What it means
Robotics combines sensing, AI decision logic, and physical actuation to operate in an environment.
Why it matters
Software errors become physical hazards when perception, planning, timing, communications, and emergency controls are not engineered together.
Practical focus
- Define the approved purpose, accountable owner, users, operating environment, and acceptable risk for robotics and autonomous systems before implementation or procurement
- Implement hazard analysis, bounded operating domains, redundant sensing, fail-safe states, manual override, and maintenance controls through documented architecture, policy, configuration, and change control
- Collect safety cases, near-miss logs, sensor-failure tests, stop-time measurements, and override exercises and compare the results with approved requirements, risk thresholds, and expected outcomes
- Exercise this failure scenario as a planned test: a sensor or network failure leaves the machine acting on stale information instead of entering a safe state
Common mistakes
- Using the label robotics and autonomous systems without defining the exact system boundary, decision, data, and people in scope
- Treating a model response or vendor claim as sufficient evidence without testing the surrounding application and operating process
- Failing to preserve safety cases, near-miss logs, sensor-failure tests, stop-time measurements, and override exercises in a form that lets reviewers reconstruct decisions and changes
- Testing only the normal path and never rehearsing what happens when a sensor or network failure leaves the machine acting on stale information instead of entering a safe state
Certification relevance
This subject appears in or supports the following certification bodies of knowledge: