In this paper, a robust adaptive sliding mode (ASM) controller for improving vehicle maneuverability and lateral stability of steer-by-wire (SBW) vehicles. The proposed stability control algorithm is composed of the following three remarkable characteristics: 1) As the vehicle sideslip angle which is very useful in vehicle stability control is often unavailable or unmeasurable for practical applications, a sliding mode-based state observer is proposed to online estimate the actual sideslip angle via the yaw rate and the lateral acceleration measurements. 2) ASM lateral stability controller, as an upper level controller, is designed to calculate the corrected steering angle for driving both the yaw rate and the sideslip angle to achieve their desired values in the presence of modelling uncertainties and external disturbances. Moreover, an adaptive law is incorporated in the control law to estimate the switching gain such that the complex uncertainty bound information can be avoided. 3) The desired steering angle is realized by a lower steering controller via ASM for an SBW system. The hardware-in-the-loop simulation results demonstrate the excellent stability control performance of the proposed control for different steering maneuvers.

In this article, a novel active front steering (AFS) control strategy including the upper controller and the lower controller is proposed to improve the yaw stability and maneuverability for steer-by-wire (SbW) vehicles. The adaptive recursive integral terminal sliding mode (ARITSM) control is adopted in the upper controller for guaranteeing the convergence performance of both the actual sideslip angle and the yaw rate with strong robustness and fast convergence rate. Then, a fast nonsingular terminal sliding mode (FNTSM) control with extreme learning machine (ELM) estimator to estimate its equivalent control is designed in the lower controller to track the desired front wheel steering angle calculated from the upper controller for driving the sideslip angle and the yaw rate to converge ideal value. It is shown that the upper controller takes two controlled variables (vehicle sideslip angle and yaw rate) and only one control input (front steering angle) into consideration, which can obtain a better performance compared with the case of using only one of these values. Since using the ELM technique in the lower controller to estimate the equivalent control of the FNTSM, not only the dependence of SbW system dynamics can be alleviated in the process of designing controller but also the excellent steering control performance can be achieved. Comparative simulations are carried out by utilizing Carsim and Matlab software to validate the excellent performance of the proposed control strategy for different steering maneuvers.

This paper deals with a bond-graph-based fault detection, isolation and estimation, which is applied to a nonlinear electromechanical system, in the presence of parametric faults and nonparametric faults. The concept of extended analytical redundancy relations (EARR) is developed where the nonparametric faults of both multiplicative and additive natures can be considered and distinguished. The integration of dependent EARR with independent EARR leads to a more efficient fault isolation where the number of potential faults could be decreased. For the purpose of fault estimation, a new pitch adjustment biogeography-based optimization (PABBO) is developed where the pitch adjustment in harmony search is embedded in BBO mutation stage to enhance the search ability of the algorithm. The proposed methodologies are validated by simulation and experiment results.

This paper proposes a real-time model-based health monitoring method for a nonlinear mechatronic system with multiple faults in both parametric and nonparametric components. A nonlinear bond graph model incorporating the influence of Stribeck friction is established to capture the dynamic behavior of the monitored mechatronic system. Based on the model, fault diagnosis is carried out via the combinative fault signature matrix which is built from independent and dependent analytical redundancy relations to enhance the isolation ability of the monitored system under multiple-fault condition. After that, an adaptive genetic algorithm-particle filter (AGA-PF) is developed for fault parameter estimation and remaining useful life prediction. The AGA-PF can mitigate the sample impoverishment problem in generic particle filter through genetic operators with adaptive mutation probability according to the fitness of the particles. The effectiveness of the proposed method is verified through simulation and experiment investigations on the nonlinear mechatronic system.

Prognosis of discrete component with intermittent fault in hybrid systems is challenging since the component has only two states (i.e., ON and OFF) and no associated physical parameter in the model can quantify the degradation. This article aims to solve the discrete component prognosis problem under the model-based paradigm. First, the fault detection and isolation module help find the possible faulty discrete components. Based on the isolated possible faulty discrete components, Levy flight biogeography-based optimization is proposed to identify the faulty discrete component states, as well as the fault appearing and fault disappearing instants. Second, a Weibull function-based degradation model which can capture the duration evolution of intermittent fault of discrete component in observation window (OW) is developed using coordinate reconstruction approach, and the degradation model coefficients can be calculated from the fault identification results. After that, the concept of failure threshold for faulty discrete component is defined based on the ratio of fault duration to OW, which enables the prognosis of intermittent fault in discrete component. Finally, the proposed methodologies are validated by experiment results. Note to Practitioners —This article is motivated by the intermittent fault prognosis problem of discrete components (e.g., relays and hydraulic valves) in hybrid systems. Existing fault prognosis researches do not consider discrete component which is an important part of hybrid systems. For the intermittent fault prognosis of discrete component, the observation window (OW) concept and coordinate reconstruction (CR) method are proposed to establish the degradation model, and the ratio of fault duration to OW is used to define the failure threshold of discrete component. To show the effectiveness of the proposed methods, an application on a hybrid circuit system is considered. It is noted that the degradation pattern (e.g., increase of frequency or duration of intermittent fault) of discrete components may vary in different systems, while the degradation process can be quantified by the OW and CR methods developed in this article, which enables the prognosis of intermittent fault in discrete component for various hybrid industrial systems. The proposed approach can be applied to industrial hybrid systems if the following conditions are satisfied: 1) the hybrid bond graph model of the monitored system can be established, based on which the fault detection and isolation can be implemented and 2) the monitored system contains multiple discrete components suffering from intermittent faults whose appearing and disappearing instants can be identified by certain method.

For autonomous vehicles, obstacle detection results using 3D lidar are in the form of point clouds, and are unevenly distributed in space. Clustering is a common means for point cloud processing; however, improper selection of clustering thresholds can lead to under-segmentation or over-segmentation of point clouds, resulting in false detection or missed detection of obstacles. In order to solve these problems, a new obstacle detection method was required. Firstly, we applied a distance-based filter and a ground segmentation algorithm, to pre-process the original 3D point cloud. Secondly, we proposed an adaptive neighborhood search radius clustering algorithm, based on the analysis of the relationship between the clustering radius and point cloud spatial distribution, adopting the point cloud pitch angle and the horizontal angle resolution of the lidar, to determine the clustering threshold. Finally, an autonomous vehicle platform and the offline autonomous driving KITTI dataset were used to conduct multi-scene comparative experiments between the proposed method and a Euclidean clustering method. The multi-scene real vehicle experimental results showed that our method improved clustering accuracy by 6.94%, and the KITTI dataset experimental results showed that the F1 score increased by 0.0629.