Collaborative Robot Safety Testing: Standards, Methods, and Compliance Requirements

Collaborative robot safety testing verifies that cobots operating alongside humans comply with international standards such as ISO 10218-1, ISO 10218-2, and ISO/TS 15066, ensuring that power, force, speed, and safety control functions remain within permissible human contact limits. This technical guide explains the applicable standards, testing samples, testing items, evaluation methods, and common questions involved in collaborative robot safety assessment and certification.

Technical Overview

Collaborative robots, commonly known as cobots, are industrial robots designed to operate in shared workspaces where humans and robots perform tasks in close proximity without traditional physical guarding. Because operators may intentionally or accidentally come into contact with the robot, collaborative robot safety testing is essential to verify that the robot's design, control system, and operating characteristics do not exceed biomechanical thresholds that could cause injury to the human body.

The safety assessment of collaborative robots is primarily governed by ISO 10218-1 and ISO 10218-2, which define the general safety requirements for industrial robots and robot systems, and by ISO/TS 15066, which specifies the technical requirements for collaborative robot operation. Supporting standards include ISO 12100 for risk assessment, ISO 13849-1 and IEC 62061 for safety-related control systems, IEC 60204-1 for electrical equipment of machines, and regional standards such as UL 1740, ANSI/RIA R15.06, and the GB 11291 series adopted in China.

ISO/TS 15066 defines four collaborative operation modes, each of which requires distinct verification approaches:

• Safety-rated monitored stop: the robot stops safely before a human enters the collaborative workspace.
• Hand guiding: the robot moves only under direct manual guidance through a safety-rated device.
• Speed and separation monitoring: robot speed is continuously adjusted to maintain a protective separation distance from operators.
• Power and force limiting: contact forces and pressures are inherently restricted to below pain onset thresholds for specific body regions.

Accredited third-party laboratories perform these evaluations to demonstrate regulatory compliance, support market access in global regions, reduce product liability risk, and provide objective evidence that the collaborative robot can be deployed safely alongside personnel.

Testing Samples

Collaborative Robot Arm Assembly

The complete articulated robot arm, including joints, actuators, encoders, linkages, and mechanical structure, submitted in a representative production configuration. Different payload classes and reach variants of the same robot family may require separate evaluation.

Robot Controller and Drive Units

The safety-rated controller, servo drives, power supply, and safety input/output interfaces. The controller firmware version is documented because safety parameters, stop performance, and monitored functions depend on software configuration.

Teach Pendants and Operator Interfaces

Hand-held programming devices, enabling switches, and operator panels are evaluated for emergency stop functionality, protective stop circuits, ergonomic design, and enabling device behavior in the three-position control logic.

End-Effectors and Grippers

Grippers, vacuum tools, and process tooling mounted on the robot flange are assessed because they frequently represent the most hazardous contact zones during collaborative operation, including pinch points and sharp edges.

Integrated Collaborative Workcell

For robot system integrations, the complete cell, including sensors, fencing, scanners, interlocks, and workstation layout, is evaluated according to ISO 10218-2 to confirm the collaborative workspace boundaries and safeguarding effectiveness.

Testing Items

A comprehensive collaborative robot safety evaluation typically covers the following test categories:

• Collaborative operation mode verification: functional confirmation of safety-rated monitored stop, hand guiding, speed and separation monitoring, and power and force limiting behavior under fault and non-fault conditions.
• Quasi-static contact testing: measurement of clamping and crushing forces and pressures sustained against body regions, compared against the permissible force and pressure limits defined in the ISO/TS 15066 body model.
• Transient contact testing: measurement of dynamic impact forces during free collision between a moving robot part and the human body model, with comparison against maximum permissible transient values.
• Speed and separation monitoring validation: verification of protective separation distance calculations, sensor response times, and robot deceleration behavior using the ISO 13855 positioning approach.
• Stop performance measurement: determination of stopping time, stopping distance, and stop angle for protective stops and emergency stops across the full speed and load range.
• Safety-rated functions verification: evaluation of safety-rated soft axis and space limiting, safety-rated speed monitoring, and monitored stop functions for achieved Performance Level and Category, typically PL d with Category 3 architecture per ISO 13849-1.
• Electrical safety testing: dielectric strength, protective bonding continuity, insulation resistance, and residual voltage tests in accordance with IEC 60204-1.
• EMC testing: conducted and radiated emissions and immunity testing per CISPR 11 and the IEC 61000 series to confirm electromagnetic compatibility in industrial environments.
• Mechanical hazard evaluation: assessment of pinch points, sharp edges, crushing zones, instability, and moving part hazards according to ISO 12100 risk assessment principles.
• Environmental and endurance testing: verification of performance under specified temperature, humidity, vibration, and duty cycle conditions relevant to the intended operating environment.
• Marking, labeling, and documentation review: confirmation of compliant instructions, warning labels, and declaration of the ISO 10218-1 collaborative operation clauses fulfilled.

