High-speed lateral stability analysis of articulated heavy vehicles using driver-in-the-loop real time simulation

Abstract

High-speed testing procedures for evaluating the lateral stability of multi-trailer articulated heavy vehicles (MTAHV) include the use of both an open-loop and a closed-loop maneuvers as specified in ISO-14791. The standard testing procedures are used to determine the Rearward Amplification (RA), which is a well-accepted performance measure for the lateral stability of MTAHVs. The open-loop testing maneuver includes the use of a single cycle sinewave steer input, while the closed-loop procedure prescribes a single lane-change path, following which a driver drives a testing vehicle. The closed-loop single lane-change maneuver can be performed by a human driver for in-vehicle tests. In numerical simulation, a driver model may be used to drive the virtual vehicle under the closed-loop maneuver for simplicity and repeatability. Very little attention has been paid to investigating into the interactions of driver-MTAHV. It has been reported that a driver is typically at fault in many heavy vehicle accidents, and many studies focus on improving the dynamic performance of the vehicle and ignore the driving skills of the driver. This thesis attempts to examine the driver-MTAHV interactions and quantify driver skills for controlling a MTAHV as compared to a single unit vehicle. This thesis proposes a method for driver skill analysis via comparing virtual drivers in various operating conditions. The driver skill analysis is conducted by evaluating the lateral stability and path-following ability of the vehicle under the control of the driver model. The evaluation is implemented using driver-in-the-loop (DIL) real-time simulations, where the RA measures derived from a group of human drivers are compared to those from the virtual driver. The numerical and DIL real-time simulation results demonstrate that the human and the virtual driver achieved similar a good agreement in terms of the performance measures, indicating the validity of the driver model, testing procedure and the interactions of driver-MTAHV.