During high-speed platooning, if the leading vehicle experiences sudden failure (e.g., tire blowout or braking loss), the following vehicles must execute simultaneous deceleration and obstacle avoidance within a minimal timeframe. This process induces significant load transfer, leading to drastic variations in vertical load distribution and tire cornering stiffness, thereby critically compromising lateral stability. Conventional planning methods, reliant on static stability boundaries and fixed rollover thresholds, fail to adapt to such dynamic perturbations, resulting in degraded control performance. This paper proposes an intelligent trajectory planning framework for emergency obstacle avoidance in heavy-duty truck platoons. The framework incorporates a physics-informed neural network to predict real-time load transfer effects on tire cornering stiffness, combined with an Energy-based Rollover Index (ERI) for dynamic stability boundary assessment. Innovatively integrating the nonlinear effects of braking-induced load transfer into coordinated trajectory-speed optimization, the method enhances platoon maneuverability while ensuring safety. Simulation results demonstrate that the proposed approach increases maximum safe lane-change speeds by 0.5%-4.3% while reducing false rollover alarms to below 2%. Co-simulation using MATLAB/Simulink and TruckSim verifies a 4.2% improvement in overall planning performance while maintaining dynamic safety standards.

Prescribed performance control (PPC) provides new insights into the path tracking problem for autonomous ground vehicles (AGVs), yet conventional algorithms struggle with fluctuating road conditions and limitations for communication bandwidth. To this end, this article develops a novel road-adaptive path tracking scheme that combines flexible prescribed performance with an event-triggered mechanism, without relying on vehicle model or parameter identification. Initially, the path tracking control is abstracted as an unknown nonlinear system, sidestepping model nonlinearity, parameter variations, and external disturbances. The runnel-shaped boundary with appointed-time prescribed performance functions (PPFs) is then introduced, managing initial constraints and mitigating overshoot simultaneously. Subsequently, the flexible PPC (F-PPC) is designed to ensure proximate appointed-time stability, where the auxiliary system adjusts performance boundaries according to road width. Moreover, an adaptive event-triggered mechanism is integrated to minimize communication frequency and bit rate, requiring only 2-bit data transmission and adjusting the trigger threshold dynamically in response to road curvature. Experimental results validate the effectiveness, robustness, and efficiency of the proposed road-adaptive path tracking scheme.

To address the vehicle lateral dynamics control in practical application with nonlinearity, uncertainty, detection faults and disturbances, this paper describes three key elements in the controller design that address the safety and comfort performance challenges, i.e., high precision modeling, redundant detection and robust control. Firstly, the vehicle dynamics model, tire model and preview driver model are augmented into a coupled lateral dynamics model, which is linearized and then transformed into a linear parameter varying (LPV) model with the varying motion states and tire cornering stiffness. Secondly, a redundant detection strategy is proposed for the lane-marker-based lateral control system to improve the reliability. According to the different detection states and sequences, an$H_{\infty } $state observer is designed for the vehicle motion state estimation, where a tracking and prediction strategy of the lane markers is considered for the constraint of the dwell time to guarantee the exponential stability. Considering the linearization errors, model uncertainty and disturbances in the LPV model, the$H_{\infty } $observer-based controller is designed to improve the stability and robustness based on the Lyapunov stability theory. Lastly, three similar experiment scenarios are given to demonstrate the effectiveness of the proposed method.

Distributed drive electric systems have gained widespread use in the automotive industry with increasing intensity of motor energy recovery. Addressing issues of wheel lock-up and vehicle instability during energy recovery on low-adhesion surfaces, a regenerative torque control strategy for four-wheel distributed drives under low-adhesion conditions is proposed, suitable for energy recovery control in scenarios like low-adhesion, joint road, etc. Initially, an adhesion estimation method based on tire longitudinal dynamics and fuzzy logic is proposed. By considering slip rate and adhesion, combined with motor characteristics and the magic formula, a tire dynamics model during regenerative braking is established. Using this model, a model predictive control (MPC-RTC) strategy for regenerative torque control is implemented, compared with traditional logic threshold control strategies. Tests in joint Carsim-Simlink simulations and real-vehicle validation have been conducted. Experimental results show that the proposed low-adhesion regenerative braking anti-lock strategy improves the intensity of regenerative braking by 10.4% and increases energy recovery efficiency by 10%, maximizing energy recovery efficiency and vehicle range while ensuring vehicle stability.

