Tyre fingerprinting for rapid and cost-effective tyre model generation
Lead Research Organisation:
Loughborough University
Department Name: Aeronautical and Automotive Engineering
Abstract
Pertinent EPSRC themes: Engineering & Manufacturing the future
Pertinent EPSRC research area: Engineering design, Artificial Intelligence Technologies
Scope and context of project:
The project will develop a new method for rapid, accurate and economical tyre model parameter identification. Mathematical tyre models are used in vehicle simulations for prediction of tyre forces. Vehicle design for passive safety and the design/calibration of active safety systems like (ABS) Anti-Lock Braking and ESP (Electronic Stability Programme), all depend on the ability to simulate tyre behaviour faithfully. All tyre models include several parameters related to their structure, material and operating conditions. This project will explore the use of non-rolling, laboratory-based methods, to allow tyre model parameterisation to be carried out quickly and accurately, using standard laboratory equipment. This will greatly enhance tyre simulation capacity in the automotive industry.
Research question: Is it possible, through the combination of non-rolling, lab-based test methods and artificial intelligence, to characterise a tyre in enough detail so that a representative mathematical model can be generated for use in vehicle simulations?
Novelty: This is the first attempt to directly relate non-rolling laboratory measurements of a tyre to its observed dynamic behaviour on-the-road, as part of a full vehicle. The approach will use a unique combination of state-of-the-art modal testing techniques, friction testing and neural-network-based artificial intelligence to characterise tyres quickly and accurately.
Objectives:
1) Develop a new efficient tyre modal testing procedure with focus on tyre structural parameters that affect its handling behaviour
2) Develop a new in-situ friction testing method
3) Develop a neural network able to predict tyre behaviour based on laboratory modal test and friction test data, as well as the measured force response for several tyres
4) Validate the method using real tyre data sets
Pertinent EPSRC research area: Engineering design, Artificial Intelligence Technologies
Scope and context of project:
The project will develop a new method for rapid, accurate and economical tyre model parameter identification. Mathematical tyre models are used in vehicle simulations for prediction of tyre forces. Vehicle design for passive safety and the design/calibration of active safety systems like (ABS) Anti-Lock Braking and ESP (Electronic Stability Programme), all depend on the ability to simulate tyre behaviour faithfully. All tyre models include several parameters related to their structure, material and operating conditions. This project will explore the use of non-rolling, laboratory-based methods, to allow tyre model parameterisation to be carried out quickly and accurately, using standard laboratory equipment. This will greatly enhance tyre simulation capacity in the automotive industry.
Research question: Is it possible, through the combination of non-rolling, lab-based test methods and artificial intelligence, to characterise a tyre in enough detail so that a representative mathematical model can be generated for use in vehicle simulations?
Novelty: This is the first attempt to directly relate non-rolling laboratory measurements of a tyre to its observed dynamic behaviour on-the-road, as part of a full vehicle. The approach will use a unique combination of state-of-the-art modal testing techniques, friction testing and neural-network-based artificial intelligence to characterise tyres quickly and accurately.
Objectives:
1) Develop a new efficient tyre modal testing procedure with focus on tyre structural parameters that affect its handling behaviour
2) Develop a new in-situ friction testing method
3) Develop a neural network able to predict tyre behaviour based on laboratory modal test and friction test data, as well as the measured force response for several tyres
4) Validate the method using real tyre data sets
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/S513672/1 | 01/10/2018 | 30/09/2023 | |||
| 2133144 | Studentship | EP/S513672/1 | 01/10/2018 | 20/06/2023 | Agate Utane |
| Description | This PhD project covers relatively simple tyre testing methods that can be used for tyre model parameterisation for vehicle handling simulations. So far, a three-dimensional experimental modal (vibration) analysis has been performed on tyres that allowed the estimation of modal parameters, such as the natural frequencies and the mode shapes. The dimensions of different sections of a tyre were used in a mechanical model of the tyre's sidewall to acquire radial and lateral stiffness properties. These data were used to parameterise a modified Brush model which is one of the simple physical models that can describe the characteristic forces and moments of a tyre as functions of slip. The model currently only covers the lateral dynamics. However, work is ongoing to extend the model for longitudinal and combined slip cases. Additionally, a new design of a friction rig has been developed. The rig's performance is yet to be examined. |
| Exploitation Route | The method developed in this project would allow for a relatively quick and inexpensive way to obtain approximate tyre characteristic curves which represent the tyre performance in vehicle handling applications. This could be applied both in industry and academia. |
| Sectors | Transport |