Doosan, Microsoft plan to construct GPT-based robots

Hearken to this text

Doosan Robotics’ E-Sequence of cobots is NSF-certified. | Supply: Doosan Robotics

Doosan Robotics has partnered with Microsoft and Doosan Digital Innovation to offer collaborative robots (cobots) a context-aware management system primarily based on generative pre-trained transformer know-how. The partnership was formalized with a memorandum of understanding for Doosan Robotics to make use of Microsoft’s Azure OpenAI Service for a GPT-based robotic management system. 

As a part of the settlement, Azure will present Doosan with entry to the GPT language mannequin, and Doosan Digital Innovation shall be chargeable for adapting the mannequin for use in Doosan’s robots. Doosan hopes this integration will shorten programming instances whereas additionally increasing the obtainable features for his or her robots. 

Doosan plans to leverage GPT’s comprehension capabilities to permit robots to achieve extra context and nuance in understanding duties. GPT is a big language mannequin that may perceive and generate human-like textual content primarily based on the huge quantities of knowledge it’s educated on. 

The corporate hopes to combine GPT into its system, cobots will be capable of self-correct errors throughout operation and anticipate and regulate their actions primarily based on the context. 

For instance, Doosan’s E-SERIES cobots are designed for meals and beverage preparation. With GPT built-in, a person may difficulty a easy voice command to organize a dish, and the robotic may work out the optimum sequence of duties to finish the meal. 

GPT integration additionally permits robots to entry and study from an present database of code, which is able to make it so programmers don’t must program each new job from scratch. 

Doosan Robotics plans to first check this know-how within the meals and beverage sector later this 12 months. Relying on the outcomes of this testing, the corporate might discover utilizing it within the manufacturing sector as effectively. 

At RoboBusiness, which takes place October 18-19 in Santa Clara, CA, a keynote panel of robotics business leaders will talk about the functions of Massive Language Fashions (LLMs) and textual content technology functions to robotics. It is going to additionally discover basic methods generative AI may be utilized to robotics design, mannequin coaching, simulation, management algorithms and product commercialization.

The panel will embrace Pras Velagapudi, VP of Innovation at Agility Robotics, Jeff Linnell, CEO and Founding father of Formant, Ken Goldberg, the William S. Floyd Jr. Distinguished Chair in Engineering at UC Berkeley, Amit Goel, the Director of Product Administration at NVIDIA, and Ted Larson, the CEO of OLogic.