Application of Learning Automata to Controller Design in Slow-Active Automobile Suspensions

This study considers a new design methodology in the context of active vehicle suspension control. The approach combines concepts from Stochastic Optimal Control with those of Learning Automata. A learning automaton effectively learns optimal control on-line in the vehicle, in an appropriate stochastic “test-track” environment. For practical application, the overwhelming advantage of this approach is that no explicit modelling is required, and considerable time savings may be expected in system development. This simulation study considers the on-line learning of optimal control in a low-bandwidth active suspension system, where control feedback is confined to a body-mounted accelerometer at each corner of the vehicle. It is shown that learning can successfully take place under a range of conditions, including the case when there is substantial transducer noise. The performance of the resulting control system is shown to depend heavily on the nature of the learning environment.     

Optimal Performance of Variable Component Suspensions

SUMMARY

Most vehicle suspension systems use fixed passive components that offer a compromise in performance between sprung mass isolation, suspension travel, and tireroad contact force. Recently, systems with discretely adjustable dampers and air springs been added to production vehicles. Active and semi-active damping concepts for vehicle suspensions have also been studied theoretically and with physical prototypes. This paper examines the optimal performance comparisons of variable component suspensions, including active damping and full-state feedback, for “quartercar” heave models. Two and three dimensional optimizations are computed using performance indicators to find the component parameters (control gains) that provide “optimal” performance for statistically described roadway inputs. The effects of performance weighting and feedback configuration are examined. Active damping is shown to be mainly important for vehicle isolation. A passive vehicle suspension can control suspension travel and tire contact force nearly as well as a full state feedback control strategy.     

DISCUSSION NOTE on Karnopp's article “Active Damping in Road Vehicle Suspension Systems”

SUMMARY

In this paper, an optimal suspension system is derived for a quarter-car model using multivariable integral control. The suspension system features two parts. The first part is an integral control acting on suspension deflection to ensure zero steady-sate offset due to body and maneuvering forces as well as road inputs. The second is a proportional control operating on the vehicle system states for vibration control and performance improvement. The optimal ride performance of the active suspensions based on linear full-state feedback control laws with and without integral control together with the performance of passive suspensions are compared.     

Optimal Linear Preview Control of Active Vehicle Suspension

SUMMARY

The problem of linear preview control of vehicle suspension is considered as a continuous time stochastic optimal control problem. In the proposed approach minimal a priori information about the road irregularities is assumed and measurement errors are taken into account. It is shown that estimation and control issues can be decoupled. The problem formulation and the analytical solution are given in a general form and hence they apply to other problems in which the system disturbances are unknown a priori, even in a stochastic sense, but some preview information is possible.

The solution is applied to a two-degree-of-freedom (2-DOF) vehicle model. The effects of preview information on ride comfort, road holding, working space of the suspension and power requirements are examined in time and frequency domains. The results show that the greatest potential is for improving road holding properties. This effect could not have been observed in previous studies based on a 1-DOF vehicle model. It is also demonstrated that the presence of preview drastically reduces power requirements, thus relieving the performance versus actuator power dilemma.     

Adaptive Control of Vehicle Suspension

SUMMARY

An adaptive control scheme for a two-degree-of-freedom vehicle model with active suspension is proposed. The performance goal is to minimize the variance of vehicle body acceleration under inequality constraints imposed on the variance of either tire or suspension deflection. An active suspension is adapted to the changes in vehicle velocity and the type of road (or terrain) surface which is assumed to be reconstructable from the accelerometer measurements. The control gain factors are obtained by the iterative method taking advantage of stochastic linear control theory. The performance of the system is evaluated and compared to that of an active system with constant gain factors and a passive system with adjustable parameters.     

Theoretical Limitations in Active Vehicle Suspensions

SUMMARY

Vehicle suspensions in which forces are generated in response to feedback signals by active elements obviously offer increased design flexibility compared to conventional suspensions using passive elements such as springs and dampers. It is often assumed that if practical difficulties are neglected, active systems could in principle produce arbitrary ideal, behavior. It is shown, using a simple linear two degree-of-freedom suspension system, model that even using complete state feed back and in the case of in which the system is controllable in the control theory sense, there still are limitations to suspension performance in the fully active case. If the ideal suspension performance is defined based on low-pass filtering of roadway unevenness inputs, an active suspension may not offer much better performance than a partially active or adaptive passive suspension depending upon the values of certain vehicle parameters.     

Active Control of Non-stationary Response of Vehicles with Nonlinear Suspensions

SUMMARY

Active control of non-stationary response of a single degree of freedom vehicle model with nonlinear passive suspension elements is considered in this paper. The method of equivalent linearization is used to derive the equivalent linear model and the optimal control laws are obtained by using stochastic optimal control theory based on full state information. Velocity squared quadratic damping and hysteresis type of stiffness nonlinearities are considered. The effect of the nonlinearities on the active system performance is studied. The performance of active suspensions with nonlinear passive elements is found to be superior to the corresponding passive suspension systems.     

