The development of vehicle intelligence and connectivity provides the opportunity to integrate more road information for vehicle control. This article proposes a preview active suspension control method based on model predictive control (MPC) with the aim of suppressing vertical vibration. A key challenge in implementing MPC lies in the heavy computational burden associated with solving Quadratic Programming (QP) problems. To address this, a QP algorithm based on non-negative least squares (NNLS) is utilized to accelerate the controller solving speed, thereby enhancing its practical feasibility. Additionally, to further improve ride comfort, a speed planning method is developed. This method addresses a multi-objective optimization problem (MOP) incorporating both travel time and comfort, then a feasible solution is chosen from the optimal solution set according to passenger preferences. Various experiment and simulation results are provided to validates the effectiveness of the controller, speed planning and the integrated design of both. Compared with traveling at a constant speed, the proposed speed planning method decreases the RMS value of body vertical acceleration by 17.3% while concurrently achieving shorter travel time.

Electricity price forecasting remains a critical research focus in modern power systems, where future electricity prices are influenced not only by historical electricity price patterns but also by exogenous factors. Traditional approaches often employ single-stream recurrent neural networks to integrate exogenous variables with historical electricity price sequences. However, since historical electricity prices and exogenous information inherently represent distinct modalities, single-stream architectures are suboptimal for processing such multimodal data, leading to inefficient feature extraction. To address this limitation, we propose DSFormer, a dual-stream transformer framework designed for electricity price forecasting. This dual-stream architecture employs parallel processing, with the parsing stream handling historical electricity price sequences and the exogenous stream processing exogenous variables. Specifically, DSFormer comprises an encoder and a decoder. The encoder incorporates two key components: a fusion mutual attention mechanism and a multispace feedforward network. The fusion mutual attention mechanism comprises a pair of mutual attention layers, followed by a fusion attention layer. The mutual attention layers independently inject historical electricity price information into the exogenous stream and exogenous information into the parsing stream, while the fusion attention layer further refines the features from dual streams. The multispace feedforward network projects the features from dual streams into subspaces and performs cross-subspace interactions based on the gating mechanism, thereby enhancing representation capabilities while achieving enhanced computational efficiency. Finally, the decoder infers future electricity prices from these refined features. Extensive experimental results demonstrate that DSFormer outperforms existing state-of-the-art methods in terms of prediction accuracy while maintaining lower computational complexity.

Vehicle-infrastructure collaborative perception has demonstrated significant potential in extending perception range and enhancing 3D object detection accuracy. However, there remain challenges due to limitations in communication bandwidth and the redundancy of shared information. To address these issues, we propose Why2comm, a novel framework for vehicle-infrastructure collaborative perception. By incorporating driving trajectories and region decoupling, Why2comm focuses information sharing on key areas that directly impact driving tasks, thereby minimizing redundant data transmission and improving collaborative perception efficiency. Specifically, we design a dynamic trajectory-aware message filtering module that investigates the potential correlations between driving trajectories and instance features, utilizing an attention mechanism to identify and extract the critical instance features. Furthermore, we propose a region-aware feature transmission module that decouples shared and exclusive regions, achieving efficient data transmission. Finally, we develop a position-guided fusion module to effectively integrate and enhance features from both vehicle and infrastructure. Evaluation results on the V2XSet, DAIR-V2X, and DeepAccident datasets show that Why2comm achieves superior detection performance compared to state-of-the-art methods while requiring merely 10.7% of the communication bandwidth.

Trajectory prediction is a continuing concern within autonomous vehicles. Psychological research shows that pedestrian traveling is a cyclic alternation. Pedestrians constantly interact with their surroundings, including social agents and physical environments, and plan paths to achieve goals. Nevertheless, most existing trajectory prediction methods are based on a single-stage design, which runs counter to traffic psychology principles. In this work, we present MPNet, a novel multi-stage progressive convolutional neural network that decomposes complicated trajectory prediction into multiple manageable components, where lightweight sub-networks handle each stage with a divide-and-conquer methodology. Specifically, our sub-networks are based on the encoder-decoder architecture, in which we capture interactive information and estimate goals to achieve trajectory prediction. The communication among sub-networks depends on a novel cross-stage fusion design. We introduce a feedback channel mutual attention mechanism and a cross-sub-network fusion unit to enable efficient information sharing across different stages. At each stage, we develop a symmetric gated supervision module to supervise future trajectory generation from coarse to fine. Extensive experiments demonstrate that MPNet achieves state-of-the-art performance, reducing Average Displacement Error (ADE) and Final Displacement Error (FDE) by 5.6%/7.4% on the ETH and UCY Datasets, 7.5%/7.7% on the Stanford Drone Dataset, and 7.1%/22.5% on the Intersection Drone Dataset, while maintaining comparable computational complexity. Additionally, our MPNet-Tiny variant reduces parameters by 90.5% and inference time by 1.83 seconds with competitive accuracy.

