Search Results

This paper gives an overview of the development of a number of loss models for the pushing metal V-belt CVT. These were validated using a range of experimental data collected from two test rigs. There are several contributions to the torque losses and new models have been developed that are based upon relative motion between belt components and pulley deflections. Belt slip models will be proposed based upon published theory, expanded to take account of new findings from this work. The paper introduces a number of proposals to improve the efficiency of the transmission based on redesign of the belt geometry and other techniques to reduce frictional losses between components. These proposed efficiency improvements have been modelled and substituted into a complete vehicle simulation to show improvements in vehicle fuel economy over a standard European drive cycle.

Recently, the development of braking assistance system has largely benefit the safety of both driver and pedestrians. A robust prediction and detection of driver braking intention will enable driving assistance system response to traffic situation correctly and improve the driving experience of intelligent vehicles. In this paper, two types unsupervised clustering methods are used to build a driver braking intention predictor. Unsupervised machine learning algorithms has been widely used in clustering and pattern mining in previous researches. The proposed unsupervised learning algorithms can accurately recognize the braking maneuver based on vehicle data captured with CAN bus. The braking maneuver along with other driving maneuvers such as normal driving will be clustered and the results from different algorithms which are K-means and Gaussian mixture model (GMM) will be compared.

The flow field and body aerodynamic loads on the DrivAer reference model have been extensively investigated since its introduction in 2012. However, there is a relative lack of information relating to the models wake development resulting from the different rear-body configurations, particularly in the far-field. Given current interest in the aerodynamic interaction between two or more vehicles, the results from a preliminary CFD study are presented to address the development of the wake from the Fastback, Notchback, and Estateback DrivAer configurations. The primary focus is on the differences in the far-field wake and simulations are assessed in the range up to three vehicle lengths downstream, at Reynolds and Mach numbers of 5.2×106 and 0.13, respectively. Wake development is modelled using the results from a Reynolds-Averaged Navier-Stokes (RANS) simulation within a computational mesh having nominally 1.0×107 cells.

The motion of a vehicle along a desired path is possible due to steering action of the driver. Hence, vehicle dynamics and control simulations should take into consideration the action of the driver. This work presents a preview based vehicle steering controller using Neural Networks which can be used in the vehicle lateral dynamics simulations. The training data for the Neural Network is being obtained using a steering controller from the existing literature and its gains are determined using Optimization. Three different architectures are being designed and conclusions are presented. These Neural Network models are validated by testing against real track data.

Transient force and surface pressure data has been measured on a range of simple geometric shapes in order to gain an understanding of the complex time dependent and separated flow around a vehicle when subjected to a crosswind. The experiments were carried out using the Cranfield University model crosswind facility. It is found that the leeward face is the dominant area of transient activity. Maximum and minimum peak yawing moments at gust entry and exit are compared