Human–robot collaboration is currently one of the frontiers of industrial robot implementation. In parallel, the use of robots and robotic devices is increasing in several fields, substituting humans in “4D”—dull, dirty, dangerous, and delicate—tasks, and such a trend is boosted by the recent need for social distancing. New challenges in safety assessment and verification arise, due to both the closer and closer human–robot interaction, common for the different application domains, and the broadening of user audience, which is now very diverse. The present paper discusses a cross-domain approach towards the definition of step-by-step validation procedures for collaborative robotic applications. To outline the context, the standardization framework is analyzed, especially from the perspective of safety testing and assessment. Afterwards, some testing procedures based on safety skills, developed within the framework of the European project COVR, are discussed and exemplary presented.

The use of collaborative robots in the industrial domain has significantly grown in the last years, allowing humans to operate in the same workspace occupied by robots without any physical barriers. Understandably, the safety of the human operator has been a major concern both for researchers and regulatory bodies. The power and force limited modality of robots is of particular interest in that sense, being used in order to bound the energy of eventual collisions when a close physical interaction with humans is necessary. Such an interaction modality allows the robotic system to operate without the use of barriers, but a measurement of the force and pressure occurring due to a contact must be provided as part of the risk assessment. However, the precise procedure to follow in order to reliably provide such measures is still unclear for users and system integrators willing to self-assess the safety of their own collaborative robotic system. In this work, the repeatability and reliability of such testing procedures and measures are analyzed with an interlaboratory comparison approach, with the aim to establish the degree of variability possibly encountered when performing the same test under slightly different conditions.

Biomechanical limits based on pain thresholds ensure safety in workplaces where humans and cobots (collaborative robots) work together. Standardization bodies' decision to rely on pain thresholds stems from the assumption that such limits inherently protect humans from injury. This assumption has never been verified, though. This article reports on a study with 22 human subjects in which we studied injury onset in four locations of the hand-arm system using an impact pendulum. During the tests, the impact intensity was slowly increased over several weeks until a blunt injury, i.e., bruising or swelling, appeared in the body locations under load. A statistical model, which calculates injury limits for a given percentile, was developed based on the data. A comparison of our injury limits for the 25th percentile with existing pain limits confirms that pain limits provide suitable protection against impact injuries, albeit not for all body locations.

Collaborative robots (cobots) become more and more important in industrial manufacturing as flexible companions, working side by side with humans without safety fences. A key challenge of such workplaces is to guarantee the safety of the human co-workers. The safeguarding Power and Force Limiting, as specified by ISO 10218-2 and ISO/TS 15066, has the objective to protect humans against robot collisions by preventing the robot from exceeding biomechanical limits. Unintended contact such as collisions can occur under unconstrained spatial conditions (a human body part can move freely) or constrained spatial conditions (a human body part is pinched). In particular, collisions under constrained conditions involve a high risk of injury and thus require the robot to stop immediately after detecting the collision. The robot’s speed has a significant influence on its stopping behavior, though, and thus on the maximum collision forces that the robot can exert on the human body. Consequently, a safe velocity is required that avoids the robot from exerting forces and pressures beyond the biomechanical limits. Today, such velocities can only be ascertained in costly robot experiments. In this article, we describe a model that enables us to determine the contact forces of a cobot as they occur in constrained collisions. Through simulations, it becomes possible to iteratively determine the maximum safe velocity for a specific contact hazard that occurs under constrained spatial conditions. Experimental tests with different cobots confirm the results of our model, albeit not for all robots. Despite the mixed test results, we strongly believe that our model can significantly improve the reliability of assumptions made today during the planning of cobots.

Collaborative robots (cobots) can be employed in close proximity to human workers without safety fences. The operation mode Power and Force Limiting requires that cobots not exceed the biomechanical limits of ISO/TS 15066 to ensure protection against injuries caused by collisions with them. Collision tests must be performed to prove that cobots cannot exceed the biomechanical limits. Such tests are performed with a biofidelic measuring device that measures contact forces and replicates the biomechanics of the human body. Biomechanical response curves serve as a reference for the calibration of such devices. In order to be able to compare measurements and limits correctly and reliably, the limits and response curves for calibration must be obtained from the same data with the same methodology. In this article, we present a new technique for developing biomechanical response curves, which employs a statistical model we used to calculate biomechanical limits for cobots in a previous study. This technique's development process entails normalizing the data over force, resampling them and then fitting the newly obtained samples to a log-normal distribution. The statistical model makes it possible to produce response curves for the same quantile we used for the limits. Our technique adds a confidence region around each response curve to express the sufficiency of the available data. We have produced response curves for 24 different body locations for which we have calculated limits. These curves will enable manufacturers of cobot testing equipment to calibrate their measuring devices precisely.

Power and force limiting (PFL) of ISO/TS 15066 sets the requirements to operate collaborative robots safely in close proximity to humans. Robots in PFL mode may not exceed biomechanical limits during accidental contact with humans to protect them from serious injury. A growing number of robot users are coming to realize that the framework behind PFL is too strict, ultimately constraining robot efficiency in collaborative applications. For instance, the framework does not provide any options for factoring the consequences of contact hazards and their probability of occurrence into the risk assessment. This paper proposes a revision of the PFL framework. Based on the findings of a literature survey, we introduce the new severity S0, which eliminates any form of injury in collaborative robotics. Instead of taking the onset of pain as the sole tolerable stress level, S0 defines three sublevels of tolerable stress that additionally correspond to different levels of risk. Each sublevel of stress is related to separate limits for two fields of application and two differently composed groups of people. The framework includes a structure for organizing limits, which is based on research findings and incorporates the most relevant load factors of accidental human-robot contacts, which govern their intensity and effects on the human body.