Real-time technologies in driving dynamics performance simulation through full vehicle multibody, 1D chassis subsystems and tire modeling integration

Research and Objective: Detailed plant modelling by means of a multi-body simulation model is the typical approach for virtual prototyping. Such models, complete of tire, work off-line. However, for the design and validation of control or intelligent systems Real-Time vehicle models are required. Typically such simulations require the model to be simplified by means of look-up tables or elasto-kinematic maps, as well as usage of empirical models of the tire, weakening the confidence of the multi-body model with respect to the real world. Scope of this project has been the development of a methodology able to maintain the complexity of the full vehicle multi-body model for Real-Time applications. Methodology: LMS has researched, developed and implemented in its multi-body tool Virtual Lab Motion a methodology able to maintain the complexity of the full vehicle multi-body model through usage of the multiple cores computing technology. The chassis subsystems, such as EPS, ABS/ESC and transmission, are also implemented in co-simulation with the multi-body model. The first part of this paper documents the changes necessary to obtain a full vehicle multi-body model in co-simulation with 1D chassis subsystems running Real-Time. In the frame of the CHASING R&D project (""Advanced Simulation Methodologies for Chassis & Suspension Engineering: Optimizing Driving Dynamics of Intelligent Vehicles"", supported by IWT Vlaanderen in the frame of the EUREKA European research project E!4907), Fraunhofer ITWM has researched and developed an innovative CDTire model suitable for comfort and durability application in hard Real-Time. The second part of this paper documents the changes necessary to obtain a complex tire model suitable for ride and durability applications running Real-Time. Results: The new tire Real-Time capable sub-model CDTire/Realtime (High Performance Solver) has been interfaced to the Real-Time capable LMS multi-body tool Virtual Lab Motion for front and rear suspensions in combination with LMS 1D tool Imagine Lab AMESim for chassis subsystems, to enable running Real-Time applications in comfort and durability scenarios. The integration in LMS Virtual.Lab Motion has been setup to perform offline applications for multi-attribute vehicle performance optimization or robust design analyses as well as hard Real-Time applications like SIL or HIL. The paper documents the changes implemented (implicit integration, parallel processing and no reduction) to obtain a complete comfort and durability Real-Time vehicle model and the good results obtained in comparison with an off-line high-fidelity solution model. A final demonstration on full Real-Time vehicle model in combination with tire is also presented. Limitations of this study: Currently the Real-Time multi-body modelling does not include any flexible part, with a limitation to extension of the frequency validity range, even if now extended till 20-25 Hz by means of the full vehicle multi-body modelling. Next step will be the implementation of main flexible parts (front and rear sub frames) and of the complete trimmed body model to further extend the frequency validity range (i.e. till 50-60 Hz). What does the paper offer that is new in the field in comparison to other works of the author_ Development of a complete vehicle multi-body modelling technology, including chassis subsystems (steer, brake, transmission) modelled in 1D environment, able to run in hard Real-Time for SiL and HiL applications. Development of a tire model suitable for multi-body comfort and durability applications in offline and hard Real-Time. Complete multi-body vehicle, chassis subsystem and tire integration able to run hard Real-Time to feature transient comfort and durability applications. Accelerated multi-attribute balancing methodology and statistical regression analysis for robust design analysis for R&H and Durability road loads data prediction through usage of offline multi-body vehicle model in combination with dedicated tire model. Conclusion: It is possible to increase the fidelity of Real-Time models, increasing the application of the frequency range, by means of usage of multi-body vehicle models, chassis subsystems models and tire semi-physical models. This increased fidelity is possible by taking advantage of implicit integration, parallel processing and no reduction. Results show good correlation with an off-line high-fidelity solution.