MavLink and Apache Hudi Integration

Input and output integration overview

This plugin collects metrics from MavLink-compatible flight controllers like ArduPilot and PX4, enabling live data ingestion from unmanned systems such as drones and boats.

Writes metrics to Parquet files via Telegraf’s Parquet output plugin, preparing them for ingestion into Apache Hudi’s lakehouse architecture.

Integration details

MavLink

The MavLink plugin is designed to gather metrics from MavLink-compatible flight controllers such as ArduPilot and PX4. It provides a mechanism to live ingest flight metrics from various unmanned systems, including drones, planes, and boats. By utilizing the ArduPilot-specific MavLink dialect, the plugin parses a wide range of messages as documented in the MavLink documentation. It enables seamless integration of telemetry data, allowing for detailed monitoring and analysis of flight operations. Users must be cautious, as this plugin may generate a substantial volume of data; thus, filters are available to limit the metrics collected and transmitted to output plugins. Additionally, configuration options allow customization of which messages to receive and how to connect to the flight controller.

Apache Hudi

This configuration leverages Telegraf’s Parquet plugin to serialize metrics into columnar Parquet files suitable for downstream ingestion by Apache Hudi. The plugin writes metrics grouped by metric name into files in a specified directory, buffering writes for efficiency and optionally rotating files on timers. It considers schema compatibility—metrics with incompatible schemas are dropped—ensuring consistency. Apache Hudi can then consume these Parquet files via tools like DeltaStreamer or Spark jobs, enabling transactional ingestion, time-travel queries, and upserts on your time series data.

Configuration

MavLink

[[inputs.mavlink]]
  ## Flight controller URL supporting serial port, UDP and TCP connections.
  ## Options are documented at
  ##   https://mavsdk.mavlink.io/v1.4/en/cpp/guide/connections.html.
  ##
  ## Examples:
  ## - Serial port: serial:///dev/ttyACM0:57600
  ## - TCP client:  tcp://192.168.1.12:5760
  ## - UDP client:  udp://192.168.1.12:14550
  ## - TCP server:  tcpserver://:5760
  ## - UDP server:  udpserver://:14550
  # url = "tcp://127.0.0.1:5760"

  ## Filter to specific messages. Only the messages in this list will be parsed.
  ## If blank or unset, all messages will be accepted. Glob syntax is accepted.
  ## Each message in this list should be lowercase camel_case, with "message_"
  ## prefix removed, eg: "global_position_int", "attitude"
  # filter = []

  ## Mavlink system ID for Telegraf. Only used if the mavlink plugin is sending 
  ## messages, eg. when `stream_request_frequency` is 0 (see below.)
  # system_id = 254

  ## Determines whether the plugin sends requests to subscribe to data.
  ## In mavlink, stream rates must be configured before data is received.
  ## This config item sets the rate in Hz, with 0 disabling the request.
  ## 
  ## This frequency should be set to 0 if your software already controls the 
  ## rates using REQUEST_DATA_STREAM or MAV_CMD_SET_MESSAGE_INTERVAL
  ## (See https://mavlink.io/en/mavgen_python/howto_requestmessages.html)
  # stream_request_frequency = 4

Apache Hudi

[[outputs.parquet]]
  ## Directory to write parquet files in. If a file already exists the output
  ## will attempt to continue using the existing file.
  directory = "/var/lib/telegraf/hudi_metrics"

  ## File rotation interval (default is no rotation)
  # rotation_interval = "1h"

  ## Buffer size before writing (default is 1000 metrics)
  # buffer_size = 1000

  ## Optional: compression codec (snappy, gzip, etc.)
  # compression_codec = "snappy"

  ## When grouping metrics, each metric name goes to its own file
  ## If a metric’s schema doesn’t match the existing schema, it will be dropped

Input and output integration examples

MavLink

1. Real-Time Fleet Monitoring: Utilize the MavLink plugin to create a centralized dashboard for monitoring multiple drones in real-time. By ingesting metrics from various flight controllers, operators can oversee the status, health, and location of all drones, allowing for quick decision-making and enhanced situational awareness. This integration could significantly improve coordination during large-scale operations, like aerial surveys or search and rescue missions.

2. Automated Anomaly Detection: Leverage MavLink in conjunction with machine learning algorithms to detect anomalies in flight data. By continuously monitoring metrics such as altitude, speed, and battery status, the system can alert operators to deviations from normal behavior, potentially indicating technical malfunctions or safety issues. This proactive approach can enhance safety and reduce the risk of in-flight failures.

3. Data-Driven Maintenance Scheduling: Integrate the data collected through the MavLink plugin with maintenance management systems to optimize maintenance schedules based on actual flight metrics. Analyzing the collected data can highlight patterns indicating when specific components are likely to fail, thereby enabling predictive maintenance strategies that minimize downtime and repair costs.

4. Enhanced Research Analytics: For academic and commercial UAV research, MavLink can be used to gather extensive flight data for analysis. By compiling metrics over numerous flights, researchers can uncover insights related to flight patterns, environmental interactions, and the efficiency of different drone models. This can foster advancements in UAV technology and broader applications in autonomous systems.

Apache Hudi

1. Transactional Lakehouse Metrics: Buffer and write Web service metrics as Parquet files for DeltaStreamer to ingest into Hudi, enabling upserts, ACID compliance, and time-travel on historical performance data.

2. Edge Device Batch Analytics: Telegraf running on IoT gateways writes metrics to Parquet locally, where periodic Spark jobs ingest them into Hudi for long-term analytics and traceability.

3. Schema-Enforced Abnormal Metric Handling: Use Parquet plugin’s strict schema-dropping behavior to prevent malformed or unexpected metric changes. Hudi ingestion then guarantees consistent schema and data quality in downstream datasets.

4. Data Platform Integration: Store Telegraf metrics as Parquet files in an S3/ADLS landing zone. Hudi’s Spark-based ingestion pipeline then loads them into a unified, queryable lakehouse with business events and logs.

Feedback

Thank you for being part of our community! If you have any general feedback or found any bugs on these pages, we welcome and encourage your input. Please submit your feedback in the InfluxDB community Slack.