The Industry 4.0 revolution wants to radically change the industrial production, favoring a tight interaction between human workers and intelligent machines. The goal of this revolution is to increase the productivity while keeping high quality standards. This workshop aims at offering a different point of view on machine learning in robotics, which is the point of view of modern manufactures. Factories have specific requirements in terms of safety, repeatability, and predictability that “intelligent” machines have to satisfy. However, some of the requirements are often neglected in academic research.
The scope of this workshop is reduce the gap between machine learning research and smart factories. We want to give the possibility to young researchers, such as Post-Docs and Ph.D. students, to discuss their ideas and preliminary results with worldwide recognized experts in the fields of machine learning, robot control, manipulation, sensor and actuator design. Moreover, an important objective is to stimulate discussion and dialogue between industrial and academic researchers to reduce the gap between the capability of lab prototypes and the requirements of industrial application.
Some of the questions we will try to answer are:
- Which factors limit the applicability of machine learning techniques in industrial scenarios?
- How learning and intuitive programming can help flexible manufacturing?
- Can we speed-up the learning process for complex robotic tasks?
- Which is the role of mobile and dexterous manipulation in smart factories?
- Are available sensors and actuators suitable for industrial tasks?
Learning for Industry 4.0: feasibility and challenges 