Object detection is a fundamental task in computer vision and supports more complex tasks, such as object tracking and trajectory prediction. Despite significant advancements in deep learning, traditional loss functions often optimize classification and localization independently, requiring manual tuning of multiple weights. This approach can lead to misalignment with mean average precision (mAP). In this article, we propose a novel mAP-based loss function that directly correlates with the mAP metric. Our contributions include a novel perspective on designing loss functions for object detection, a new strategy for assigning positive and negative samples, and the establishment of a new state-of-the-art approach for loss function design. Our approach improves both interpretability and performance, effectively overcoming the limitations of conventional methods that rely on manually designed weights. Extensive experiments on standard benchmarks demonstrate that our method consistently outperforms existing State-of-the-Art techniques.

Incorporating both RGB and depth images has proven effective for enhancing the performance of semantic segmentation. However, current RGB-D semantic segmentation methods tend to overlook the critical role of cross-modal difference information during fusion, leading to the undesired suppression of discriminative cues and a failure to achieve potent cross-modal complementary fusion. In this article, a novel RGB-D semantic segmentation approach that realizes the efficient utilization of multimodal information is proposed. To address the issue of the suppression of cross-modal difference information, we propose a dynamic frequency-spatial difference-aware fusion module adept at explicitly emphasizing cross-modal differences, capturing vital features in the frequency domain, and using them to aggregate spatial context information of multimodal features. We also present a novel soft-edge loss to meticulously handle complex scenes by supervising different regions respectively. In addition, a progressive calibration context module is designed to enhance global contextual information by capturing multiscale multimodal representations. Extensive experiments on two public RGB-D datasets demonstrate that the proposed DFNet achieves highly competitive performance compared to state-of-the-art methods, making it well-suited for assisting indoor robots.

Diffusion probabilistic models (DPMs) have recently achieved brilliant achievements in computer vision. Inspired by the success of DPMs, we present TrajDiff, a model based on conditional diffusion probabilistic models for agent future trajectory prediction, which speculates the agent future states through a series of stochastic iterative denoising processes. Specifically, we map the trajectory prediction task into the latent heatmap space, translating hard keypoint prediction into soft cluster center learning. The core architecture is a U-shaped encoder-decoder network (U-Net) that is trained with a denoising objective. During inference, conditioned on the observed past trajectory heatmaps, random pure Gaussian noise is initialized to drive the reverse sampling process. The U-Net iteratively removes various levels of Gaussian noise from initialized images, resembling Langevin dynamics, and generates multi-modal predicted future trajectory heatmaps. Furthermore, we introduce a novel residual block with a mutual attention mechanism that can elegantly consider the interactions between the agent and the surrounding environment at multiple scales, assisting in generating physically and socially acceptable trajectories. We verify TrajDiff on the Stanford Drone Dataset and the ETH and UCY Datasets. The experimental results show that TrajDiff outperforms previous state-of-the-art methods with considerable accuracy gains, while significantly reducing computational requirements.

Monocular 3D object detection finds applications in various fields, notably in intelligent driving, due to its cost-effectiveness and ease of deployment. However, its accuracy significantly lags behind LiDAR-based methods, primarily because the monocular depth estimation problem is inherently challenging. While some methods leverage additional information to aid in network training and enhance performance, they are hindered by their reliance on specific datasets. We contend that many components of monocular 3D object detection lack the necessary adaptability, impeding the performance of the detector. In this paper, we propose six adaptive methods addressing issues related to network structure, loss function, and optimizer. These methods specifically target the rigid components within the detector that hinder adaptability. Simultaneously, we provide theoretical insights into the network output and propose two novel regression methods. These methods facilitate more straightforward learning for the network. Importantly, our approach does not depend on supplementary information, allowing for end-to-end training. In comparison with existing methods, our proposed approach demonstrates competitive speed and accuracy. On the KITTI dataset, our method achieves a 17.72% AP3D(IOU =0.7, Car, Moderate), outperforming all previous monocular methods. Additionally, our approach prioritizes speed, achieving a runtime of up to 52 FPS on an RTX 2080Ti GPU, surpassing all previous monocular methods. The source codes are at:https://github.com/jiayisong/AMNet.