Integrated-Electronic-Parking-Braking (EPBI) system is an important component of vehicles. Intelligent wheel speed sensor is the essential basic sensor of intelligent vehicle. The stable wheel speed signal is the key factor to realize EPBI control. Based on uncertain algorithm, a new type of circuit and fusion signal processing (FSP) is proposed, which solves the effective signal screening and signal fluctuation problem in traditional process system of intelligent wheel speed sensor. The wheel speed processing circuit includes an adjustable voltage regulator supply module (RSM) and signal processing module (SPM). On the basis of the circuit, FSP is designed, which can reduce the sensitivity of wheel speed signal to environmental noise to get a stable signal. The results show that this method processes the intelligent wheel speed signal effectively, improves the stability of wheel speed signal and ensure the basic braking function of EPBI.

The steer-by-wire (SbW) system has gained recognition as the future of intelligent vehicles due to its attributes, such as safety, simplification, and flexibility. However, the elimination of mechanical linkage necessitates the provision of artificial steering feedback torque (SFT), which is crucial for the potential driver to manipulate the vehicle. To enhance the steering feel, this article extends the SFT design range and proposes an SFT prediction scheme based on the sequence-to-sequence (S2S) network with the switcher-assisted (SA) training algorithm. The models of the electric power steering (EPS) and SbW are first established to analyze the input features. The S2S network with gated recurrent units (GRU) is then presented, where the encoder scheme incorporates the squeeze-and-excitation (SE) operation to achieve adaptive feature recalibration. Subsequently, the SA algorithm is proposed to alleviate the exposure bias based on the principle of multitask learning (MTL), wherein online training is regarded as the main task and offline training is treated as the auxiliary task. The weighting coefficients between tasks are optimized using an assisted network named switcher, facilitating a gradual transition from MTL to the single main task, thereby avoiding complex tuning processes. The validation results indicate the proposed scheme outperforms existing methods regarding estimation and prediction. The ablation experiments are further conducted to illustrate the effectiveness of SE blocks and the SA algorithm. Finally, the SFT construction simulation, involving target prediction and torque tracking, is conducted, validating that variable-length prediction can adapt to various conditions and improve tracking performance.

The trajectory planning module generates optimal collision-free paths, which is essential to automated driving. Existing approaches have focused on traffic flow situations and the geometrical feasibility of local trajectories. However, high nonlinearity of vehicle dynamics could cause handling instability or even severe accidents in cases of tracking dynamic infeasible paths. Meanwhile, the variety and complexity of driving environments bring intractable challenges to safe trajectory planning. This article aims to address this crucial issue and proposes a chassis global dynamics-oriented trajectory planning scheme. This work designs comprehensive performance indices to represent handling dynamic status based on chassis dynamics modeling. As the basis of trajectory planning, a reference path is extracted from global digital map data, and the curvilinear coordinate frame is exploited. Numerical optimization is proposed to solve the trajectory planning problem. Thus, an optimal and dynamically feasible trajectory, which satisfies geometrical smoothness, chassis global dynamics indices, and feasible safety, can be generated and transformed. Various scenarios are involved in carrying out simulation tests, and those results demonstrate the excellent capability and effectiveness of the proposed scheme to provide optimal trajectories in a variety of driving situations.

Vehicle chassis coordinated control always has been an appealing topic in academia and industry because of the increasing number of chassis electronic actuators with the rapid development of automated vehicles. The optimization of multiple performance targets with multiactuators is intractable, which involves trajectory tracking and handling stability. Additionally, the optimization of tire friction usage remains a knotty problem. Therefore, this article develops a global chassis multiactuator integrated control framework, named by the chassis domain controller (CDC), to realize chassis global dynamics optimization for automated vehicles. Aiming at realizing more efficient, reliable, and flexible mobility, this framework defines each individual wheel to be fully adjustable and controllable to overcome the individual actuation limitation of traditional chassis structures. Global chassis dynamic modeling is formulated based on the analysis of distributed and controllable tire modules and vehicle dynamics motions. A game-theoretical control scheme is proposed to formulate chassis multiactuator integrated control, and the chassis global dynamics can be optimized by guaranteeing a Nash equilibrium for this game. Various experimental results demonstrate the feasibility and effectiveness of the proposed control method, and it suggests the CDC merits further studies to enhance the dynamics performance of automated vehicles in full situations.

