On Robust Steering Based Lateral Control of Longer and Heavier Commercial Vehicles

Doctoral thesis, 2018

Rapid growth in the transportation of goods has led to raised concerns about
environmental effects, road freight traffic, and increased infrastructure usage.
The increasing cost of fuel, and issues with congestions and gas emissions,
make longer and heavier commercial vehicles (LHCVs) an attractive alter-
native to conventional heavy vehicles. However, one major issue concerning
LHCVs is their potential impact on traffic safety. A typically dangerous be-
haviour happens during evasive steering maneuvers, which causes amplified
lateral motions in the towed units. These amplified motions can lead to the
towed units’ oscillation, large offtracking and, in a worst case scenario, cause
rollover.
The main objective of this thesis is to develop robust steering-based con-
trollers for improving the lateral performance of LHCVs at high speeds by
suppressing unwanted amplified motions in the towed units. Robust control
methods aim to achieve an adequate level of robustness against model un-
certainties and disturbances, while at the same time satisfying the desired
closed-loop system performance specifications. The proposed robust control
syntheses are formulated based on an H ∞ static output-feedback (SOFB)
in which only one easily measurable state variable is required. The control
synthesis problems are solved by using linear matrix inequality (LMI) op-
timizations. As the measurement of the driver steering input is available,
a combined version of SOFB and dynamic feed-forward (DFF) is also de-
veloped and several techniques for designing DFF are proposed. The theo-
retical contributions of this research mainly lie in the derivation of a novel
LMI conditions for integral quadratic constraints on the states and also in
the derivation of a set of new LMI conditions for the DFF design method.
From a practical point of view, the proposed controllers are simple and easy
to implement, despite their theoretical complexity.
The effectiveness of the designed controllers is verified through numerical
simulations performed on linear vehicle models as well as high-fidelity ve-
hicle models. The verification results confirm a significant reduction in yaw
rate rearward amplification, lateral acceleration rearward amplification and
high-speed transient off-tracking, thereby improving the lateral stability and
performance of the studied LHCVs.