Industry and academia in machine learning collaboration

26-06-2019 |   |  By Nnamdi Anyadike

A number of companies are engaging with academic institutes to examine how best to develop and commercialise technologies for ‘intelligent’ production. One such potentially groundbreaking project is the five year collaboration known as ‘Engineering for Smart Manufacturing’ (E4SM). It was launched this May by the Technical University of Ilmenau in Germany and is funded by the Carl Zeiss Foundation as part of the Intelligent Systems Research Program. It links high-profile companies, including: Robert Bosch GmbH; the Honda Research Institute Europe; the machine and plant designer LASO tech Systems from Suhl; the manufacturer of service robots and robot platforms MetraLabs from Ilmenau; the engineering company Henkel and Roth from Ilmenau and the TÜV Thuringia. It aims to provide innovative technologies for intelligent manufacturing and assembly in the Industry 4.0 age.

Concrete applications that can be expected from the expansion of artificial intelligence (AI) and machine learning (ML) include: robot-assisted device-free laser beam welding and collaborative assembly processes. But in fact, the applications spread much wider as machines have emerged as a vital part of many enterprise productivity strategies. ML algorithms can gather data about customers thereby helping businesses to provide better service. ML can also use the information it gathers from data to provide insights into a company’s future. Indeed, when combined with other AI tools, the Internet of Things and CPaaS technology, it is possible for a business network to begin to solve a problem before the company’s management is even aware of its existence. This could prove crucial in predicting potential outages or fixing issues with hardware.

SONARO pushes healthcare robot technology

An early success from the Ilmenau Technical University project team is in healthcare. ‘Smart’ health care relies on innovative digital technologies and increasingly intelligent, interactive systems are used for human-machine interactions. Today, assistance robots already support nursing staff in the healthcare sector. But in the future, they will have the ability to hand patients their medication or take objects from them. In that regard, the development of ever more dexterous robots is crucial. In May, the research group SONARO (Smart Object Takeover and Transfer for User-Centered Mobile Assistance Robotics) was launched at the Ilmenau Technical University. As well as Ilmenau, SONARO’s founders include: the Thuringian Center for Mechanical Engineering; the Schmalkalden University of Applied Sciences and the Schmalkalden Manufacturing Technology and Development Company; the Honda Research Institute Europe; Metralabs GmbH; Vision & Control GmbH; Hörisch precision GmbH; Götting KG; and SCS Robotik UG. The project leaders said, “It will explore novel methods for smart object handover and takeover in human-robot collaboration that go far beyond the current state of interactive assistance robotics.”

SONARO’s technology will allow assistive robots, when interacting with humans, to situationally adapt their actions to the person and their current activity. The team explained, “If the robot wants to pick up an object from a human, it must be able to recognise its holding pose and the grasping position of the hand in order not only to grasp the object safely but also not to endanger the human being. Then he has to be able to to transport the inherited object independently and safely to another person and to hand it over without endangering it. For the scientists and engineers of the research group, this means that they have to develop methods to contactlessly monitor and analyze the common interaction space of humans and robots and their respective activities.”

Oil and gas industry leads the way

Within the process industries it is the oil and gas industry, followed by the consumer packaged goods and the materials, minerals and mining sectors, that is the trail blazer in the use of ML and data science. Offshore oil and gas companies are better able to monitor complex internal operations and respond quickly to concerns that human operators may not have been able to detect. Earlier this year, BP invested in the Houston-based technology start-up Belmont Technology to bolster the company’s AI capabilities, developing a cloud-based geoscience platform nicknamed ‘Sandy.’ This allows BP to interpret geology, geophysics, historic and reservoir project information, creating unique ‘knowledge-graphs.’ The Oil and Gas Authority (OGA) is also making use of AI in similar ways, with the UK’s first oil and gas National Data Repository (NDR), launched in March. Also in March, Aker Solutions partnered with tech company SparkCognition to enhance AI applications in its ‘Cognitive Operation’ initiative. Last September, Shell adopted similar AI software when it partnered with Microsoft to incorporate the Azure C3 Internet of Things software platform into its offshore operations.


A recent analysis of the ML industry by Gartner, the US based leading research and advisory company, claims that the industry is now growing at a whopping 48.3% CAGR. The report comments, “Investment in AI and machine learning is expected to reach $100 billion by the year 2025... We’re already beginning to see virtual agents take over the roles of many old-fashioned IVR menus.”


By Nnamdi Anyadike

I have 30 years experience as a freelance business, economy and industry journalist, concentrating on the oil, gas and renewable energy, telecommunications and IT sectors. I have authored a number of well received in-depth market intelligence reports. And I have also spoken at conferences.

Related articles