Thorough performance evaluation of automated vehicles (AVs) is an essential prerequisite for AVs' release and deployment. The challenges posed by dynamics performance appraisal of AVs are centered around the complexity of chassis dynamics, performance diversity, and lack of unified quantitative metrics. Therefore, this article proposes a novel quantitative evaluation metric for AVs' chassis-domain performance. We reveal mathematically explicit chassis steady boundaries of various vehicle maneuvers based on the modeling of chassis-domain dynamics and vehicle spatiotemporal signal analysis for safety-critical AVs. By defining and analyzing the multiperformance appraisal problem, this article gives mathematically prerequisites for evaluation metrics. Then, a rigorous metric is developed to quantify AVs' safety and comfort performance comprehensively. Wherein, the steady boundaries are leveraged to the metric normalization. We demonstrate the effectiveness of the proposed quantitative evaluation methodology in various scenarios. Test results illustrate that the proposed method provides a quantitative way to test AVs' integrated dynamics performance.

Autonomous vehicles must be comprehensively evaluated before deployed in cities and highways. However, most existing evaluation approaches for autonomous vehicles are static and lack adaptability, so they are usually inefficient in generating challenging scenarios for tested vehicles. In this paper, we propose an adaptive evaluation framework to efficiently evaluate autonomous vehicles in adversarial environments generated by deep reinforcement learning. Considering the multimodal nature of dangerous scenarios, we use ensemble models to represent different local optimums for diversity. We then utilize a nonparametric Bayesian method to cluster the adversarial policies. The proposed method is validated in a typical lane-change scenario that involves frequent interactions between the ego vehicle and the surrounding vehicles. Results show that the adversarial scenarios generated by our method significantly degrade the performance of the tested vehicles. We also illustrate different patterns of generated adversarial environments, which can be used to infer the weaknesses of the tested vehicles.

Delayed measurements can create difficulties for real-time feedback control. To address the advanced driver assistant system (ADAS) in practical application with delayed measurements, false alarms, false dismissals and disturbances, we propose a novel mixed$H_{2}/H_\infty $observer-based controller in this study, which enables object tracking and car-following. Especially, the additive and multiplicative noises of the adaptive cruise control (ACC) system can be attenuated by the proposed mixed$H_{2}/H_\infty $method. We first analyze the adaptive cruise and Radar tracking characteristics. Then, a definition of$H_{2}/H_\infty $guarantee performance is introduced to ensure satisfying target tracking and safety car-following performances. Based on$H_\infty $theory, the design criterion of the proposed mixed$H_{2}/H_\infty $observer-based controller for ACC is established by linear matrix inequality (LMI) technique. Lastly, some experiment scenarios are given to demonstrate the effectiveness of the proposed method.

Automatic lane-changing is a complex and critical task for autonomous vehicle control. Existing researches on autonomous vehicle technology mainly focus on avoiding obstacles; however, few studies have accounted for dynamic lane changing based on some certain assumptions, such as the lane-changing speed is constant or the terminal state is known in advance. In this study, a typical lane-changing scenario is developed with the consideration of preceding and lagging vehicles on the road. Based on the local trajectory generated by the global positioning system, a path planning model and a speed planning model are respectively established through the cubic polynomial interpolation. To guarantee the driving safety, passenger comfort and vehicle efficiency, a comprehensive trajectory optimization function is proposed according to the path planning model and speed planning model. In addition, a dynamic decoupling model is established to solve the problems of real-time application to provide viable solutions. The simulations and real vehicle validations are conducted, and the results highlight that the proposed method can generate a satisfactory lane-changing trajectory for automatic lane-changing actions.

