Call for Papers

As the demand for individualization is ever-increasing, products become platforms with ecosystems built around them. Integration is the key of such ecosystems. Mechatronics is characterized by its unique integrated view and understanding of systems. Machine vision complements mechatronics by supplying environmental information much needed for acting intelligently. Increasing computing power and decreasing sensor and actuator costs enable artificial intelligence methods to be deployed for consumer and industrial use. The 25th edition of the Mechatronics and Machine Vision in Practice (M2VIP) Conference aims to explore how the latest advances in mechatronics, machine vision and artificial intelligence shape the way products and production systems are designed, developed, manufactured (or implemented) and deployed. Thus, this conference provides a forum for researchers, academics and students to exchange and present their latest results and best practices.

All the papers presented at the conference will be submitted to IEEE Xplore for publication.

Keynote Speakers

Ralf Bucksch

Technical Executive Watson IoT, IBM

AI - the Era of Augmented Intelligence -
How can artificial intelligence (AI) provide insights from structured and unstructured data?

All IoT solutions are based on a platform on which data is collected. For this purpose, production level data – if necessary with an intermediate step via edge devices – must be transferred to a cloud platform and collected centrally. To make them readable and understandable, the data is put into an integrated context with the help of reporting functions. This makes it possible to establish dependencies between the data, the various production areas, product quality and machine availability. In the next step, the application of analytical functions helps to gain insights, for example to achieve improvements in product quality.

IBM complements these solutions with two elements: From predictive maintenance you move to prescriptive maintenance, i.e. it is not only said what happens, but what is the best action. The second element is the analysis of unstructured data. Unstructured data can take the form of images, manuals, machine repair tickets, flyers, Post-its, etc. The analysis system is taught-in, i.e. initially filled with structured and unstructured data. Depending on the quality and quantity of the data, the initialization takes different lengths of time. The result is faster and more flexible quality assurance and production optimization.

One example is Cognitive Visual Inspection. The basis is an algorithm for pattern recognition. It is taught what pictures of good and bad parts look like. Know-how from production, quality assurance, production and design data and external data such as the weather are taken into account. The parts in production are detected by a camera and analyzed for quality defects in (almost)
real time. Unlike an automated classic visual inspection, Watson learns, improves with each day and does not need to be readjusted.

The keynote presents various application scenarios that are also suitable for small and medium-sized enterprises and different batch sizes and complexities. The range of digitization of industry (Industry 4.0) and current developments in the field of artificial intelligence is stretched. Furthermore, an outlook is given from current to future capabilities in the area of hardware and software in IT.

Damon Kohler

Tech Lead, Google Munich’s cloud robotics team

Life-long Mapping with Cartographer and ROS: a Cloud Robotics Story
Keynote #3 tba
Invited Workshop

Soft Robotic Actuators and Sensors


Dr Jian Zhu and Dr Hongliang Ren (TBC), National University of Singapore; Singapore
Dr Mingzhu Zhu and Dr Mengyin Xie, Ritsumeikan University; Japan
Prof Peter Xu, The University of Auckland; New Zealand

Related Event

27 – 29 November 2018: SPS/IPC/Drives: smart and digital automation
Exhibition Centre Nuremberg
Messeplatz 1
90471 Nuremberg, Germany