Deep feature fusion plays a significant role in the strong learning ability of convolutional neural networks (CNNs) for computer vision tasks. Recently, works continually demonstrate the advantages of efficient aggregation strategy and some of them refer to multiscale representations. In this article, we describe a novel network architecture for high-level computer vision tasks where densely connected feature fusion provides multiscale representations for the residual network. We term our method the ResDNet which is a simple and efficient backbone made up of sequential ResDNet modules containing the variants of dense blocks named sliding dense blocks (SDBs). Compared with DenseNet, ResDNet enhances the feature fusion and reduces the redundancy by shallower densely connected architectures. Experimental results on three classification benchmarks including CIFAR-10, CIFAR-100, and ImageNet demonstrate the effectiveness of ResDNet. ResDNet always outperforms DenseNet using much less computation on CIFAR-100. On ImageNet, ResDNet-B-129 achieves 1.94% and 0.89% top-1 accuracy improvement over ResNet-50 and DenseNet-201 with similar complexity. Besides, ResDNet with more than 1000 layers achieves remarkable accuracy on CIFAR compared with other state-of-the-art results. Based on MMdetection implementation of RetinaNet, ResDNet-B-129 improves mAP from 36.3 to 39.5 compared with ResNet-50 on COCO dataset.

Developing a solution to suppress vibration with low energy consumption is an interesting and significant topic for vehicle suspension. This paper proposes an approximation-free prespecified time energy-efficient fault-tolerant control scheme for active suspensions. Inspired by the X-shaped asymmetric structures that existed in animal limbs, the energy consumption can be significantly reduced, without changing the hardware structure and optimization, only by introducing bionic dynamics. Moreover, the states can converge to a neighborhood of zero within a pre-specified finite time interval by employing the designed bionic control framework. Note that the settling time is independent of the initial values of states and the controller parameters. Meanwhile, system uncertainties, external perturbations, and actuator faults can be handled effectively by the presented approximation-free controller. The knowledge of actuator faults, perturbation, and suspension model is not required for controller design. By using the designed approximation-free prespecified time bionic fault-tolerant controller, excellent ride comfort can be achieved with low energy consumption. Experiments are performed and corresponding comparison results are provided to demonstrate the superiority of the developed control method.Note to Practitioners—Vehicle active suspensions offer a higher degree of control freedom to isolate vibrations, however, are associated with expensive energy consumption. Model uncertainties, non-linearities, external disturbances, and actuator faults in real-world environments can lead to performance degradation. Motivated by this, this paper designs a bio-inspired energy-saving prespecified time approximation-free fault-tolerant controller. By employing dynamical properties inspired by X-type structures that exist in animal limbs or bones, efficient energy-saving control can be achieved. The transient responses are constrained to a prescribed region and converge to a given steady-state region within a pre-assigned time. The proposed controller does not require approximation structures such as neural networks or fuzzy rules. Consequently, the controller exhibits a simple structure that can be readily implemented and deployed in practical suspension systems. Experimental results obtained from practical suspension platforms under different road excitations demonstrate satisfactory energy saving and vibration suppression performance of the presented control scheme.

