IDIOM - Integrated Design and Optimization of Mechatronic Products

Products based on sound integration of mechanics, electronics, software and control are and will be important for the competitiveness of Swedish industry. There is a lack of design and verification methods as well as corresponding tools that adequately support efficient multi-domain synthesis and optimization of these products. The goal of the project is to develop a new model-based design method for mechatronic products. The hypothesis is that this new design method will lead to more optimal products, higher product quality as well as faster and more flexible product development thus leading to an increased innovation capacity in Swedish industry. One or two prototypes in the field of visionary brake and steering system for road vehicles will be designed with the new methodology and built for verification purposes.

Background

The method developed in the project is based on research previously conducted at the department by Fredrik Roos. In his thesis it is outlined that in order to reach the optimal design of a mechatronic system it needs to be considered in its whole in contrast to treating the different subparts individually.

Mechatronic servo
Mechatronic servo. Figure from Roos´ Ph.D. thesis.

Roos´ methodology is based on using simplified static models instead of complex dynamic ones, thus allowing for much faster evaluation and optimization computations. However, his method is for now limited to a fixed system configuration, which contains the basic parts of a mechatronic servo (electric motor driver, DC-motor and gearbox ). His method also mainly focuses on the optimization of volume.

The idea with IDIOM is to create a new method based on Roos´ ideas which will allow optimization of arbitrary system configurations. New components (e.g. shafts, motors, etc.) as well as different types of models will also be added. New model types could include for instance models describing dynamic behavior of the components. To test and validate the design method one or two innovative by-wire steering or braking prototypes will be developed. To support the new design methodology a tool framework will be designed and refined as the project progresses.

Current Progress

Matlab-kod

During 2010 a state of the art/practice survey about vehicular mechatronics, specifically design of by-wire systems, was conducted to serve as a basis for method and prototype design. This survey resulted in a report as well as a conference paper published at the IEEE International Conference on Mechatronics in Istanbul, Turkey, 2011. One of the most important conclusions in this report were that nearly all research focuses on single or few aspects of the system in question, for instance fuel consumption. However, to analyze if mechatronic braking and steering system are viable for use in a real production vehicles, there is a wide range of properties which needs to be considered in conjunction. Hence research with a more holistic approach even from an early stage is needed. Another conclusion drawn was that most research is only validated in software simulations or in some cases hardware-in-the-loop simulations. Thus most of the published research in the field misses the important validation in real conditions.

During 2011 research was focused on methodology. Roos´ method was analyzed and tested on a Brake-by-Wire system for heavy vehicles designed by Haldex.

Haldex EMB+ Brake-by-Wire system
Haldex EMB+ Brake-by-Wire system.

An investigation on how concepts are evaluated during design processes in practice was conducted in the form of a case study with five companies. The idea was to use the result as a basis for the support tool designed in the project. The study resulted in a conference paper which has been submitted for publication. Following, some of the conclusions drawn from the study:

  • The time it takes time it takes to model, simulate and evaluate a concept is of high importance (5 minutes can already be too long since the user will lose focus meanwhile).
  • The ability to quickly evaluate concepts with modeling and simulation instead of physical prototypes becomes vital (even large companies consider it too expensive to run multiple concepts in parallel).
  • Simple, fast, yet powerful, and well integrated tools still unavailable (simple enough to allow everyone to quickly describe and test their ideas, as illustrated in the figure below).
Illustration of a tool simple enough to capture and communicate design ideas at a common level of understanding between experts and non-experts in the field.
Illustration of a tool simple enough to capture and communicate design ideas at a common level of understanding between experts and non-experts in the field.

During late 2011 and early 2012 focus has been on developing the new methodology as well as a supporting tool. Early tests have shown promising results which were first published in late 2012.

In 2013 the methodology was significantly extended to now also consider system dynamics. The optimization does not only dimension a concept given constant constraints, e.g. peak torque, but now takes into account the time-varying constraint and its implications on the system. This forms the basis for integrating control design into the mechatronic design methodology for early concepts. Results were published at the IFAC Symposium on Mechatronic Systems as well as in the Proceedings of the Advanced Intelligent Mechatronics Conference which both were held during 2013.

Recently, an extended article was submitted to the Proceedings of the IMechE, Part I: Journal of Systems and Control Engineering. Further, two conference papers were submitted to the Mechatronics Forum International Conference which will take place in Karlstad, Sweden in June 2014.

