unused simulated_sensor_base.hpp (YAGNI)
mfaferek93
approved these changes
Jan 26, 2026
bburda
added a commit
that referenced
this pull request
Jan 26, 2026
Fixes from @mfaferek93: - Update copyright year 2025 → 2026 in all source files - Add early return in report_fault() to avoid spamming FaultManager - Remove unused normal_dist_ member from camera_sim_node - Unify diagnostic topic to /diagnostics for all sensors (IMU, GPS, LiDAR, Camera) Fixes from @Copilot: - Add rcl_interfaces to ament_target_dependencies for lidar/camera nodes - Fix docker-compose jq filter: .[] → .items[] - Add exec_depend for ros2_medkit_diagnostic_bridge and fault_manager - Fix anomaly_detector topic subscriptions to match sensor namespaces - Add runtime scan_rate validation in lidar parameter callback - Add num_readings validation with fallback to 360 Additional fixes: - Update restore-normal.sh to clear IMU_SIM and GPS_SIM faults
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.
Description
This PR adds a lightweight, Gazebo‑free sensor diagnostics demo for ros2_medkit, including simulated sensors, an anomaly detector, Docker/docker‑compose integration, and CI wiring. It provides REST‑driven fault injection and a walkthrough script to showcase monitoring, configuration, and fault detection.
Changes:
Related Issue
closes #13
Checklist