A Systematic Literature Review of Computer Vision Applications in Robotized Wire Harness Assembly

TL;DR

The paper surveys computer vision applications in robotized automotive wire harness assembly, framing the challenge as automating perception for manipulation of deformable, tree-like wire structures. It catalogs 15 CV-focused works across clamp manipulation, connector mating, wire recognition, and bag segmentation, highlighting shifts from marker-based to learning-based, and from 2D to multi-modal 3D sensing. Key contributions include a synthesis of methodologies, evaluation practices, and a critique of the gap between laboratory demonstrations and production deployment, with clear callouts for datasets, benchmarks, and production-oriented metrics. The study concludes that robust, multi-modal perception and human-robot collaboration, along with redesigned harness components and standardized evaluation, are essential for industrialization of robotized wire harness assembly.

Abstract

This article provides a systematic literature review of computer vision applications in robotized wire harness assembly.

Figures (6)

• Figure 1: An example of the floor harnesses to be installed into passenger cabins of automobiles. (Courtesy of Volvo Car Corporation) Color should be used for this figure in print and size should be calibrated for the camera-ready version (This sentence will be removed for the final submission).
• Figure 2: The electrical infrastructure of a Volvo XC40 Recharge, consisting of front, central, cabin, and rear cable systems. (Courtesy of Volvo Car Corporation) Color should be used for this figure in print and size should be calibrated for the camera-ready version (This sentence will be removed for the final submission).
• Figure 3: The remarkably increasing length of wires in passenger cars over time. (Courtesy of Volvo Car Corporation) Color should be used for this figure in print and size should be calibrated for the camera-ready version (This sentence will be removed for the final submission).
• Figure 4: PRISMA page2021prisma flow diagram of review process. Color should be used for this figure in print and size should be calibrated for the camera-ready version (This sentence will be removed for the final submission).
• Figure 5: The number of publications per year and the total number of publications and citations by each year (data retrieved from Scopus on September 6, 2023). The total number of publications and citations for the year 2023 are excluded considering the incomplete statistics at the time of literature searching. Color should be used for this figure in print and size should be calibrated for the camera-ready version (This sentence will be removed for the final submission).