AttentiveMachines – Man and Machines in ICT based Production Systems of the Future

Sensitive Machines cooperate with Plant Personnel. Attention-aware Machines adapt to the human’s performance and constitute the production systems of the future.

Short Description

Future ICT-based manufacturing systems aim at a radical individualization of products (Lot-size 1) under the pre-condition of highly flexible mass production environments.

A key challenge towards such systems is the implementation of confluent collaborative cooperation between humans and machines. The Attentive Machines project aims at attention-aware self-adaptive machines pacing the interaction with human workers based on observed user attention and task performance and up-to-the minute adaption of the level of assistive support provided to human workers during the manufacturing, semi-manual assembly process.

This project will realize attentive machines based on

  • formal models of human attention,
  • multisensory recognition architectures based on machine learning and pattern recognition approaches and
  • realization of embedded subtle assistive support mechanisms in real-world production systems.

The developed system will be empirically evaluated in the scope of two cases of Austrian manufacturing excellence to significantly improve product quality assurance (Fischer Sports GmbH) and to significantly advance worker safety in heavy duty metallic manufacturing processes (Aumayr GmbH).

Based on the fusion of multi-sensor data (eye tracking, behavior analysis via motion tracking, capturing of somatic and psycho-physiologic attention indicators), activity, cognitive workload as well as orientation and quality of attention are supposed to be detected and modeled to provide suitable assistance feedback (visual, acoustic, haptic) to support the correct execution of the respective collaborative task.

The explicit goal of this project is not to replace workers via extended automation of production systems, but to optimize the collaborative cooperation between humans and machines via an encompassing automatic perception of the human being, its intentions and attention distributions.

Project Partners

  • Institut für Pervasive Computing, JKU Linz
  • Research Studios Austria FG, PCA
  • Fischer Sports GmbH
  • Aumayr GmbH

Contact Address

Univ.Prof. Mag. Dr. Alois Ferscha
E-Mail: alois.ferscha@jku.at