Technische Universität München Robotics and Embedded Systems



Autonomous, self organizing Sensor Networks for Process Monitoring

In this project, a robust and self organizing architecture for sensor networks is being developed for the process monitoring domain. The results will be directly applied to industrial manufacturing processes in tire production and reflow soldering. A sensor system suitable for these processes consists of statically mounted and flexible placeable sensors. Using sensors with a wireless communication provides the flexibility to easily adapt the network. After the deployment phase it is easily possible to optimize the sensor coverage or to provide additional sensing resolution at specified locations for calibration tasks only by placing additional sensors there.

System architecture

In contrast to the cabled sensors, which have a constant power supply, the wireless sensors need to be powered by batteries. To increase the lifetime, energy harvesting techniques will be used in some setups. A further aspect to increase node lifetime is to reduce energy consumption. Usually the most power is needed for communication. To minimize the network traffic and thus power consumption, the sensor data will be pre-processed by the sensor nodes to minimize the size and number of the network packages.

Example nodes MicaZ node

Facing these challenges a new software architecture needs to be developed that is suitable for wired and wireless sensor nodes in the same way and which is easily implementable in industry processes.

To show the feasibility of this approach, a demonstrator will be created and, in further steps of the project, this sensor system will be used in a tyre production process and in a soldering plant operated by project partners.

Demonstration setup modeling an industrial production process

This work is funded by the German Ministry of Education and Research (BMBF).





[1] Stephan Sommer, Michael Geisinger, Christian Buckl, Gerd Bauer, and Alois Knoll. Reconfigurable industrial process monitoring using the CHROMOSOME middleware. In The Fifth International Workshop on Adaptive and Reconfigurable Embedded Systems (APRES 2013). ACM, April 2013. [ .bib | .pdf ]
[2] Gokul Balakrishnan, Michael Geisinger, and Christian Buckl. Multifunk: Self-organizing sensor networks for industrial process monitoring. In Jian-Jia Chen, editor, Proceedings of the 17th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'11), Work-in-Progress Session, pages 57-60, 2011. [ .bib | .pdf ]
[3] Christian Buckl and Michael Geisinger. Middleware-Architekturen zur Integration von Systemen in Systems-of-Systems. In Tagungsband Embedded Software Engineering Kongress, pages 38-42, Sindelfingen, Germany, 2011. [ .bib | .pdf ]
[4] Christian Buckl, Irina Gaponova, Michael Geisinger, Alois Knoll, and Edward A. Lee. Model-based specification of timing requirements. In Proceedings of the 10th ACM International Conference on Embedded Software (EMSOFT 2010), pages 239-248, Scottsdale, Arizona, USA, 2010. Association for Computer Machinery. [ DOI | .bib | .pdf ]