Open Source Robotics

A curated collection of high-quality open source robotics projects, tools, and software to propel the robotics community forward.

Protective Stop

  • Also called: safeguard stop
  • Category: Control › Human in the loop
  • Tags: safety, industrial-robots, collaborative-robots, risk-assessment, robot-controllers

A stop triggered automatically by a robot's safety system or safeguarding devices when a monitored limit or condition is violated, as opposed to an emergency stop pressed by a person.

Definition

A protective stop is a stop of robot motion that the safety system initiates in response to a safeguarding event. The trigger may be a light curtain or area scanner detecting a person in a hazard zone, a door interlock opening, or a safety-rated monitoring function finding that a joint torque, contact force, speed, position or tool orientation has gone beyond its configured limit. Industrial robot safety standards such as ISO 10218 require this kind of stop as one of the basic safety functions, alongside the emergency stop.

The distinction from an emergency stop matters. An emergency stop is a manual action, a person hitting a red mushroom button, intended for situations nobody anticipated. A protective stop is automatic and part of the designed safeguarding of the cell. Machinery safety standards also define stop categories, ranging from immediately cutting power to the drives, through a controlled deceleration followed by power removal, to a controlled stop with power kept on and standstill monitored. Which category a protective stop uses is a design choice made in the risk assessment; stopping along the planned path with power applied avoids the arm sagging or drifting, while removing power gives a stronger guarantee that it cannot move.

After a protective stop the robot usually needs an explicit reset or acknowledgement before it resumes, and the controller may require the program to be restarted or the robot to be moved clear. On collaborative arms, protective stops triggered by force or collision limits are a normal part of operation rather than a fault. A protective stop differs from a safety filter on this site: a safety filter modifies commands continuously to stay inside a safe set, while a protective stop is a discrete reaction of the robot's safety system once a limit has already been crossed.

Why it matters

Protective stops are how a robot cell turns sensor events and limit violations into a guaranteed halt. Engineers need to know what triggers them and how recovery works, because frequent unplanned stops are a common source of downtime, and tuning limits to avoid them must never weaken the protection they provide.

Key applications

  • Stopping a robot when a light curtain, scanner or door interlock is triggered
  • Collision and force-limit reactions on collaborative arms
  • Speed and separation monitoring when people approach a cell
  • Enforcing joint, workspace and tool-orientation limits
  • Safe recovery logic in research setups that push robots near their limits

In practice

A research team uses an industrial arm to play table tennis against people. Fast swings sometimes push a joint near its configured torque or speed limit, and the safety controller then issues a protective stop. The team's software watches the controller state; when a stop occurs mid-rally, it logs the event, marks the point as void so it is replayed, waits for the stop to be acknowledged and moves the arm slowly back to a ready pose. Over time they reduce the number of stops by smoothing commanded accelerations near the limits rather than by raising the safety thresholds.

Common misconceptions

  • A protective stop is the same as an emergency stop — an emergency stop is manually initiated for unforeseen hazards, while a protective stop is triggered automatically by safeguarding
  • A protective stop means the robot is broken — on collaborative arms it is an expected response to a limit being exceeded and is cleared after the cause is checked
  • Every protective stop cuts motor power — standards allow different stop categories, and some keep power on while monitoring standstill
  • Raising limits is the right fix for frequent stops — the cause is usually the motion, payload settings or cell layout, which should be fixed without weakening protection

Who should know this

  • Robot cell integrators and safety engineers
  • Engineers programming collaborative and industrial arms
  • Researchers running fast or contact-rich experiments on real robots

Related technologies

  • Emergency Stop
  • Safety PLC
  • Light Curtains and Safety Laser Scanners
  • Speed and Separation Monitoring
  • Safety-Rated Monitored Stop

Related terms

  • Cobot
  • Safety Filter
  • Pinch Point
  • Physical Human-Robot Interaction
  • Sensorless Force Estimation
  • Force/Torque Sensor

Related projects

  • NASA Ogma: NASA runtime monitor generator for safe robotics and flight systems

This website requires JavaScript to function properly. Please enable JavaScript to explore our collection of open source robotics projects.