Optimal Preview Control of Rear Suspension Using Nonlinear Neural Networks

SUMMARY

The performance of neural networks to be used for identification and optimal control of nonlinear vehicle suspensions is analyzed. It is shown that neuro-vehicle models can be efficiently trained to identify the dynamical characteristics of actual vehicle suspensions. After trained, this neuro-vehicle is used to train both front and rear suspension neuro-controllers under a nonlinear rear preview control scheme. To do that, a neuro-observer is trained to identify the inverse dynamics of the front suspension so that front road disturbances can be identified and used to improve the response of the rear suspension. The performance of the vehicle with neuro-control and with LQ control are compared.     

Adaptive Suspension Concepts for Road Vehicles

SUMMARY

Most vehicle suspensions are composed of passive spring and damper devices, although improved suspension performance is possible if an active system is used to control forces or relative velocities. The complexity, power requirements, and cost of fully active suspensions have restricted their use. Various partially active suspensions have been proposed and suspensions with slow load levelers and variable dampers are in widespread use. Here we analyze a class of basically passive suspensions the parameters of which can be varied actively in response to various measured signals on the vehicle. These suspensions can come close to optimal performance with simpler means than many of the active or semi-active schemes previously proposed.     

On Control Laws for Vehicle Suspensions Accounting for Input Correlations

SUMMARY

Input correlations involving time delays are common in active vehicle suspension system problems. One approach to control law derivation fur such systems is to restrict attention to slate feedback laws in the interests of practicality and it is then of interest to determine the law which is, in some sense, the best. Under assumptions which are common in this area. relating to input, system and cost Function forms, a new derivation of the expression for the cost, accounting for time delays, is given. The use of the expression in numerical procedures for determining effective control gains is discussed and an example for a half car planar vehicle model is described. By comparing results with existing ones which are truly optimal, an estimate is made of the loss of performance which results from the restriction of the control law form in this case. Some generalisation of the results is attempted and they are placed in a contemporary context at the conclusion.     

An Active Suspension with Optimal Linear State Feedback

SUMMARY

In this paper modern optimal control theory is applied to the design of an active suspension system for a motor vehicle. The road profile is assumed to be continuous and random with a power spectral density (P.S.D.) which varies inversely with the square of the frequency. The quadratic integral type performance index employed is a weighted sum of the integral squares of body acceleration, dynamic tyre deflection and relative body-to-axle displacement. A solution is obtained for the infinite time case which is both computationally and physically realizable as an active suspension in which the only continuous measurements required are the body absolute velocity and the body displacement relative to the road. The performance is compared with that of a conventional type passive suspension and found to be significantly better in practically all respects.     

The Response of Active and Semi-active Suspensions to Realistic Feedback Signals

SUMMARY

A simple vehicle model is presented incorporating passive, active, and semi-active suspensions. When the desired feedback variables are ideally available, the system response is well understood and excellent sprung mass isolation results. More often than not, the measured variables must be signal processed in some manner prior to their use in some control algorithm. This paper presents the expected response of a simple vehicle with an active and/or semi-active suspension, subject to non-ideal feedback information.     

DSP-based self-organising fuzzy controller for active suspension systems

This study proposed a self-organising fuzzy controller (SOFC) for controlling an active suspension system to evaluate its control performance. During the control process, the SOFC continually updated the learning strategy in the form of fuzzy rules. The fuzzy rule table of this SOFC could be initially set to zero. This not only overcame the difficulty in finding appropriate membership functions and control rules for designing a fuzzy controller, but also solved the database problem where the fuzzy rules of a fuzzy controller, once determined, remained fixed and could not suitably regulate them in real time to optimise the dynamic response of the system required to gain the desired control performance. To demonstrate the applicability of the proposed SOFC for active suspension systems, a quarter-car hydraulic-servo suspension system was designed and constructed to evaluate the feasibility of active suspension control. Additionally, to conform to real-time application requirements in the vehicular industry, the SOFC was implemented with a digital signal processor to control the hydraulic-servo suspension system so that the control performance could be determined. The SOFC has shown a better control performance in suppressing the vibration amplitude of the vehicle body for enhancing the structural safety of the vehicle and increasing the life of the suspension system. It also effectively suppressed the amplitude of the vehicle body acceleration and reduced the tire deflection for improving the ride and the handling quality of a vehicle better than a passive control, as verified in experimental results.     

Switched control of vehicle suspension based on motion-mode detection

This paper presents a study on switched control of vehicle suspension based on motion-mode detection. This control strategy can be potentially implemented via the interconnected suspension such as hydraulically interconnected suspension by actively switching its interconnection configuration in terms of the dominant vehicle body motion-mode. The design of the switched control law is developed focusing on three vehicle body motion-modes: bounce, pitch, and roll. At first, an H_∞ optimal controller will be designed for each motion-mode with the use of a common quadratic Lyapunov function, which guarantees the stability of the switched system under arbitrary switching functions. Then, a motion-mode detection method based on the calculation of the motion-mode energy is introduced. And then, the possible implementation of the control system in practice is discussed. Finally, numerical simulations are used to validate the proposed study.     

