[Paper Review] Effect of Active and Passive Protective Soft Skins on Collision Forces in Human-robot Collaboration

Petr Švarný, Jakub Rozlivek+5 more|arXiv (Cornell University)|Mar 18, 2022

Robot Manipulation and Learning23 references22 citations

TL;DR

This study investigates how active and passive protective soft skins reduce collision forces in human-robot collaboration, using 2,250 empirical collision measurements on UR10e, KUKA LBR iiwa, and KUKA Cybertech robots. It demonstrates that protective skins can safely enable up to four times higher velocities than prescribed by ISO/TS 15066—up to 0.5 m/s—when combined with proper safety stop settings, and proposes an extended model incorporating skin stiffness and compressible thickness for improved force prediction.

ABSTRACT

Soft electronic skins are one of the means to turn an industrial manipulator into a collaborative robot. For manipulators that are already fit for physical human-robot collaboration, soft skins can make them safer. In this work, we study the after impact behavior of two collaborative manipulators (UR10e and KUKA LBR iiwa) and one classical industrial manipulator (KUKA Cybertech), in presence or absence of an industrial protective skin (AIRSKIN). In addition, we isolate the effects of the passive padding and the active contribution of the sensor to robot reaction. We present a total of 2250 collision measurements and study the impact force, contact duration, clamping force, and impulse. The dataset is publicly available. We summarize our results as follows. For transient collisions, the passive skin properties lowered the impact forces by about 40 %. During quasi-static contact, the effect of skin covers -- active or passive -- cannot be isolated from the collision detection and reaction by the collaborative robots. Important effects of the stop categories triggered by the active protective skin were found. We systematically compare the different settings and the empirically established safe velocities with prescriptions by the ISO/TS 15066. In some cases, up to the quadruple of the ISO/TS 15066 prescribed velocity can comply with the impact force limits and thus be considered safe. We propose an extension of the formulas relating impact force and permissible velocity that take into account the stiffness and compressible thickness of the protective cover, leading to better predictions of the collision forces. At the same time, this work emphasizes the need for in situ measurements as all the factors we studied -- presence of active/passive skin, safety stop settings, robot collision reaction, impact direction, and, of course, velocity -- have effects on the force evolution after impact.

Motivation & Objective

• To evaluate the impact of passive padding and active sensing in protective soft skins on collision forces in human-robot collaboration.
• To isolate and quantify the contributions of passive cushioning and active collision detection to robot safety.
• To investigate how robot-specific factors—such as safety stop settings, collision reaction, impact direction, and end-effector velocity—affect post-impact forces.
• To compare empirically measured safe velocities with ISO/TS 15066 prescriptions and identify conditions where higher velocities remain safe.
• To develop and validate an extended model for predicting collision forces that accounts for protective skin compliance, improving upon ISO/TS 15066 equations.

Proposed method

• Conducted 2,250 controlled collision experiments across three robots: UR10e, KUKA LBR iiwa, and KUKA Cybertech, using a custom impact rig with a human-like mass.
• Measured key collision metrics: peak impact force, contact duration, clamping force, and impulse under varying conditions.
• Used the AIRSKIN protective skin—available in passive-only and active-sensing variants—on all robots to isolate passive and active effects.
• Systematically varied: end-effector velocity (0.1–0.5 m/s), safety stop categories (Stop 0, 1, E-stop), impact direction, and robot collision reaction settings.
• Proposed an extended version of the ISO/TS 15066 force-velocity equation (Eq. 6) incorporating skin stiffness and compressible thickness to improve prediction accuracy.
• Publicly released the full dataset at https://osf.io/gwdbm for future model development and benchmarking.

Experimental results

Research questions

• RQ1 To what extent do passive protective skins reduce peak impact forces in transient human-robot collisions?
• RQ2 How do active protective skins (with sensor integration) influence robot collision detection and reaction, and how does this affect post-impact forces?
• RQ3 What is the combined effect of robot safety stop settings, skin type (passive/active), and impact direction on safe operating velocities?
• RQ4 Can safe velocities exceed ISO/TS 15066 prescriptions when protective skins are used, and under what conditions?
• RQ5 How well does the extended ISO/TS 15066 model, incorporating skin compliance, predict actual collision forces compared to the original standard?

Key findings

• Passive protective skins reduced transient impact forces by approximately 40% compared to bare robot surfaces.
• For the UR10e robot, connecting the active AIRSKIN to the E-stop allowed safe operation at 0.5 m/s—four times the 0.13 m/s limit prescribed by ISO/TS 15066 for clamping scenarios.
• On the KUKA iiwa, safe velocities reached up to 0.4 m/s when the active skin was connected to Stop 0 and external torque limits were disabled, exceeding the standard’s 0.16 m/s limit.
• The modified ISO/TS 15066 equation (Eq. 6), which includes skin stiffness and compressible thickness, provided more accurate predictions of collision forces than the original standard.
• The second phase of collision response in the KUKA iiwa (Type 2 impact) significantly altered force evolution, highlighting the need for in-situ measurements over theoretical models.
• Empirical measurements revealed that safety stop settings and skin connection to robot safety inputs had non-trivial, sometimes counterintuitive, effects on force profiles and safe velocity limits.