The Internet of Things (IoT) is the network of objects that exchange data with other devices and systems over the internet. Machine-to-machine (M2M) communication has existed for a long time, but smart affordable sensors and improved connectivity in combination with cloud computing platforms and artificial intelligence have boosted the applications for IoT. NORCE is mainly involved with industrial IoT, especially with respect to instrumentation and smart sensors with embedded signal processing and interaction with cloud technologies for advanced data analysis, visualization and automation. Typical applications for industrial IoT are Smart manufacturing (Industry 4.0), automated process control, preventive and predictive maintenance, smart transport and logistics, smart cities and energy systems. NORCE have expertise on a wide range of IoT-related disciplines, and are currently involved in several projects where IoT is a significant part of the system design and innovation.
NORCE has contributed to the development of several IoT- solutions, like device tracking, process control and environmental monitoring. Our scientists have achieved extensive competence and experience on implementation of smart sensors based on the latest standards and innovative design, and have established a versatile platform for research and development of competitive solutions.
Smart Sensors
Interaction between machines is crucial in the digital changeover, and smart machines depend on smart sensors to observe conditions and surroundings. Miniaturization and increased computing power enable small and low-cost sensors to perform demanding signal processing, and to interact with other sensors and data sources to provide a good description of what they are observing. Sensors in networks (sensor fusion) increase both data quality and reliability, allowing for high density distributed monitoring and machine learning.
Wireless sensors are more easily mounted in existing production environments and in areas without any infrastructure. The installation cost is therefore far less compared to traditional measurement methods. Efficient signal processing and data transfer give a battery lifetime of many years, and adaptive measurement algorithms can provide high performance when needed and still maintain low average power consumption.
Real-time analysis of sensor data in combination with historical data and additional online information enables prediction of future events. Thus, condition-based maintenance and process optimization can reduce the risk of accidents or unwanted environmental impacts, and provide more cost-efficient operation. Machine learning based on large amounts of data can identify phenomena that are impossible to reveal manually, and thereby provide completely new opportunities for utilizing available information, both from own data sources and other databases.
