China's First Industrial Embodied Quality Inspection Robot Mass-Produced and Delivered in Luoyang
2026-06-18 14:44
Favorite

en.Wedoany.com Reported - On June 16, CASIVISION Technology (Luoyang) Co., Ltd. achieved the batch production and delivery of China's first industrial embodied quality inspection robot, CASIVIBOT, marking its official entry into mass production and large-scale application, propelling industrial quality inspection into the 3.0 era of "embodied intelligence + ecological synergy."

CASIVISION is a national-level specialized and innovative "Little Giant" enterprise incubated by the Institute of Automation, Chinese Academy of Sciences in Luoyang. Leveraging over a decade of technical expertise in glass appearance defect detection, the company has built an industrial intelligence system featuring "hand-eye-brain" coordination, and released the first CASIVIBOT prototype in August last year.

Sun Erdong, R&D Director of the Embodied Robot Division at CASIVISION, stated that the mass-produced version has been optimized for practical application scenarios, addressing issues such as low efficiency, high rates of missed and false detections, and persistently high costs in manual quality inspection. It boasts core advantages including "dexterous hands," "sharp eyes," and "intelligent brain."

"Dexterous hands" refer to the product's integrated "hand-eye-brain" collaborative architecture, where multi-degree-of-freedom humanoid dual arms, combined with AI algorithms, enable 360-degree scanning without blind spots. "Sharp eyes" denote a multispectral sensing system that achieves human-eye-like multimodal imaging with micron-level detection accuracy, with motion and detection precision reaching internationally advanced levels.

"Intelligent brain" means the product is equipped with a million-level precisely annotated defect sample library and a vertical industry large model, avoiding the "AI hallucination" issues common in general-purpose large models. It supports identifying new defects guided by natural language descriptions and enables cloud-based "one-machine training, group deployment," compressing the detection and calibration cycle from dozens of hours in traditional solutions to minutes. Its judgment logic and recognition capabilities are comparable to those of senior quality inspectors.

Currently, the product has achieved small-batch mass production and is stably applied in industrial inspection scenarios such as automotive motor stators, cemented carbide blades, precision tools, and packaging products. It can accurately identify defects like chipping, missing pieces, pores, broken magnets, foreign matter contamination, and adhesive residue, with false and missed detection rates far below manual levels.

Sun Erdong noted that the product has passed practical testing, providing effective solutions to pain points such as insufficient flexibility in traditional quality inspection, high reliance on manual labor, and difficulties in data closed-loop management. It is expected that shipments will exceed 100 units this year.

A representative from the Luoyang Municipal Science and Technology Bureau stated that Luoyang has been actively cultivating emerging and future industries in recent years, incorporating artificial intelligence and robotics into 12 key advantageous industrial chains for focused development. Leveraging leading enterprises like CASIVISION and CITIC Heavy Industries, the city has continuously advanced technological R&D and scenario expansion, achieving a series of progress.

In the next step, Luoyang will continue to harness the driving effect of leading enterprises, promote coordinated development across the industrial chain, expand the scale of the AI and robotics industry, and provide momentum for cultivating new quality productive forces and building a modern industrial system with advanced manufacturing as its backbone.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com