Mission
Goals, environment, and success conditions for a robot task.
Early research prototype
ZephaRobot
An early research prototype for organizing a ROS 2 mobile-robot run as a traceable Episode that keeps problems, timelines, and original evidence connected.
RESEARCH QUESTION
How can a robot failure become a replayable, explainable, and verifiable engineering record?
Why Episode
Robot data needs task context.
A topic segment, error log, or map rarely explains a complete failure. ZephaRobot begins with the mission and preserves the robot, environment, software, runtime signals, and key events.
Episode
Timeline, state, and full context during one mission run.
Incident
An anomaly, failure, or behavioral change worth investigating.
Evidence
Signals, logs, parameters, and time ranges supporting a conclusion.
Episode pipeline01 02 03 04
From robot data to reviewable diagnostic evidence.
Run data
rosbag2 · logs · TF · parameters · maps
Task context
robot · mission · environment · software
Episode
timeline · events · signals · annotations
Review
incidents · evidence · checks · report
Research prototype
Current engineering foundation.
The prototype currently validates offline Episodes, incident rules, and review reports with mock and simulated data. It does not yet provide a real ROS 2 connector or rosbag2 import.
Episode data model
The current prototype includes server-side Robot, Mission, Episode, Signal, Event, Incident, and Report models and APIs.
Repeatable mock scenarios
Generated run data validates event timelines, incident rules, and the reporting workflow.
Explainable incident rules
Diagnosis summaries retain the triggering rule, time range, and related signals instead of returning an untraceable conclusion.
Episode review reports
Run summaries, incidents, evidence, and follow-up checks are assembled into a consistent HTML report.