Current Status

A PhD thesis  by Daniel Malmquist on holistic design optimization of mechatronic systems and the supported software IDIOM framework was concluded in 2015. Another PhD thesis on this topic and mostly focusing on control design and optimization is close to be concluded. And currently, the integration of design and optimization is further developed and is built upon the concluded doctoral thesis. The realization of non-linear multi degree of freedom mechatronic products and systems are being analysed and several new physical components in terms of physical dimension, dynamic and static models are added to the library to examine the concepts and discussions in these areas. An article focusing on early phase design optimization of nonlinear systems is published in ICCMA conference in October 2017. The mathematical formulation of the abovementioned framework is in progress to be published in a journal paper and discussions on the design optimization of more complex nonlinear multi DOF robotic manipulator case is under study to be implemented in the aforesaid framework.

Prototype

During the project's progress concepts for different prototype alternatives have been studied and the conclusions so far are that focus will be put on building one main prototype instead of two. To demonstrate the methodology's potential, a real design case has been carried out where the methodology was applied. In particular, a haptic steering device is designed which is will be integrated into KTH's very own , of which a CAD rendering is given below.

KTH Research Concept Vehicle (rendering)

Project Goals

The main goals of the project are:

  • To provide a new integrated design and optimization methodology based on previously developed methods of the whole design process.
  • To provide supporting software tools for the design methodology. Prototypical tools should have reached a maturity level such that it can be applied for testing in supporting companies in the later stages of the project.
  • To demonstrate one or two visionary, integrated and optimized product concepts designed with the developed method and tools.
  • To demonstrate that the method and tools improve development efficiency and product performance.

Research Approach

The Goals of the project will be realized by following the approach described here.

  1. Investigate two high-end product cases including design methods and tools
  2. Develop and further refine existing development methodology and tools
  3. Create/refine two new and visionary and strategic product concepts (design integration and optimization)
  4. Partially design the two visionary products and refine the methodology
  5. Implement and verify product prototypes
  6. Conclude, discuss and disseminate

Publications

  • Frede, D., Khodabakhshian, M., Malmquist, D., and Wikander, J. - "A survey on safety-critical vehicular mechatronics", in Proceedings of "Mechatronics (ICM), 2011 IEEE International Conference on", pages 176-181, Istanbul, Turkey, April 2011.
  • Frede D., Khodabakhshian M., and Malmquist D. - "A state-of-the-art survey on vehicular mechatronics focusing on by-wire systems". Technical report, ISSN 1400-1179, Department of Machine Design, KTH, 2010.
  • Malmquist, D., Frede, D., and Wikander, J. – "Optimization of Mechatronic Systems for Early Design Decisions", in Proceedings of the 13th Mechatronics Forum International Conference, Linz, Austria, September 2012.
  • Frede, D. , Malmquist, D, and Wikander, J. – "Holistic Design Optimization in Mechatronics", in Proceedings of the 6th IFAC Symposium on Mechatronic Systems, Hangzhou, China, April 10-12, 2013

  • Malmquist, D., Wikander, J., and Frede, D. – "Optimal Design of Harmonic Drive Servo", in Proceedings of the 2013 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, July 9-12, Wollongong, Australia.
  • Rahimi, F., Feng, L., Wikander, J., Frede, D., Early Phase Design-Optimization of Mechatronic Systems. International Conference on Control, Mechatronics and Automation, Edmonton, Canada, October 2017.

Funding

ProViking, Stiftelsen för Strategisk Forskning

IDIOM is funded by ProViking which is a program run within the Swedish Foundation for Strategic Research . The goal of the program is to support research in the field of product realization (product development, manufacturing, product support and maintenance in a life-cycle perspective) .

Project Partners

Haldex  - Developes innovative commercial vehicle technology with focus on brake and suspension products

SKF  - The leading global supplier of products, solutions and services within rolling bearings, seals, mechatronics, services and lubrication systems.

Scania  - A leading manufacturer of heavy trucks, buses and coaches as well as industrial and marine engines.

Research Team:

Daniel Malmquist
Ph.D. Student
malmqui@kth.se
+46 8 790 78 03
Daniel Frede
Ph.D. Student
frede@kth.se
+46 8 790 63 08
Jan Wikander
Jan Wikander
Professor, Mechatronics
jan@md.kth.se
+46 8 790 73 70

Ph.D. Student
frahimi@kth.se
+46 8 790 67 48

Associate Professor, Mechatronics
lfeng@kth.se
+46 8 790 76 49