Rollover warning system for tractor-semitrailer using a modified odenthal rollover index algorithm

• Main Author: Harun, Mohamad Hafiz
• Format: Thesis
• Language: English
• Published: 2021
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://eprints.utm.my/id/eprint/101816/1/MohamadHafizHarunPSKM2021.pdf; http://eprints.utm.my/id/eprint/101816/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:145012

Any accident involving a tractor-semitrailer could significantly affect life and component damage as well as the surrounding environment due to the size of the vehicle. One of the main factors that causes tractor-semitrailer accidents is vehicle rollover instability. Therefore, this study aimed to develop a vehicle instability avoidance system for a tractor-semitrailer by implementing it on an accurate tractorsemitrailer model. In developing the model, a new approach was proposed by adopting a virtual Pacejka tire model in modelling the hitch joint of the tractor-semitrailer. The virtual Pacejka tire model has included a 16 degree-of-freedom tractor-semitrailer within MATLAB/Simulink software and later verified using the TruckSim model and validated with published data. It is observed from the verification and validation results, the tractor-semitrailer model using the virtual Pacejka tire model for the hitch joint showed a similar response to the behaviour of the TruckSim model and published data. In terms of vehicle instability avoidance system, the fastest response of the tractor-semitrailer rollover index based on early warning indicator was selected by utilizing several types of rollover index algorithm proposed by the previous researchers. The step steering manoeuvres simulation at a various speed was conducted using MATLAB/Simulink software to obtain the rollover index. It can be observed from the results that the rollover index algorithm proposed by Odenthal provides the fastest index based on the early warning indicator on the tractor unit. In order to optimise the rollover index performance, the Odenthal rollover index algorithm was modified and optimised using Particle Swarm Optimisation (PSO). Finally, the rollover index algorithm was proposed by integrating the modified Odenthal rollover index algorithm with driver steering and vehicle speed inputs instead of lateral acceleration. The modified Odenthal rollover index algorithm performance was evaluated by conducting an experiment involving the step steering manoeuvres, subjected to various vehicle speeds and load conditions through the Hardware-in-the- Loop (HIL) simulation in the TruckSim driving simulator and MATLAB/Simulink software. It was observed from the experimental results that the modified Odenthal rollover index algorithm produced 12.4% faster Time-To-Warn (TTW) than the Odenthal rollover index for the driver. Thus, the modified Odenthal rollover index algorithm demonstrated a better early warning system for the driver to initiate the corrective action.