Daeyi Jung

This paper presents a novel fault estimation strategy for electric motors in Steer-By-Wire (SBW) systems, based on a Semi-Supervisory Artificial Neural Network (SSANN) observer. Our previous work [22] explored a purely supervisory ANN-based approach (SANN), which demonstrated significantly improved estimation performance over traditional model-based methods under various fault scenarios. However, this approach showed limitations in accurately tracking high-frequency faults and adapting to unexpected perturbations throughout the estimation process. To address these challenges, the proposed SSANN architecture is introduced to further enhance estimation performance and ensure robustness across a broader range of fault scenarios, including chirp-like and step-like signals. The SSANN employs a hybrid adaptive structure, where the network is initialized with the pre-trained weights and biases from SANN, serving as a stable baseline. This baseline is then complemented by an additional set of online-adjustable weights and biases, which are updated during operation using a newly designed loss function tailored to motor dynamics in SBW systems. Unlike conventional semi-supervisory ANN approaches, the proposed SSANN introduces adaptive parameters that are decoupled from the baseline supervisory model, ensuring that online learning does not destabilize the core estimation framework. This design enables the observer to maintain baseline stability while effectively adapting to unanticipated changes in the fault signal, resulting in robust and accurate estimation performance. The proposed method has been rigorously validated through Hardware-in-the-Loop Simulation (HILS) and testing on a compact simulator with human participants. Experimental results confirm that SSANN significantly outperforms the original SANN approach, reducing final cumulative estimation error by 25–41%. Furthermore, SSANN successfully captures fault signals with higher frequency components, which were previously difficult to estimate using purely supervisory methods. It is also worth emphasizing that the SSANN framework is general in its formulation and can be applied to a wide range of nonlinear estimation and control problems with minimal structural modifications, making it a versatile and practical solution for real-world applications.

Various model-based mass estimation approaches have been discussed for a long time. However, estimation performance often deteriorates in some driving situations and, in particular, slow convergence and excessive overshoot of estimates are a major issue for model-based approaches. Meanwhile, mass estimation approaches using ANN models have recently emerged to propose better solutions, but their usefulness has not been fully investigated. Therefore, this paper presents a vehicle mass estimation strategy using practical supervisory artificial neural networks to achieve more accurate results with better convergence. Here, the perturbed engine torque and vehicle longitudinal acceleration are selected as the inputs of the ANN (instead of the original engine torque and vehicle acceleration), which allows for the faster convergence of estimates with high accuracy; these inputs are existing sensor values already available in the vehicle system. The effectiveness of the proposed ANN approach was verified using simulation and software-in-the-loop simulation (SILS) with field test data, and it was found that the convergence speed of the proposed ANN is almost twice as fast as that of the model-based approach, the accuracy is much better, and the estimation quality is constantly stable without any excessive transient responses. This study will provide further insights into mass estimation using the ANN approach.

Recently, to produce a large steering torque or ensure fault-tolerant control (FTC), dual motorized actuators for a steering system in automobiles have been emerged. In particular, the commercial vehicle requires an electrical-powered multiple motor based steering system generating large steering torque, which can replace the existing hydraulic steering one. However, the control of such system become more complicate than controlling of a single actuator and must deal with the unavoidable non-synchronous nature of multiple motors. In line with the trends and issues, many automobile manufacturers are currently paying attention to the robust and efficient motion controls of a dual-motor driving steer-by-wire (DSBW). Therefore, we proposed a novel synchronous motion control system of DSBW using passive decomposition and sliding mode control (SMC) techniques guaranteeing the passivity and robustness of the overall synchronized motion. Furthermore, the performance of proposed control system is validated through HILS of an actual DSBW using a developed vehicle model and an optimal finite preview control (OFPC) scheme tracking a specific test scenario, Double Lane Change (DLC). It is found that the motor-to-motor synchronization error and tracking one for the target reference trajectory are less than 1 degree, which sufficiently meets the requirement of automobiles manufacture company and is comparable to other existing approaches. We also investigated the tracking performance of the control system based on the different steering characteristics of OFPCs generated by the several preview times. Moreover, we investigated the tolerance level of proposed technique (without any additional FTC) against a partial fault condition in one of the motors and compared it with other control schemes in terms of tracking performance. This work will be valuable asset for those who wish to design the robust synchronous motion control of DSBW and construct the corresponding HILS of system considering different steering characteristics.

