Research New Patent for AI-Based Energy Efficiency

Related Vendors

FAULHABER Drive Systems

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

SIMTOS (Seoul International Machine Tool Show)

The Salzburg University of Applied Sciences (Austria) and the Machine Automation Division of ABB (B&R) have filed a joint patent application in the field of energy-optimized drive systems. The innovation is a key result of the collaboration at the Josef Ressel Center for Intelligent and Safe Industrial Automation (JRZ ISIA).

The patent application focuses on a novel approach for energy-optimized motion control of drive systems in industrial automation —for example in robots, machine tools or automated production lines, in which highly dynamic motion sequences such as positioning, acceleration, deceleration or cycle sequences must be precisely controlled. While conventional control methods are based on mathematical models of the system that are as exact as possible, the new approach starts precisely where such models reach their limits: with real, often difficult-to-detect energy losses that are measurable but not fully modeled or known in detail.

Learning Agent on the Real System

By using artificial intelligence methods, in particular reinforcement learning (RL), a learning agent is used directly on the real system. This agent adaptively optimizes the movement strategies by independently learning through interaction how different movement sequences contribute to energy losses and adapting its control accordingly—without the need for a complete system model.

A key innovative contribution of the patent lies in a novel mathematical design of the learning strategy: it makes it possible to significantly accelerate the RL methods, which have often been too slow and data-intensive for industrial applications up to now, and to achieve better results at the same time. This makes practical use in cyber-physical systems economically and technically viable for the first time—with the aim of making motion sequences significantly more energy-efficient and taking real operating conditions fully into account.

AI Research With Industrial Relevance

"This patent application impressively demonstrates how scientific excellence and industrial practice go hand in hand at the Josef Ressel Center. Our goal is not to let research end in the laboratory, but to develop concrete technological innovations for industry," emphasizes Stefan Huber, Head of Research at FH Salzburg. "Especially in the field of artificial intelligence, we in Austria and Europe need first-class research at the cutting edge of technology that has a direct impact on the business areas of our industry."

ABB's Machine Automation Division (B&R) also sees the collaboration as strategically relevant: "The close cooperation with the Salzburg University of Applied Sciences enables us to quickly transfer innovative methods from research into industrial applications. The patent application submitted is a strong signal of the innovative strength that this partnership generates," says Martin Haidacher, Innovation Manager at B&R.

Joint Research—Grown Since 2020

The underlying research has a longer development history: initial work dates back to 2020 and was initiated in the EU Interreg project KI-Net. Since 2022, the topic has been further developed at the Josef Ressel Center—together with industrial partners from the Machine Automation Division (B&R), COPA-DATA and others.

Cooperation As A Driver for New Solutions

The joint application underlines the importance of long-term research cooperation between universities and—particularly in a field of technology that is of key relevance for energy efficiency, sustainability and the international competitiveness of Europe as an industrial location.