This paper investigates the issue of event-triggered adaptive saturated fault-tolerant control (ESFC) for uncertain nonlinear systems with time-varying full state constraints (TFSCs), actuator saturation and faults as well as unknown control direction. A bounded function with an auxiliary variable is constructed by utilizing a novel dynamics of the auxiliary system, which contributes to reducing the adverse impact of actuator saturation. Different from the previous backstepping-based event-triggered control methods such specifications by either using fuzzy approximation or by employing neural approximation techniques, this paper skillfully addresses the unknown nonlinearities, actuator saturation and faults without involving any approximation structures, and thus, we proposes the ESFC on the basis of low-complexity design framework as contributing to communication and computational resource reduction. A rigorous theoretical analysis shows that the proposed control method is an effective way to handle with the problems of actuator saturation and faults, full state constraints, and unknown system uncertainties, while simultaneously simplifying the backstepping design and avoiding the issue of explosion of complexity. The asymptotic stability of the closed-loop system is guaranteed and the Zeno behavior can be effectively removed. We present an application example of a linear motor scenario to illustrate the effectiveness of the method.Note to Practitioners—Since state constraints, actuator saturation and faults, unknown mechanism model, and limited bandwidths exist extensively in practical engineering systems, which constantly degrade the operation performance of the plant. To handle these disadvantages, this paper is focus on providing simple but effective ESFC methods to ensure the asymptotic stability and enhance reliability. Compared to existing results, the presented method only uses the state signals of system without using system dynamic functions under mild conditions, which provides a theoretical basis, and has the advantages of low-complexity design, and easy implementation in practical engineering. Preliminary physical experimental comparisons demonstrate that this method is applicable to practical liner-motor platform, and achieves satisfactory control performance.

This paper develops an event-triggered fault-tolerant tracking strategy for block-triangular MIMO uncertain nonlinear systems with sensors and actuators polluting by multiplicative/additive faults and unknown control directions, to address the time-varying asymmetric output constraint control. By combining an adaptive fault-tolerant control law with event-triggered mechanisms (ETMs), the presented method possesses the properties of low structure and calculation complexity, and can effectively conserve the system resources of communication and computation. To solve the issues of unknown control directions and time-varying asymmetric output constraints, the proposed method utilizes the integrated design of the Nussbaum function and barrier Lyapunov function (BLF) to realize a novel constrained tracking control with strong robustness, and can eliminate the adverse effects of the output tracking caused by all state (except for output) sensor faults. The proposed controller operates without having to use any approximating techniques, has the capability to handle the coupling uncertain terms derived from unknown system functions, sensor and actuator faults, and ETMs, and avoids the issue of explosion of complexity as in traditional backstepping procedure. The closed-loop stability can be guaranteed based on Lyapunov stability analysis with contradiction, while ensuring the boundedness of all signals, maintaining the output constraint, and preventing the Zeno behavior. Finally, the potential of application is investigated by means of experiments on a Linear Motor system, illustrating the effectiveness. Note to Practitioners— Due to the existence of output and bandwidth constraints, sensor and actuator faults, and unknown system models in practical plants, the operational performance of the system may inevitably degraded. To address this issue, this study primarily focuses on developing a low-complexity adaptive control method that aim to guarantee the overall performance of the system. Existing approaches mainly rely on system dynamic functions or adopt the approximation technology, our methodology can only utilizes the state signals of the system, which possesses advantages of low-complexity design and easy implementation. Preliminary experiments conducted through physical experiments demonstrate the applicability of this method to practical linear-motor platforms, yielding satisfactory control performance.

Dual-drive crossbeam system has been an significant part of high-end equipment, whose synchronous accuracy and robustness determine the performance if equipment. Thus, a robust iterative learning control method is proposed, which is based on the cross-coupling model. Robust term is to deal with the non-repetitive disturbance and iterative learning term is to compensate the unknown dynamics by repeating the same task. Finally, different control methods are conducted to verify the effectiveness of the proposed method.

This paper proposed a neural networks-based adaptive control scheme for linear motors considering cogging force compensation. The cogging force is modeled and compensated for improving the tracking performance. An indirect parameter adaptive strategy is proposed to address the problem of parametric uncertainties. Compared with the direct adaptive strategy, this strategy can promote the parameter estimations to converge to the true values. In addition, radial basis neural networks are designed to estimate remaining system uncertainties, including unmodeled dynamics, model errors, and external disturbances. Comparative experiments are conducted on an iron-core permanent magnet linear synchronous motor platform. The experimental results show that the proposed control scheme can achieve excellent control performance.

Synchronization control and coordination control are the core issues of dual-linear-motor-driven gantries (DLMDG). This paper proposes a novel integrated realtime compensation super-twisting sliding-mode control (RCSTSMC) framework that aims to simultaneously improve the contouring accuracy and synchronization performance of DLMDG for highly repetitive tracking tasks. The established control scheme combines the fast convergence and immunity to disturbance of the higher order sliding mode and the advantage of model-based realtime compensation. The results of the simulation confirm the superiority of the RCSTSMC. Compared to the ARC, the RCSTSMC has improved synchronization performance and contouring performance by 72.43 % and 69.11 % respectively, and the maximum synchronization and contouring errors have been reduced by 5.19 μrad and 0.96 μm.