Recently, with the trend of redundancy design, the importance of synchronous motion control of multiple motors has been emphasized in various fields such as automotive, construction, and industrial engineering. Therefore, this paper proposed a novel passive decomposition-based robust synchronous control strategy for a multi-motor system, guaranteeing that both the tracking error of each motor and the synchronous error between motors are ultimately and synchronously bounded, even under the presence of parametric uncertainty and unstructured external disturbance. Specifically, a passive decomposition is used to obtain the locked and shape systems from the original system, and then a sliding mode control system along with robust compensations is designed for each decomposed system to achieve the precise synchronous motion control of the n number of motors. Here, the controller for the locked system reduces the tracking errors of motors for a given desired trajectory, while the controller for the shaped system decreases the synchronous errors between motors. Furthermore, the control system is generally and conveniently formulated to adopt the arbitrary n number of motors that must track a given desired trajectory and be synchronized. Compared to other related studies, this work especially focused on increasing the robustness of the entire system using both high-order sliding mode control and two separate compensation terms for model uncertainty and unstructured external disturbance. Finally, to validate the effectiveness of the proposed synchronous control strategy, the extensive experimental studies on two/three/four-geared BLDC motors with a high dead-zone effect were conducted, and we also compared the synchronous control performance of the proposed control strategy with the other representative control approaches, a master–slave control scheme and an independent one to address the superiority of the proposed control system. Regardless of the number of motors, due to the robustness of the control system, it is found that the proposed control ensures the tracking and synchronous errors are less than 1 degree for the sine-wave trajectory while it guarantees that the errors are below 1.5 degree for the trapezoidal trajectory. This control approach can be widely and generally applied to the multiple motor control required in various engineering fields.

Our previous study focused on the development of ANN classifier estimating the maximum static road friction coefficient for each wheel of the vehicle based on the Carsim/Simulink co-simulation environment. As follow-up to the previous study, this paper investigated the effectiveness of the proposed artificial neural network (ANN) classifier using the field test data in Software-In-Loop-Simulation (SILS) environment. Furthermore, we extended the input of ANN classifier to improve an estimation performance. Specifically, the braking pressure and the pressure gradient of each wheel were added to the original input of the classifier, the initial speed at braking, the vehicle deceleration, the wheel slip ratio, and the vehicle mass. Hence, the benefits of additional inputs have been clearly explored here. Moreover, the proposed scheme was challenged to other, more complicate road surface conditions (including jump and split friction roads). The original classifier guarantees a fairly accurate estimate, but we found that including brake pressure information into the classifier yielded estimates in better quality and the estimation results capture 86~95% accuracy for normal braking regardless of roads and 70~84% accuracy for extreme roads. This work will be a valuable asset for those who wish to develop the practical estimation methods for road friction coefficient via an ANN classifier using only inertial sensor information already available in most standard cars.

In this paper, we present the fault estimation of a motor in rack-type electrical power steering (R-EPS) system using an artificial neural network (ANN) observer and the comparison study of estimation performance between ANN observer and model-based ones. For various amplitudes and frequencies of fault, it is not easy to obtain the accurate fault estimation using the model-based observers inherently possessing the accumulated errors and disturbance effect. Such model-based methods often result in undesired consequences; hence, this study employed the “model-free” ANN observer, without using any dynamics and parameters of a motor, to accomplish the better outcomes. Furthermore, the advantages of ANN observer for the fault estimation of the motor have been clearly investigated under several control/fault scenarios, and the effectiveness of proposed work has been validated through an actual experimental study. It is found that the performance of model-based approaches is degraded when the frequencies and amplitudes of fault and control scenarios are changed, but the ANN observer guaranteed performance that was almost the same regardless of fault and control scenarios. Notably, ANN observers showed 84% to 95% of estimation accuracy with almost no delay between estimates and actual faults while model-based approaches did 68% to 86% accuracy along with noticeable delay.

The wind suddenly blown to the side of the car (i.e, cross-wind effect) is considered as one of the major lateral disturbances, which causes the unstable motion of the vehicle and the persistent driving fatigue for the driver who tracks the desired travel path. In particular, a commercial vehicle having a large side area is greatly affected by this effect. Therefore, many related automotive/car manufacturers still wish to equip their Advanced Driving Assistant System (ADAS) with the crosswind disturbance compensation control system. Meanwhile, in recent advanced vehicle systems, a rack-type motor-driven power steering (R-MDPS) system is more widely used than a column-type MDPS (C-MDPS) due to the structure advantage and the effective steering assistance. Recognizing those two issues, this paper investigates a novel anti-control and estimation strategy of crosswind disturbance for the R-MDPS system of vehicles with non-negligible side surfaces. Specifically, an adaptive disturbance observer (D.O.B) has been proposed to estimate the crosswind effect. Furthermore, using optimal control theory, the compensation control system is designed to assist the driver in two possible situations. One is for when the driver continues to steer the vehicle under the effect of the crosswind, and the other is for when the driver temporarily loses steering control due to the effect. In addition, the control mode selection conditions between two controls are clearly presented to maximize the efficiency and performance of the proposed control system. Finally, the effectiveness of the proposed control system has been evaluated based on the Simulink/Carsim Co-simulation and HILS environments.