Safety is a critical concern when deploying reinforcement learning agents for realistic tasks. Recently, safe reinforcement learning algorithms have been developed to optimize the agent’s performance while avoiding violations of safety constraints. However, few studies have addressed the nonstationary disturbances in the environments, which may cause catastrophic outcomes. In this paper, we propose the context-aware safe reinforcement learning (CASRL) method, a metal-earning framework to realize safe adaptation in non-stationary environments. We use a probabilistic latent variable model to achieve fast inference of the posterior environment transition distribution given the context data. Safety constraints are then evaluated with uncertainty-aware trajectory sampling. Prior safety constraints are formulated with domain knowledge to improve safety during exploration. The algorithm is evaluated in realistic safety-critical environments with non-stationary disturbances. Results show that the proposed algorithm significantly outperforms existing baselines in terms of safety and robustness.

The automated steering controller is crucial for smooth and accurate path tracking of autonomous ground vehicles (AGVs). However, time-varying uncertainties and disturbances may deteriorate the path tracking performance. Moreover, it is difficult for the steering system to strictly follow the desired steering angle in practice. Therefore, this paper proposes an automated steering control algorithm consisting of two parts: 1) an output feedback model predictive controller (MPC) to solve the path tracking problem, which is formulated as an optimization problem in this paper, with strong robustness against time-varying uncertainty and disturbance; 2) a feedforward compensator for the steering angle calculated by MPC using deep deterministic policy gradient (DDPG) algorithm so that the steering system can execute the desired steering angle more quickly and more accurately. Simulation results demonstrate that the proposed control scheme can significantly improve response speed and accuracy for path tracking of AGVs with strong robustness.

The automated steering control technology is crucial for autonomous vehicles, and lane-keeping control systems have been developed extensively, but various traffic conditions, and obstacles ahead of the ego vehicle would cause serious traffic collision accidents. Therefore, this paper proposes a virtual fluid-flow-model (VFFM) based lane-keeping integrated with collision avoidance control system (LKCA) to realize the function of both lane-keeping and collision avoidance. Firstly, a novel lane-keeping model is proposed which is inspired by viscous fluid flowing between two parallel plates or around a cylinder. Then, based on the proposed VFFM-based lane-keeping model, the ego vehicle’s path can be planned, and it ensures the performance of lane-keeping and obstacle-avoidance. Finally, an optimal-preview-driver-model-based path tracking controller is designed to track the desired VFFM-based path. The proposed control system is evaluated by both co-simulations of MATLAB/Simulink & CarSim and real-bus tests. Results show the effectiveness of the proposed control system, and it can ensure lane-keeping performance on the premise of obstacle-avoidance.

To address the advanced driver assistant system (ADAS) in practical application with some disturbances, we propose a novel finite frequency H∞observer-based method in this study, which enables effectively object tracking to restrain the disturbance during middle and low frequency ranges. We first analyze the Radar tracking characteristics. Then, an H∞observer is established. Based on the H∞theory and Kalman-Yakubovic-Popov (KYP) lemma, the design criterion of the finite frequency H∞observer is established by a linear matrix inequality (LMI). Lastly, two real world experiments are given to demonstrate the effectiveness of the proposed method.

To address the advanced driver assistant system (ADAS) in practical application with some packet dropouts, false alarms, false dismissals and disturbances, we propose a novel$H_\infty $observer-based method in this study, which enables object tracking and car-following. We first analyze the adaptive cruise and Radar tracking characteristics. Then, the adaptive cruise control (ACC) system with packet dropouts is modeled as a class of discrete-time linear switched system with four subsystems. Combining of switched system and$H_\infty $theory, the design criterion of the proposed$H_{\infty }$observer-based ACC is established. Lastly, four similar experiment scenarios are given to demonstrate the effectiveness of the proposed method.

A novel distributed cooperative formation control method inspired by the aggregate behaviors of fish groups is proposed for plug-in hybrid electric vehicle (PHEV) formation. Firstly, a hierarchical control architecture is established for formation keeping with fuel consumption (FC) optimization simultaneously. The top layer is to generate a leader with the optimal performance based on the nonlinear model predictive control (MPC) technique. A fish swarm optimization (FSO) algorithm is proposed to solve the nonlinear MPC problem by imitating the predation behaviors of the fish swarm. The middle layer is a decentralized intelligent cruise control (ICC) for follower vehicles to track their leader imitated the behaviors of fish swarm, and some design criteria are presented based on the Lyapunov stability theory. The under layer is to achieve a satisfying performance for the hybrid powertrain systems of followers. Finally, the bio-inspired method applied for PHEV formation is verified with a satisfying robustness, fuel economy, car-following and also real-time processing performances. According to the results, the PHEV formation using the proposed method represents a better car-following performance compared with the normal adaptive cruise control (ACC) method and a 21.26% improvement of FC compared with the rule-based energy management strategy (EMS). The computational burden is also reduced by the bio-inspired method.

