Abstract:
Tyre parameterisation is fundamental to vehicle dynamics research, as tyre properties have a significant impact on the behaviour of a vehicle. One way in which some of the primary tyre parameters can be extracted is a trailer-based tyre tester, multiple versions of which exist at the CSIR. Fundamental to this type of testing is the accurate measurement of the wheel slip angle. In the case of trailer-based tyre-testing, this is measured relative to the trailer body, and so accurate measurement of the trailer side-slip angle is critical. This may be carried out using a suitable sensor such as the VBox 3, OxTs RT3000, or Corrsys Datron Correvit sensor. The latter is used in the application by the CSIR. Each of these sensor options have drawbacks in terms of cost, size and measurement limitations, which have prompted investigations into alternative side slip angle sensing solutions. In this paper we present the implementation of a camera-based system, using open source software libraries, on the CSIR’s medium tyre-tester. The system comprises a single camera mounted to the trailer and observing the ground, and a processing unit. The software was developed in Python, making use of Lucas-Kenade feature tracking, an outlier rejection scheme, and simple filtering. The system was evaluated in a number of tests and has demonstrated good potential for application to future tyre testing research. The system was further evaluated in limited tests involving the CSIR’s Stress-In-Motion technology for possible additional future research functionality.
Description:
Papers presented at the 38th International Southern African Transport Conference on "Disruptive transport technologies - is South and Southern Africa ready?" held at CSIR International Convention Centre, Pretoria, South Africa on 8th to 11th July 2019.