Advanced Control Methods of Active Suspension

SUMMARY

This paper describes new control methods for the active suspension. For improving ride comfort further, preview control rule is proposed. For improving stability further, roll stiffness distribution control rule is examined by the test vehicle. Simulations and vehicle driving tests are conducted to confirm the effect of these new control methods. The results of simulations and vehicle driving tests show in our research phase that preview control can achieve a substantial improvement in ride comfort and application of roll stiffness distribution control provides a large improvement in stability     

Performance of Low-bandwidth,Semi-active Damping Concepts for Suspension Control

Abstract

Active damping has been shown to offer increased suspension performance in terms of vehicle isolation, suspension packaging, and road-tire contact force. It can even approximate the performance of full state feedback control without requiring the difficult measurement of tire deflection. Many semi-active damping strategies have been introduced to approximate the response of active damping with the modulation of passive damping parameters. These strategies have typically required a relatively high bandwidth for actuator response. This paper investigates the simulation performance and “frequency response” of two concepts in low-bandwidth semi-active suspension control, one that sets a damping force directly and another that sets the damping resistance. The electronically controlled bandwidth of these actuators is approximately an order of magnitude less than other semi-active devices; high frequency control is handled mechanically. A quarter-car model is studied with the controlled damping replacing both passive and active damping of typical control schemes. Both low-bandwidth damping strategies perform remarkably well compared to both active and high-bandwidth, semi-active damping. In certain dynamic performances, the new semi-active strategies outperform active damping and what the author calls “nominal” semi-active damping.     

Optimal control of brakes and steering for autonomous collision avoidance using modified Hamiltonian algorithm

ABSTRACT

This paper considers the problem of collision avoidance for road vehicles, operating at the limits of friction. A two-level modelling and control methodology is proposed, with the upper level using a friction-limited particle model for motion planning, and the lower level using a nonlinear 3DOF model for optimal control allocation. Motion planning adopts a two-phase approach: the first phase is to avoid the obstacle, the second is to recover lane keeping with minimal additional lateral deviation. This methodology differs from the more standard approach of path-planning/path-following, as there is no explicit path reference used; the control reference is a target acceleration vector which simultaneously induces changes in direction and speed. The lower level control distributes vehicle targets to the brake and steer actuators via a new and efficient method, the Modified Hamiltonian Algorithm (MHA). MHA balances CG acceleration targets with yaw moment tracking to preserve lateral stability. A nonlinear 7DOF two-track vehicle model confirms the overall validity of this novel methodology for collision avoidance.     

Road roughness estimation based on discrete Kalman filter with unknown input

ABSTRACT

The road roughness acts as a disturbance input to the vehicle dynamics, and causes undesirable vibrations associated with the ride and handing characteristics. Furthermore, the accurate measurement of road roughness plays a key role in better understanding a vehicle dynamic behaviour and active suspension control systems. However, the direct measurement by laser profilometer or other distance sensors are not trivial due to technical and economic issues. This study proposes a new road roughness estimation method by using the discrete Kalman filter with unknown input (DKF-UI). This algorithm is built on a quarter-car model and uses the measurements of the wheel stroke (suspension deflection), and the acceleration of the sprung mass and unsprung mass. The estimation results are compared to the measurements by laser profilometer in-vehicle test.     

Coordinated control strategies for active steering,differential braking and active suspension for vehicle stability,handling and safety improvement

ABSTRACT

In this paper, a coordinated control strategy is proposed to provide an effective improvement in handling stability of the vehicle, safety, and comfortable ride for passengers. This control strategy is based on the coordination among active steering, differential braking, and active suspension systems. Two families of controllers are used for this purpose, which are the high order sliding mode and the backstepping controllers. The control strategy was tested on a full nonlinear vehicle model in the environment of MATLAB/Simulink. Rollover avoidance and yaw stability control constraints have been considered. The control system mainly focuses on yaw stability control. When rollover risk is detected, the proposed strategy controls the roll dynamics to decrease rollover propensity. Simulation results for two different critical driving scenarios, the first one is a double lane change and the other one is a J-turn manoeuvre, show the effectiveness of the coordination strategy in stabilising the vehicle, enhancing handling and reducing rollover propensity.     

Semi-Active Attitude and Vibration Control

SUMMARY

The paper discusses the attitude and vibration control of a passenger car on the basis of a full vehicle model. The analysis presented consists of two parts: (I) The introduction of a newly developed semi-active anti-roll/pitch system, (ii) An example of an actively suspended full vehicle model using a simple control strategy to improve ride comfort. The attitude control using semi-actively generated compensation forces prevents the car from rolling in curves and pitching during braking or accelerating. The strength of the system is the small energy consumption. The performance of the combination of both attitude and vibration control can compete with a fully active suspension system.