Autonomous vehicles’ dynamics stability control is one key issue to ensure safety of self-driving. However, vehicle uncertainties and time-varying parameters could weaken the performance of autonomous vehicle stability control. Therefore, this article proposes a novel robust linear matrix inequality (LMI)-based$H$-infinite feedback algorithm for vehicle dynamics stability control, and this algorithm controls vehicle steering system and brake system via direct yaw moment control (DYC) and active front steering control (AFS). The presented controller is robust against vehicle parametric uncertainties, including the vehicle mass and vehicle longitudinal velocity. A linear parameter varying lateral model is constructed utilizing polytopic uncertainty method considering time-varying vehicle longitudinal velocity and mass, where a polytope that contains finite vertices is established to contain all of the possible selections for uncertainty parameters. Then, the$H$-infinite feedback controller integrating DYC and AFS is derived via LMI technique. Finally, experimental results based on hardware-in-the-loop (HIL) platform illustrate that the presented controller has better performance of ensuring autonomous vehicle dynamics stability than other controllers.

The longitudinal collision avoidance controller can avoid or mitigate vehicle collision accidents effectively via auto brake, and it is one of the key technologies of autonomous vehicles. Moreover, the vehicle lateral stability is very crucial in emergency scenarios. Due to complex traffic conditions and various road frictions, emergency brake may cause a vehicle to lose its lateral stability. Therefore, this paper proposes a lateral-stability-coordinated collision avoidance control system (LSCACS) based on the model predictive control (MPC). First, the proposed LSCACS decides which control mode to be implemented based on vehicle dynamics states, including a normal driving mode, a full auto brake mode, and a brake and stability mode. The MPC is used in the upper controller to calculate the desired deceleration and additional yaw moment. The lower controller calculates the desired tire forces of four wheels and realizes them by certain wheel cylinder hydraulic pressures. The LSCACS is validated by hardware-in-the-loop (HIL) tests, and the results show LSCACS's effectiveness and great performance of the collision avoidance and lateral stability.

Due to increasing complexity of systems in industrial applications, decoupling control of interconnected or cascaded systems has been discussed a lot recent years. A traditional way to deal with the system is to decouple the system states so that the states of each subsystem can be dealt with independently. This works theoretically well in most cases, however, for systems with high dimensions or with constraints on specific subsystems, this method may not be efficient or flexible enough in practical applications. In view of this, this paper proposes a backpropagation algorithm for decoupling control of a linear time invariant cascaded system with inner feedback loops, which simplifies the complicated system into several simpler ones, so that the system can be dealt with more efficiently and flexibly. Accordingly, the stability of the system is analyzed in a feedforward way, establishing the sufficient as well as necessary conditions for asymptotic stability of the system. Simulation experiment on the steering module of a steer-by-wire system verifies the effectiveness of the proposed algorithm. Finally, future researches following this paper are pointed out to perfect the algorithm proposed.

Plug-in hybrid electric buses (PHEBs), compared with traditional fuel-driven vehicles, can achieve higher fuel economy and lower pollution emissions. For a PHEB with a single-shaft parallel powertrain, a major challenge for researchers is to find approximate optimal energy management strategies that can run in real time. Motivated by this idea, this paper aims at minimizing PHEB fuel consumption with a temporal-difference (TD) learning method. First, historical driving cycle data from real-world bus routes are collected and processed and parameter variables of TD are introduced. Specially, this process is completed offline. Then, the configuration and main parameters of PHEB are presented, and a control-oriented dynamic system of the PHEB is constructed. Thereafter, the TD learning method based on historical data is introduced. Furthermore, the approximate optimal control strategy for energy management is proposed. Compared with the traditional optimal control strategy, the proposed method can realize real-time running without sacrificing the accuracy of optimization, because the learning method updates the estimates based on other learned estimates without calculating a final outcome. This method can learn directly from the data of running PHEBs without a simplified model of the PHEB, which can avoid the influence of model error. Finally, to verify this method, several different strategies are used for comparison. In addition, experimental results in real-world driving cycles demonstrate that the proposed method can improve the fuel economy obviously by up to 21% compared with a traditional charge-deleting, charge-sustaining scenario. Therefore, this novel method has great potential in realistic applications.

