Robust H ∞ H_{\infty} Position Controller for Steering Systems
Tushar Chugh, Fredrik Bruzelius, Balázs Kulcsár · 2024
Read Robust H ∞ H_{\infty} Position Controller for Steering Systems by Tushar Chugh, Fredrik Bruzelius, Balázs Kulcsár (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-19161.txt.
Introduction
Tushar Chugh, Fredrik Bruzelius and Balázs Kulcsár wrote a 2024 paper on a robust position controller for electric steering systems. Steering feel, driver aids and self-driving all need exact position control. The driver's arm adds shifting inertia when hands grip the wheel. Single-signal controllers lose grip and tracking under that sway.
Their design also feeds the sensed driver torque into the controller. Solving a robust optimization shapes the loop gain as wanted. Low frequencies get strong gain for accuracy; high ones stay tame for safety. Bench tests on real hardware showed better tracking and sturdiness than rival controllers.
Catalog record
| CREATOR | Tushar Chugh, Fredrik Bruzelius, Balázs Kulcsár |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-19161
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Robust H ∞ H_{\infty} Position Controller for Steering Systems Tushar Chugh, Fredrik Bruzelius, Balázs Kulcsár Abstract This paper presents a robust position controller for electric power assisted steering and steer-by-wire force-feedback systems. A position controller is required in steering systems for haptic feedback control, advanced driver assistance systems and automated driving. However, the driver’s physical arm impedance causes an inertial uncertainty during coupling. Consequently, a typical position controller, i.e., based on single variable, becomes less robust and suffers tracking performance loss. Therefore, a robust position controller is investigated. The proposed solution is based on the multi-variable concept such that the sensed driver torque signal is also included in the position controller. The subsequent solution is obtained by solving the LMI − H ∞ -H_{\infty} optimization problem. As a result, the desired loop gain shape is achieved, i.e., large gain at low frequencies for performance and small gain at high frequencies for robustness. Finally, frequency response comparison of different position controllers on real hardware is presented. Experiments and simulation results clearly illustrate the improvements in reference tracking and robustness with the proposed H ∞ H_{\infty} controller. I Introduction With the advancement in driving automation technology, the transportation sector aims to realize semi-automated and automated vehicles. In the context of driving autonomy, there exists different levels of automation systems, depending on the undertaken dynamic driving task , see [ 1 ] for definitions. At operational level, to perform these tasks, both longitudinal and lateral vehicle motion should be controlled with a certain performance and safety guarantees [ 2 , 3 ] . For lateral control, in the higher levels of driving automation especially in level 4 and level 5 , the primary actuation must be done by the steering system to maintain the required trajectory. This task is executed by a steering position controller. Typically, it has a hierarchical structure: a higher level and a lower level control. The former computes an optimal reference trajectory, representing the vehicle motion control application. Whereas, the tracking is achieved with the latter by minimizing the error between the reference and actual position signals. For the higher level control, there has been an extensive research for more than a decade. For example using classical feedback-feedforward approach in [ 4 ] , linear model predictive control (MPC) in [ 3 ] , piecewise linear MPC for the non-linear vehicle-tire interaction in [ 5 ] , non-linear MPC method in [ 6 , 7 ] to name a few. Most of them have reasonably assumed an ideal lower level control; however, we focus on that in this paper. The higher level control requests a reference signal which can be either steering (or pinion/wheel) angle or rack position [ 8 , 9 , 10 ] . In robotics, it is also referred to as admittance control, see e.g. [ 11 , 12 , 13 ] . Regardless of the system dynamics, a position feedback controller minimizes the error in one of the interaction port’s motion trajectory variables, i.e., in terms of angular position and/or velocity. From the lateral vehicle control aspect, the maximum relevant frequency for periodic steering inputs is between 4 - 5 Hz [ 14 ] , assuming no environmental disturbance and irrespective of the steering action performed by the driver or an advanced driver assistance function. This can be corroborated from the test scenarios considered in the higher level control design, see [ 3 , 4 , 6 , 7 , 5 ] . Despite the fact, that most of the existing steering position controllers, for example in [ 8 , 9 , 10 ] , would likely provide a good lateral vehicle control performance, the question is why do we seek another solution. The reason is coupling port’s uncertainty that affects the nominal tracking. In an electric power assisted steering (EPAS), there are two interaction ports: (a) the steering wheel responsible for the driver interaction; and (b) the steering rack responsible for the environment interaction, see Fig. 1 (a). Although the driver is not required in level 5 driving automation, but it is still a fail-safe fall back solution up to level 4 . As a result, the driver coupling could be a source of parametric uncertainty due to its own physical impedance, especially the coupled arm inertia. Whereas, an unknown tire-road interaction at the environment port (i.e. defined in terms of steering rack force) can be considered as a disturbance [ 15 ] . For this, another possibility is to include a vehicle model at the environment port, thereby extending the state variables with: vehicle sideslip, yaw rate, and lateral tire slip angles, see e.g. [ 16 , 17 ] . In this approach, the vehicle and tire parameters, such as the peak tire-road friction coefficient, are assumed to be known; the model equations are also linearized with a limited tire model fidelity. Hence, this results in a real parametric uncertainty, such that the rack force becomes a function of vehicle speed and other parameters. Therefore, [ 16 , 17 ] have proposed a gain scheduled feedback controller based on some of the tire parameters and vehicle speed. To avoid these dependencies and for simplicity, we design a robust position controller regardless of the environmental interaction. The steering feedback control is an important objective when the driver is in the loop. Hence, our goal is to use the same position controller for generating the required haptic feeling for consistency, unlike switching between the different controllers [ 18 ] . Consequently, an outer loop with a position reference generator is necessary for the closed-loop interaction, see [ 19 ] . From the steering feedback perspective, the driver-environment transparency must be realized up to 15 Hz, and so must the reference tracking performance [ 20 ] . For a typical steer-by-wire (SbW) system in Fig. 1 (b), there are two actuators: (a) force-feedback (FFb) with the driver interaction port and (b) road wheel actuator with the environment interaction port. They are responsible for the steering feedback control and the vehicle lateral control objectives, respectively. A state-of-the-art road wheel actuator is based on position control such that the actual FFb angle defines the rack position reference. The road wheel actuator is disregarded in this paper because the environment port is mechanically disconnected from FFb and we assume no direct implication of this port’s uncertainty on FFb [ 21 , 22 , 23 ] . However, the SbW-FFb system has the same tracking performance requirement as EPAS. Fig. 1: A typical illustration of: (a) an electric power assisted steering and (b) a steer-by-wire system. In the literature, there are numerous examples of classical single variable position controllers, e.g. [ 24 , 25 , 26 , 10 ] . They have limited robustness and a depreciating reference tracking performance during a port’s uncertainty caused by the coupling impedance [ 20 ] . Hence, there exists a research gap when it comes to a robust position controller for haptic applications. On the contrary, there are available torque/force controllers, for example [ 27 , 26 , 15 ] , where the feedback control principle is inherently robust against the coupling uncertainties due to its causality [ 20 ] . Moreover, a