Camera and LiDAR are indispensable perception units in autonomous driving, providing complementary environmental information for 3D semantic segmentation. It is the key point that fuses the information of two modalities to accurate and robust semantic segmentation. However, three major factors will restrict the performance of fusion-based methods, i.e., the reliability of image features, the contribution of different image features, and the trade-off between results of image and point cloud. This paper proposes a novel multi-phase fusion network for 3D semantic segmentation. For the first factor, this paper takes the lead in regarding the problem that image features may be wrong due to the lack of dense annotations in the common datasets as a weak supervision problem and introduces the weakly supervised loss. Second, the proposed attention based feature fusion module can filter and reweight the image features effectively. Third, the results of the two modalities are further fused by self-confidence based late fusion module at pixel-level to complement their advantages. The proposed scheme has been evaluated on nuScenes and SemanticKITTI benchmarks, and the results show the competitiveness with state-of-the-art methods. The ablation studies demonstrate the superiority of the method in sparse classes segmentation. In addition, the robustness is also evaluated, and the results of the proposed method can keep relatively accurate even when faults in one of the sensors.

Autonomous driving has witnessed rapid development with the application of artificial intelligence technology in recent years. Lane detection is one of the tasks of environment perception, which affects the planning and decision-making directly, and requires the algorithm to meet both high precision and high efficiency. Most of the existing methods extract pixels belonging to lanes in the image, which should be postprocessed, otherwise it cannot be applied to subsequent tasks like planning. This article proposes an end-to-end lane detection method that utilizes auxiliary supervision and knowledge distillation based teaching-test module to predict the parameters of polynomials of lanes directly. The teaching-test module guides the polynomial regression branch to learn the shape features from the segmentation branch to improve the fitting accuracy under complex road conditions. The proposed method is validated on TuSimple and CULane datasets, and is competitive with state-of-the-art methods in efficiency and accuracy.

Using light-weight architectures or reasoning on low-resolution images, recent methods realize very fast scene parsing, even running at more than 100 FPS on a single GPU. However, there is still a significant gap in performance between these real-time methods and the models based on dilation backbones. To this end, we proposed a family of deep dual-resolution networks (DDRNets) for real-time and accurate semantic segmentation, which consist of deep dual-resolution backbones and enhanced low-resolution contextual information extractors. The two deep branches and multiple bilateral fusions of backbones generate higher quality details compared to existing two-pathway methods. The enhanced contextual information extractor named Deep Aggregation Pyramid Pooling Module (DAPPM) enlarges effective receptive fields and fuses multi-scale context based on low-resolution feature maps with little time cost. Our method achieves a new state-of-the-art trade-off between accuracy and speed on both Cityscapes and CamVid dataset. For the input of full resolution, on a single 2080Ti GPU without hardware acceleration, DDRNet-23-slim yields 77.4% mIoU at 102 FPS on Cityscapes test set and 74.7% mIoU at 230 FPS on CamVid test set. With widely used test augmentation, our method is superior to most state-of-the-art models and requires much less computation. Codes and trained models are available athttps://github.com/ydhongHIT/DDRNet.

This article investigates the completely unknown autonomous vehicle tracking issues with actuator faults through model-free adaptive dynamic programming (MFADP) approaches. Because partial parameters are measured difficultly or inaccurately, the model-based control theories are imperfect for the vehicles. Therefore, the proposed multiplayer optimal control method in this work, which is not necessary to know the prior system knowledge, achieves the purpose of unknown vehicle tracking control via a novel MFADP theory. Besides, the control strategies are robust, which contain adaptive regulators to eliminate the disturbance of the vehicle systems caused by actuator faults, modeling errors, and curvature interference. To reduce the computational burden of control, a single neural network (NN) architecture is constructed with minimal computational cost and fast response speed. In addition, the convergence analysis of the NN structure, the stability and robustness analysis of identification, and the control schemes in this work are supplied. Finally, two driving scenario simulations are shown to prove the effectiveness of the established controller.