Spatial-temporal traffic state estimation and the prediction of urban expressways is a vital component of traffic management and information systems. The adaptive smoothing method is one of the most frequently used approaches to estimate traffic states. However, the fixed filter parameters used in existing approaches sometimes fail to characterize traffic dynamics well. To better capture generation, propagation, and mitigation dynamics of traffic congestion, we propose an adaptive rolling smoothing (ARS) approach by dynamically tuning the filter parameters in a rolling horizon scheme for online applications. The fusion of heterogeneous traffic data combines aggregate traffic measurements (e.g., traffic flow rate, time occupancy, and speed collected by remote microwave sensors) and disaggregate information (e.g., timestamps of individual vehicles detected by license plate recognition cameras). A nonlinear traffic flow filter based on the virtual trajectory algorithm is established to reconstruct the spatial-temporal traffic state and estimate experienced travel times of individual vehicles. The results demonstrate the capability and effectiveness of the proposed ARS approach in the historical traffic state estimation and short-term traffic flow prediction. Complicated traffic states of weaving, merging, and diverging segments can be well distinguished by reconstructing time-space speed diagrams. The proposed approach can be extended to develop efficient missing data imputation algorithms and hierarchical control strategies for heterogeneously congested urban expressways.

This paper considers the adaptive fuzzy output feedback tracking control problem for a class of uncertain nonlinear switched systems with time delay and unmodeled dynamics. Based on a kind of switched K-filters, a prescribed performance control scheme is proposed to guarantee the tracking performance and restrain the fluctuation caused by switches between submodes as well. In addition, fuzzy logic systems (FLSs) are use to approximate unknown nonlinear functions and dynamic surface control (DSC) method is employed to eliminate the explosion of complexity problem inherent in traditional backstepping method. The proposed controllers of corresponding subsystems guarantee that all closed-loop signals remain bounded under a class of switching signals with average dwell time (ADT). A numerical simulation is performed to illustrate the effectiveness of the proposed approach.

Gear engagement is the most important part in gear-shift process of automated manual transmission (AMT). However, it is practical to encounter complicated nonlinearities, uncertainties, and multistage characteristics in the system model, so the controller design for the AMT gear engagement becomes challenging. This paper proposes a dual-loop self-learning fuzzy control framework. In the outer loop, the self-learning rules based on fuzzy logic is designed to adjust desired trajectory of actuator motor. In the inner loop, the gear engagement is divided into three stages, and a fuzzy controller with model reference self-learning algorithm is designed, which controls the actuator motor to track the desired trajectory. Besides, the control parameters could be adjusted to be optimal automatically when the parameters change. Results of simulations and experiments indicate that the proposed method is able to realize the smooth and fast control of gear engagement. In addition, the self-learning fuzzy controller can be extended to deal with other nonlinear systems with uncertain and even unknown parameters.

Driving cycles and road slope are two important factors affecting fuel saving performance of plug-in hybrid electric buses (PHEBs) in Chinese cities. Moreover, onboard auxiliary equipment (e.g., Global Position System receiver and General Packet Radio Service (GPRS) wireless module) of PHEB may provide potential means to communicate with the control center of the bus company, allowing for driving cycle prediction through data communication between foregoing buses and the control center. With this general approach in mind, and by utilizing driving data clustering and driving cycle classifier, this paper presents a multimode switched logic control strategy, targeting fuel economy improvement of the PHEB team for a particular city bus route. First, the normal feature parameters are extracted from the sampled driving history cycles, and the composed feature parameters are given by a mapping of normal feature parameters in this approach. A novel improved hierarchical clustering algorithm is applied for driving cycles' data clustering into four groups. Then, on the basis of the clustering results, support vector machine method is used to predict the current driving cycle. Finally, a switched driving controller is presented according to current type of driving cycle and slope information. Simulation results are compared with those of traditional methods in the given real-world driving cycles of city bus, showing significant improvement, which may offer a theoretical solution with engineering application. Experimental results also demonstrate that the proposed control approach is feasible in the tested bus routes.