???? Taiwan at the Forefront: Robots, AI, and the Future of Semiconductor Innovation
Taiwan’s Techman Robot CEO: Nvidia Boosts AI Robot Training
Will the combination of AI and robots lead to humanoid robots in our everyday lives? Haw Chen, CEO of the world’s second-largest collaborative robot manufacturer, Techman Robot, told the CommonWealth Magazine Economic Forum that thanks to the AI-backed training of novel robots, robot development has entered a new realm.
Techman Robot began working on machine vision in 2012, invested in AI in 2018, and subsequently developed AI applications.
Presently, our product lines include everything from standalone robotic arms and system integration to AI model training platforms. In 2022, we repositioned our product as AI Cobot (AI collaborative robot).
Products Use Ranging from Cars to Toys
Our solutions are used by customers in the electronics, semiconductor, automobile, metal processing, plastics, furniture, and food industries.
In 2019, for instance, we helped a carmaker in Taiwan create a vision inspection system used to inspect car equipment.
Using four suspended robotic arms together with 30 cameras, we could inspect 120 items within 80 seconds, helping the customer ensure that orders match the cars about to leave the factory.
Because, depending on the orders, the cars’ model numbers, lamp shades, vehicle glass, and interior equipment differ. Wheel rims for the Taiwanese market don’t need protective film, while export models do. In the past, this inspection relied on human labor.
Another example of applying AI, vision, and robotic arms is a toymaker among our customers who exclusively produces action figures. The heads of these action figures must be inserted into a machine to plant hair, requiring vision-based positioning and inspection.
Since the heads of the action figures are round, it’s very challenging for the robotic arms to insert them into machines for processing. We used visual identification to help confirm whether the head was positioned correctly. If not, the system issues a warning.
In the metal processing field, we helped a customer distinguish between unprocessed raw materials and processed materials using machine vision. Before loading materials, visual recognition ensures no mixing or misalignment. The system also alerts if materials are tilted too much.
Read the full article:
领英推荐
The Future of TSMC’s CoWoS: Hybrid Bonding and Microfluidic Cooling
Techman Robots Makes World’s First AI-Vision Robotic Arm
Taiwan Enters Angstrom Era with ASML’s High-NA EUV
Podcast｜Taiwanology Ep.42
Subscribe to CWE newsletter (every Thursday)