In this article, an adaptive event-triggered fault-tolerant asymptotic tracking control problem guaranteeing prescribed performance is addressed for a class of block-triangular multi-input and multioutput uncertain nonlinear systems with unknown nonlinearities, unknown control directions, and actuator faults. Through a systematic co-design of the adaptive control law and the event-triggered mechanism, including fixed and relative threshold strategies, a control scheme with low structure and calculation complexity is designed to conserve system communication and computation resources. In this design, the output asymptotic tracking is achieved. The Nussbaum gain technique is incorporated to overcome unknown control directions with a new adaptive law, and a type of barrier Lyapunov function is adopted to handle the prescribed performance control problem, which contributes to a novel control law with strong robustness. The robust controller can address the uncertainties and couplings derived from the system structure, actuator faults, and event-triggered rules, without using approximating structures or compensators. Besides, the explosion of complexity is avoided. It is proved that all signals of the closed-loop system remain bounded, and system tracking errors asymptotically approach 0 with the prescribed performance, while the Zeno behavior is prevented. Finally, the effectiveness of the proposed control scheme is evaluated via an application example of the half-car active suspension system.

Online multi-object tracking (MOT) is an active research topic in the domain of computer vision. Although many previously proposed algorithms have exhibited decent results, the issue of tracklet inactivation has not been sufficiently studied. Simple strategies such as using a fixed threshold on classification scores are adopted, yielding undesirable tracking mistakes and limiting the overall performance. In this paper, a conditional random field (CRF) based framework is put forward to tackle the tracklet inactivation issue in online MOT problems. A discrete CRF which exploits the intra-frame relationship between tracking hypotheses is developed to improve the robustness of tracklet inactivation. Separate sets of feature functions are designed for the unary and binary terms in the CRF, which take into account various tracking challenges in practical scenarios. To handle the problem of varying CRF nodes in the MOT context, two strategies named as hypothesis filtering and dummy nodes are employed. In the proposed framework, the inference stage is conducted by using the loopy belief propagation algorithm, and the CRF parameters are determined by utilizing the maximum likelihood estimation method followed by slight manual adjustment. Experimental results show that the tracker combined with the CRF-based framework outperforms the baseline on the MOT16 and MOT17 benchmarks. The extensibility of the proposed framework is further validated by an extensive experiment.

Convolutional neural networks (CNN) have been proved to be an effective method in the field of artificial intelligence (AI), and large-scale deploying CNN to embedded devices, no doubt, will greatly promote the development and application of AI into the practical industry. However, mainly due to the space-time complexity of CNN, computing power, memory bandwidth and flexibility are performance bottlenecks. In this paper, a framework containing model compression and hardware acceleration is proposed to solve the above problems. This framework consists of a mixed pruning method, data storage optimization for efficient memory utilization and an accelerator for mapping CNN on field programmable gate array (FPGA). The mixed pruning method is used to compress the model, and data bit-width is reduced to 8-bit by data quantization. Accelerator based on FPGA makes it flexible, configurable and efficient for CNN implementation. The model compression is evaluated on NVIDIA RTX2080Ti, and the results illustrate that the VGG16 is compressed by 30× and the fully convolutional network (FCN) is compressed by 11× within 1% accuracy loss. The compressed model is deployed and accelerated on ZCU102, which is up to 1.7× and 24.5× better in energy efficiency compared with RTX2080Ti and Intel i7 7700.

Modern autonomous vehicles are required to perform various visual perception tasks for scene construction and motion decision. The multiobject tracking and instance segmentation (MOTS) are the main tasks since they directly influence the steering and braking of the car. Implementing both tasks using a multitask learning neural network presents significant challenges in performance and complexity. Current work on MOTS devotes to improve the precision of the network with a two-stage tracking by detection model, which is difficult to satisfy the real-time requirement of autonomous vehicles. In this article, a real-time multitask network named YolTrack based on one-stage instance segmentation model is proposed to perform the MOTS task, achieving an inference speed of 29.5 frames per second (fps) with slight accuracy and precision drop. The YolTrack uses ShuffleNet V2 with feature pyramid network (FPN) as a backbone, from which two decoders are extended to generate instance segments and embedding vectors. Segmentation masks are used to improve the tracking performance by performing logic AND operation with feature maps, proving that foreground segmentation plays an important role in object tracking. The different scales of multiple tasks are balanced by the optimized geometric mean loss during the training phase. Experimental results on the KITTI MOTS data set show that YolTrack outperforms other state-of-the-art MOTS architectures in real-time aspect and is appropriate for deployment in autonomous vehicles.

