Benchmarking human-robot collaborative assembly tasks

TL;DR

The paper addresses the lack of standardized benchmarks for human-robot collaborative assembly by introducing the CT benchmark. It proposes a city landscape object set with seven sub-assemblies and five common tasks, implemented with open STL models and evaluated across manual, collaborative, and automated modes using assembly time and NASA-TLX workloads. Key contributions include a reusable benchmarking framework, a rich, 3D-printable component set, and comparative results showing that collaborative assembly reduces operator workload but increases total time by about 70.8% compared with manual. The work provides a practical, reproducible platform for evaluating HRC assembly strategies and informing deployment decisions in manufacturing.

Abstract

Manufacturing assembly tasks can vary in complexity and level of automation. Yet, achieving full automation can be challenging and inefficient, particularly due to the complexity of certain assembly operations. Human-robot collaborative work, leveraging the strengths of human labor alongside the capabilities of robots, can be a solution for enhancing efficiency. This paper introduces the CT benchmark, a benchmark and model set designed to facilitate the testing and evaluation of human-robot collaborative assembly scenarios. It was designed to compare manual and automatic processes using metrics such as the assembly time and human workload. The components of the model set can be assembled through the most common assembly tasks, each with varying levels of difficulty. The CT benchmark was designed with a focus on its applicability in human-robot collaborative environments, with the aim of ensuring the reproducibility and replicability of experiments. Experiments were carried out to assess assembly performance in three different setups (manual, automatic and collaborative), measuring metrics related to the assembly time and the workload on human operators. The results suggest that the collaborative approach takes longer than the fully manual assembly, with an increase of 70.8%. However, users reported a lower overall workload, as well as reduced mental demand, physical demand, and effort according to the NASA-TLX questionnaire.

Figures (8)

• Figure 1: Human and robot collaborate to complete the assembly of the city landscape model proposed in the benchmark.
• Figure 2: The object set of the CT benchmark, depicting a city landscape model combining five common manufacturing tasks which can be performed by humans or robots. The benchmark comprises seven different sub-assemblies related to the various buildings.
• Figure 3: Schematic of a proposed benchmark sub-assembly sequencing for each setup: fully manual, collaborative and fully automatic. Tasks related to each sub-assembly are detailed in Table \ref{['table:Table']}.
• Figure 4: Snapshots of the proposed benchmark being used in fully manual assembly (M1 to M4), human-robot collaborative assembly (C1 to C4), and fully automatic assembly performed by a robot (A1 to A4). In the collaborative setup, the robot is equipped with a gripper, while in the fully automatic setup, the robot is equipped with either a gripper or a screwdriver tool.
• Figure 5: Total time registered for each participant after performing the manual and collaborative assemblies. Participants 1 to 8 first executed the manual assembly and participants 9 to 15 initially executed the collaborative assembly. The last result illustrates the total time registered for the fully robotic assembly.