Bonsai and Siemens Demonstrate Applicability of AI to Industry by Reducing Machine Calibration Time More than 30x in a Joint Proof-of-Concept

Unique deep reinforcement learning platform combines the best of human and machine intelligence to train machine tools

AI startup Bonsai today announced, together with Siemens, a technological highlight in deploying AI on a real-world machine in a test environment. Using Bonsai’s AI Platform, Siemens subject matter experts trained an AI model to auto-calibrate a Computer Numerical Control (CNC) machine more than 30x faster than an expert human operator (estimation performed by Commonwealth Center for Advanced Manufacturing (CCAM) in Virginia). This marks the first time deep reinforcement learning has been successfully applied to auto-calibrate real-world CNC machines.

CNC machines, or computer-controlled machine tools, have revolutionized manufacturing since their inception in the 1940s. However, the value that CNC machines provide global manufacturers is constrained by high maintenance costs. To achieve highest possible quality of production, CNC machines need to be recalibrated frequently, as even minor friction leads to errors that result in costly manufacturing imperfections. Manufacturers have to fly in specialist engineers to do the job, which can take hours. While machines are decommissioned for maintenance, downtime and service costs can run up to several thousand dollars. Costs run especially high when unplanned errors arise outside the regular maintenance schedule.

Balancing human and machine intelligence
To explore the possibilities of AI for their CNC business (see editor’s note), Siemens partnered with Bonsai, which has pioneered an AI platform based on deep reinforcement learning. At the platform’s core is an innovative ‘Machine Teaching’ technique, which enables subject matter experts such as specialist engineers to train machines to efficiently perform complex task. Using a simple scripting language, experts can design the ‘lessons’ and ‘rewards’ required to train each task. Bonsai’s AI Engine supports a wide range of state-of-the-art deep reinforcement learning algorithms, along with the logic for choosing the best-fit algorithms and guiding the training. In this way, the experts are able to leverage AI without themselves having to gain a deep understanding of machine learning.

Teaching a CNC machine to calibrate itself significantly faster
To build this proof of concept, the team used Bonsai’s AI engine to build a predictive model that would calibrate the CNC machine. Each model produced by Bonsai is referred to as a BRAIN (Basic Recurrent Artificial Intelligence Network). The AI engine trains each BRAIN using cutting-edge deep reinforcement learning algorithms.

After six months of PoC, including training of the algorithms in a simulation environment, CCAM tested a BRAINs’ ability to calibrate a Siemens CNC controlled machine. The results were a great step forward. The most successful BRAINs calibrated a CNC machine more than 30x faster than the human operators, while achieving precision of less than two microns (as estimated by Commonwealth Center for Advanced Manufacturing (CCAM) in Virginia).

Siemens is delighted with the results, which are an important step in using AI to improve significantly the productivity, cost and quality of the CNC machine calibration process.  Michal Skubacz, Siemens Vice President and Head of Industry Software at Siemens Motion Control  concluded, “The results we achieved using Bonsai demonstrate that organizations can deploy the latest AI technologies in a noisy real-world system. The solution possible based on the proof-of-concept with Bonsai could augment and scale the work of our best operators. Instead of having operators carry out the same work repeatedly, they can focus on training the machines to perform better and more advanced tasks.”

Mark Hammond, CEO and co-founder of Bonsai, “Our successful project with Siemens represents a huge milestone in industrial AI, demonstrating the powerful results that can be achieved by combining machine teaching and machine learning. The beauty of this approach is that it balances the best of human and machine intelligence. Applied across the whole industrial manufacturing sector, the implications are staggering.”