Projects / Programmes
Modelling and optimising human skill in controlling dynamic systems
Code |
Science |
Field |
Subfield |
2.07.07 |
Engineering sciences and technologies |
Computer science and informatics |
Intelligent systems - software |
Code |
Science |
Field |
P176 |
Natural sciences and mathematics |
Artificial intelligence |
artificial intelligence methods, machine learning, modelling, optimisation, genetic algorithms, process control, human skill, communication interface
Researchers (3)
no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
1. |
09006 |
PhD Mihael Junkar |
Manufacturing technologies and systems |
Researcher |
1998 - 1999 |
552 |
2. |
11562 |
PhD Viljem Križman |
Computer science and informatics |
Researcher |
1998 - 1999 |
30 |
3. |
04586 |
PhD Tanja Urbančič |
Computer science and informatics |
Head |
1997 - 1999 |
290 |
Organisations (2)
Abstract
The aim of the project is to enhance understandability and efficiency of operators'' activities in dynamic systems control. To this end, we develop methods for automated synthesis of models of human skill and combine them with multidimensional optimisation methods. In the three step process, operator''s activities are recorded, modelled by machine learning from the records, and optimised. by genetis algorithms. The process results in control rules that mimic human operator at the qualitative level of control strategy, but exhibits better characteristics with respect to the chosen optimisation criterion. Nonoptimised and optimised models of human skill are analysed in order to study individual characteristics of the operator''s strategy and to advice the subject how to improve his or her skill.The analysis of generated models can reveal features important in design of user friendly communication interfaces that can adapt to individual specificities of operators. The method is tested in two problem domains: control of a model container crane, and electrical discharge machining (EDM).