This article is devoted to the event-triggered adaptive control design for uncertain nonlinear systems with full state constraints. A robust adaptive control method enabling the codesign of event-triggering mechanism is proposed, in which the communication burden between controllers and actuators is reduced, and both the physical limitation of the plant with uncertainties and the measurement errors introduced by event-triggering mechanisms can be simultaneously addressed. In addition, a priori knowledge of the signs of unknown virtual control coefficients is not required in the presented controller design methods. Furthermore, Lyapunov stability analysis guarantees that all states in the closed-loop nonlinear system are bounded, the state constraints are not violated, and the tracking errors are driven to a compact set. Finally, a designed example is given to illustrate the effectiveness and advantages of the presented design approaches.

In this paper, an adaptive fault-tolerant method is proposed to synthesize a compensation controller for the uncertain nonlinear pure-feedback systems possessing dead-zone actuators and stochastic failures. Each actuator's failure mode is described by a scalar Markovian type function, which is not only much more practical in control engineering but also challenging in control theory. By exploring the adaptive backstepping methodology, a compensation scheme for actuator failure is presented to guarantee that the solution of the closed-loop system is a unique and bounded in probability. Importantly, the proposed control method can achieve arbitrarily small tracking error in the presence of nonlinear actuators with random failures. The case study of active suspension system using the adaptive fault-tolerant compensation controller shows the effectiveness of the presented method.

In this paper, an output feedback finite-time control method is investigated for stabilizing the perturbed vehicle active suspension system to improve the suspension performance. Since physical suspension systems always exist in the phenomenon of uncertainty or external disturbance, a novel disturbance compensator with finite-time convergence performance is proposed for efficiently compensating the unknown external disturbance. Moreover, the presented compensator is advantageous over the existing ones since it is continuous and can completely remove the matched disturbance. From the viewpoint of practical implementation, continuous control law will not lead to chattering, which is desirable for electrical and mechanical systems. For the nominal suspension system without disturbance, a homogeneous controller with a simple filter is constructed to achieve a finite-time convergence property, where the filter is applied to obtain the unknown velocity signal. Thus, the nominal controller combines a disturbance compensator into an overall continuous control law, which provides two independent parts with a separate design unit and a high flexibility for selecting the control gains. According to the geometric homogeneity and finite-time separation principle, it can be shown that the active suspension is finite-time stabilized. A designed example is given to illustrate the effectiveness of the presented controller for improving the vehicle ride performance.

This paper is concerned with the problem of adaptive tracking control for a class of nonlinear systems with parametric uncertainty, bounded external disturbance, and actuator saturation. In order to achieve robust output tracking for the saturated uncertain nonlinear systems, a combination of adaptive robust control (ARC) and a novel terminal sliding-mode-based nonlinear disturbance observer (TSDO) is proposed, where the modeling inaccuracy and disturbance are integrated as a lumped disturbance. Specifically, the observer errors of estimating the lump disturbances converge to zero in finite-time for improving the precision of estimation. The estimated disturbances are then used in the controller to compensate for the system's lumped disturbances. The analytical results show that the proposed scheme is stable and can guarantee the asymptotic tracking with the tracking error converging to zero even in the presence of disturbances. Finally, the developed method is illustrated the effectiveness by the application to control of a quarter-car model with active suspension system.

This paper presents the problem of finite-time stabilization for vehicle suspension systems with hard constraints based on terminal sliding-mode (TSM) control. As we know, one of the strong points of TSM control is its finite-time convergence to a given equilibrium of the system under consideration, which may be useful in specific applications. However, two main problems hindering the application of the TSM control are the singularity and chattering in TSM control systems. This paper proposes a novel second-order sliding-mode algorithm to soften the switching control law. The effect of the equivalent low-pass filter can be properly controlled in the algorithm based on requirements. Meantime, since the derivatives of term with fractional power do not appear in the control law, the control singularity is avoided. Thus, a chattering-free TSM control scheme for suspension systems is proposed, which allows both the chattering and singularity problems to be resolved. Finally, the effectiveness of the proposed approach is illustrated by both theoretical analysis and comparative experiment results.