Collaborative Robot (Safety) Testing

Collaborative robot safety testing evaluates whether a cobot can share a workspace with human operators without unacceptable risk of injury. The test object is the complete collaborative application: the robot arm, its end-effector, the workpiece, and the safeguarded workspace considered together. Testing follows the framework of ISO 10218-1/-2 together with ISO/TS 15066, which defines four collaborative operating modes and their verification requirements. The modules below describe the scope and applicable standards, the risk assessment points that drive test selection, the force and pressure measurement procedures, the performance metrics and acceptance thresholds, the parameters that can be co-tested alongside safety, and typical application scenarios. Together they show how an accredited laboratory converts hazard analysis into measurable evidence, supporting conformity declaration and safe deployment.

scope & applicable standards

Testing covers the robot system as integrated, including the manipulator, controller, end-effector, and any collaborating-operator zone defined in the risk assessment. The governing documents are ISO 10218-1 for the robot itself and ISO 10218-2 for the integrated robot system and cell; ISO/TS 15066 supplies the technical detail for collaborative applications. Regional requirements may add machinery directives or low-voltage and EMC standards, since a cobot is simultaneously a machine and electrical equipment. Each of the four collaborative modes — safety-rated monitored stop, hand guiding, speed and separation monitoring, and power and force limiting — carries its own verification clause, so the test plan must state which mode applies before any measurement begins. Functional safety of the safety-rated control system is assessed against ISO 13849-1 or IEC 62061, because speed limits, stop functions, and force caps all depend on software-embedded safety functions whose reliability must be demonstrated.

Test items & risk assessment points

The test programme originates in the hazard identification and risk estimation performed under ISO 12100. Review the task catalogue first: every operator interaction, such as loading, teaching, cleaning, and fault clearing, defines a contact scenario that must be tested. For power and force limiting applications, map each potential contact event to a body region, because the permissible force and pressure depend on the affected body part. Distinguish quasi-static contact, where the operator is trapped against a fixed surface, from transient contact, where the moving robot strikes the operator and rebounds; the limits for the former are stricter. Additional test items include protective stop performance, safety-rated soft axis and space limiting, speed monitoring accuracy, and the response of the safety controller to injected faults. Verify that the risk assessment documents worst-case parameters — maximum payload, maximum speed, sharp edges on the end-effector — since testing at reduced settings invalidates the conclusions.

Test

methods & procedures — ISO/TS 15066 measurement
Force and pressure measurement is performed with a calibrated biomechanical test rig that mimics the stiffness and geometry of the relevant body region. A typical setup combines a six-axis force-torque sensor with an elastomer pad and pressure-indicating film placed over a rigid backing. Position the rig at the contact location identified in the risk assessment, orient it perpendicular to the approaching direction, and command the robot through the specific motion that produces the modeled contact. Record the peak force and, from the pressure film, the contact area and peak pressure at the most loaded point. Repeat each scenario at the maximum speed and payload declared for the application. Speed and separation monitoring is verified with a calibrated measurement system that compares actual robot TCP speed and the distance to a target with the safety controller's monitored values, checking both accuracy and response time. Stop-function tests record stopping time and distance across the full speed range.

Performance

metrics & acceptance criteria
The primary metrics are peak contact force in newtons, peak contact pressure in N/cm², and contact area. Acceptance is judged against the body-region limit tables in ISO/TS 15066, which assign force and pressure thresholds to specific body areas, with separate values for quasi-static and transient contact. For most body regions, transient limits are set at roughly twice the quasi-static values, reflecting shorter loading duration. A test point passes only when both force and pressure remain within the applicable limits; exceeding either one constitutes failure. Secondary metrics include speed-monitoring accuracy, positioning accuracy of the safety-rated soft limits, protective-stop time and distance, and the performance level of each safety function. Where a measured value lies within measurement uncertainty of the limit, treat the result as non-compliant or refine the measurement, and document the decision. All calibration traceability must be reported alongside the results.

Co-testable

parameters & certification pathway
Safety verification is usually bundled with performance and electromagnetic testing in a single campaign, since the same sample preparation and instrument setup serve multiple purposes. Co-testable items include path accuracy and repeatability per ISO 9283, payload and velocity verification, noise emission, EMC immunity and emission behavior under IEC-based machinery standards, and electrical safety of the controller. When a cooperative development route is used, preliminary measurements can be taken during design to tune speed and force limits, followed by formal type testing of the final configuration. The certification pathway proceeds from a documented risk assessment through a test plan agreed with an accredited laboratory, execution of the measurements, corrective iteration where limits are exceeded, and a final test report. That report, together with the technical file and the manufacturer's Declaration of Conformity, supports market placement under the applicable machinery regulations.

Application scenarios & industry cases

Collaborative robot safety testing applies wherever humans and robots share an unshielded workspace. Common deployment scenarios include assembly assistance, machine tending, palletizing of light parts, screwdriving, quality inspection stations, and laboratory sample handling. In each case the value of testing lies in confirming that the engineered combination of reduced speed, rounded end-effectors, and force limitation keeps contact events within tolerable biomechanical limits. Sector requirements differ: medical and pharmaceutical installations add hygiene and cleanroom considerations, while food-contact applications introduce washdown constraints that affect end-effector design and thus contact geometry. Integration matters as much as the robot itself, because a compliant arm fitted with a sharp tool can still fail verification. For this reason end-users and integrators typically commission application-level testing rather than relying solely on the arm manufacturer's certificates, so that the tested configuration matches the production configuration exactly.