LPV Static Output Feedback for Constrained Direct Tilt Control of Narrow Tilting Vehicles

Abstract

This paper presents a new direct tilt control (DTC) design to improve the lateral stability and the driving comfort of narrow tilting vehicles. To this end, a conceptual model is constructed from the vehicle dynamics, a simplified model of the driving environment, and the vehicle perceived acceleration. This latter is considered as the main performance variable of the related H2 control problem. The conceptual model is then represented in a polytopic linear parameter varying (LPV) form for control purposes. To avoid the use of costly vehicle sensors while favoring the simplest control structure for real-time implementation, a new LPV static output feedback (SOF) control method is proposed. Thanks to Lyapunov stability arguments, physical constraints on both system states and DTC actuator are explicitly considered in the design procedure to improve the safety and the comfort of passengers. Moreover, a parameter-dependent Lyapunov function is exploited for theoretical developments. In this way, the finite bounds on vehicle speed and acceleration are effectively taken into account in the control design to reduce the conservatism. The H2 control design is recast as an LMI-based optimization which can be easily solved with numerical solvers. The resulting robust DTC controller is evaluated with realistic driving scenarios.

Linear System Identification Versus Physical Modeling of Lateral–Longitudinal Vehicle Dynamics

“…To avoid local minima of bad quality, each identification experiment was repeated 1000 times with different starting values within the constraint set Θ, except for the data-driven models where the parameters are estimated in a single instance of optimisation. The optimal vector of parameters amongst the 1000 results was chosen as the one with lowest cost function (8). Despite the 1000 starts it is not possible to guarantee that the chosen parameters are indeed the true vehicle parameters, nor the global optimum to our optimisation problem.…”

“…The last two decades have seen a steep increase in research efforts to deploy (semi) autonomous ground vehicles both for private and commercial use [1]- [3]. An important part of these efforts has focused on devising mathematical models of vehicle dynamics for the purpose of control design [4]- [8]. Dynamic models for control design are often permitted to be relatively simple approximations because feedback control methods are highly tolerant of model error -a key advantage of using feedback.…”

Robust LFT-LPV H∞ Control of an Underactuated Inverted Pendulum on a Cart with Optimal Weighting Functions Selection by GA and ES

“…The LPV techniques are commonly used in sensible systems that need rigorous rejection of disturbances, uncertainties' effects, and faults like unmanned aerial vehicles (López-Estrada et al, 2016;Hasseni and Abdou, 2018). The authors of (Nguyen et al, 2020) constructed a conceptual model of narrow tilting vehicles in a polytopic LPV form, and they then proposed a static output feedback control method as the simplest structure to avoid the use of costly vehicle sensors. In (Xu et al, 2019), a novel robust fault detection approach is proposed.…”

Research on Model Predictive Control for Automobile Active Tilt Based on Active Suspension

“…An H 2 control design based on a linear matrix inequality based optimization was built. The resulting robust DTC controller was evaluated with realistic driving scenarios [12].…”