Testing Methods

Biomechanical Threshold Evaluation

Force and pressure limiting verification is performed using calibrated force-pressure measuring devices with biofidelic properties that simulate human skin, soft tissue, and bone stiffness. The measuring device is placed against a rigid surface to represent worst-case clamping, and the robot is commanded at maximum permitted speed and payload. Measured peak forces and pressures are compared with the maximum permissible quasi-static and transient limits for each contactable body region defined in ISO/TS 15066, such as the skull, face, neck, chest, abdomen, arms, hands, thighs, legs, and feet.

Stop Performance and Separation Distance Measurement

Protective stop performance is measured with precision timing and position instrumentation to record the response time of the safety function, the total stopping time, and the path distance traveled during deceleration. The results feed into the minimum protective separation distance calculation for speed and separation monitoring, applying the ISO 13855 formula that combines human approach speed, robot response latency, sensor performance, and stopping distance, including measurement uncertainty.

Safety Control System Verification

The safety-related parts of the control system are analyzed and tested per ISO 13849-1, ISO 13849-2, and IEC 62061. Fault injection and failure mode testing confirm that the system achieves the specified Performance Level or Safety Integrity Level. Random hardware failure rates, diagnostic coverage, common cause failure defenses, and software safety lifecycle documentation are reviewed as part of the assessment.

Risk Assessment and Validation of the Collaborative Workspace

A documented risk assessment per ISO 12100 supports the selection of collaborative operation modes and safeguarding measures. Verification confirms that the declared collaborative workspace is correctly configured, that safety configuration parameters cannot be altered without authorization, and that warning systems and mode indicators function as designed. Integration-level validation per ISO 10218-2 confirms the interaction of the robot, operator, and process task within the production environment.

Electrical and EMC Test Procedures

Electrical tests follow the insulation, grounding, and residual voltage requirements of IEC 60204-1, while electromagnetic compatibility is evaluated in shielded chamber facilities using the methods of CISPR 11, CISPR 16, and the IEC 61000-4 series for immunity against electrostatic discharge, radiated fields, fast transients, surges, and power supply disturbances.

Applications

Collaborative robot safety testing supports deployment and market acceptance across a broad range of industries where humans and robots share tasks:

• Automotive assembly lines, including screw driving, door and panel installation, and part presentation alongside line operators.
• Electronics and semiconductor manufacturing, such as precise component insertion, connector fitting, and circuit board handling.
• Machine tending operations, where cobots load and unload CNC machines, injection molding equipment, and presses under speed and separation monitoring.
• Packaging, palletizing, and kiosk-based order fulfillment with direct human-robot task sharing.
• Laboratory automation, pharmaceutical compounding, and sample preparation requiring hygienic, low-force collaborative operation.
• Quality inspection and testing stations where robots present parts to human inspectors within a shared workspace.
• Educational institutions and research facilities where students and researchers work in proximity with robot arms.
• Small and medium-sized enterprises adopting flexible automation without traditional hard guarding.

Successful safety assessment enables manufacturers to obtain certification marks, satisfy CE conformity obligations under the Machinery Directive framework, support North American market entry under ANSI/RIA R15.06 and UL 1740, and meet regional acceptance where GB standards apply.

FAQ

Which standards are most important for collaborative robot safety testing?

The core standards are ISO 10218-1 for the robot itself, ISO 10218-2 for the integrated robot system, and ISO/TS 15066 for collaborative operation requirements. These are supplemented by ISO 13849-1 and IEC 62061 for safety control functions, IEC 60204-1 for electrical safety, and ISO 13855 for protective separation distance.

What is the difference between quasi-static and transient contact testing?

Quasi-static contact refers to near-stationary clamping or crushing contact, such as when a body part is trapped between the robot and a rigid surface, and it has stricter force and pressure limits. Transient contact refers to a dynamic impact in which the robot strikes the body without clamping, and it is evaluated against separate, typically higher permissible limits.

Is safety testing mandatory for collaborative robots?

Before placing a robot on the market in most jurisdictions, manufacturers must conduct a risk assessment and demonstrate conformity with applicable machinery safety regulations, such as the Machinery Directive framework in Europe. Testing by accredited third-party laboratories is the recognized way to provide objective evidence of conformity and is frequently required by customers, insurers, and end users.

What Performance Level must collaborative robot safety functions achieve?

Under ISO 10218-1, safety-related control functions for collaborative operation are generally required to achieve Performance Level d with Category 3 architecture as described in ISO 13849-1, meaning a single fault must not lead to loss of the safety function and must be detected at or before the next demand whenever reasonably practicable.

How long does a typical collaborative robot safety evaluation take?

The duration depends on the scope of testing. A focused power and force limiting verification may be completed within a few weeks, while a full program covering mechanical, electrical, EMC, functional safety, and all four collaborative operation modes typically requires several weeks to a few months, including documentation review and any corrective engineering iterations.

Can a robot that passes testing be used without any risk assessment at the deployment site?

No. Robot certification covers the robot as supplied, while the integrator and end user must perform an application-specific risk assessment per ISO 10218-2 and ISO 12100, considering the actual task, workpiece, end-effector, workspace layout, and operator exposure before commissioning.