ABB-B&R Patents AI System to Cut Energy Use in PLC Motion Control

Detail

The patent, announced this week, targets the core of industrial motion: positioning, acceleration, deceleration, and cyclic movement sequences. These highly dynamic operations, executed millions of times daily in robots, CNC machine tools, and automated production lines, account for a dominant share of factory-floor electricity consumption. The new system uses artificial intelligence to optimize these motion profiles in real time without compromising the precision that PLC-driven automation environments demand.

This patent signals a strategic inflection point. Rather than treating energy efficiency as a system-level afterthought addressed through regenerative braking or efficient motors, ABB-B&R is embedding AI-driven optimization directly into the motion control layer , the firmware that governs every movement a drive executes. This bottom-up approach could yield compound savings across entire production lines.

The collaboration is anchored at the Josef Ressel Center for Intelligent and Secure Industrial Automation (JRZ ISIA) in Salzburg , a research hub specifically designed to bridge advanced AI theory with industrial-grade implementation. The center's mandate is to develop digital assistants that increase the autonomy of industrial machines through artificial intelligence, and this patent represents one of its most commercially significant outputs to date.

Central to the innovation is a new mathematical formulation of the learning strategy. Traditional reinforcement learning approaches require vast datasets and extended training cycles , luxuries rarely available on live production lines. The patented formulation dramatically reduces both the volume of data and the time needed for the AI to learn optimal motion profiles, making the technology viable for deployment in cyber-physical systems operating under real-world conditions.

The system applies multi-objective reinforcement learning to industrial drive control. During operation, the AI continuously evaluates motion sequences , positioning moves, acceleration ramps, deceleration curves, and cyclic patterns , and adjusts parameters to minimize energy draw while maintaining required precision thresholds. Unlike static pre-programmed motion profiles, the AI learns and adapts to the specific mechanical characteristics of each machine, including inertia, friction, and load variations. The key breakthrough is the mathematical formulation that compresses the learning phase, enabling the system to converge on optimal control strategies with significantly fewer iterations than conventional reinforcement learning methods.

The timing of this patent is not coincidental. ABB's own internal studies reveal that more than 70% of a customer robot's carbon footprint stems from electricity consumption during operational life. Meanwhile, a recent Energy Efficiency Movement report found that 58% of global manufacturers cite high energy costs as a direct threat to competitiveness, and over 93% plan to invest in energy efficiency improvements within the next three years.

The global industrial automation market, valued at approximately USD 272.5 billion in 2025, is projected to expand at a CAGR of 9.8% through 2034. Within this growth trajectory, energy efficiency has emerged as a non-negotiable purchasing criterion for PLC and motion control systems. European manufacturers, who commanded a 32.99% market share in 2025, face particularly stringent regulatory pressure under the EU's evolving Energy Efficiency Directive and carbon border adjustment mechanisms.