Deep reinforcement learning (DRL) is a promising way to develop autonomous driving decision-making models. However, poor driving decisions and low sample efficiency for multiple DRL coupled training hinder its applications in driving decision-making models. This article proposes an innovative framework to combine two different DRL algorithms as the upper- and lower-layer planner to make car-following and lane-changing decisions respectively. The upper- and lower-layer models are trained simultaneously, and the double-layer model outputs a composite driving action. In the upper-layer model, using TD3 algorithm generates continuous vehicle speed. This article proposes the action exploration mechanism where the TD3 algorithm selects one of two action policies with a probability and then outputs an action value in the early training phase. Moreover, the proposed Q-value auxiliary networks guide our upper-layer algorithm to compute the Q-value based on a Q-value from the trained TD3 algorithm. Dueling-double DQN is used to address the issue of how a vehicle changes lanes in the lower-layer model and to output discrete values to instruct the vehicle to change lanes. To validate our model in autonomous driving applications, different training and testing scenarios simulating expressways are designed through SUMO. The experiments show that our methods address the difficulty of coupling training for the integrated model and improve its performance under different traffic flow scenarios. Compared with other models, our model enhances driving velocity while ensuring vehicle safety.

The Transformer has been successfully used in medical image segmentation due to its excellent long-range modeling capabilities. However, patch segmentation is necessary when building a Transformer class model. This process ignores the tissue structure features within patch, resulting in the loss of shallow representation information. In this study, we propose a Heterogeneous Swin Transformer with Multi-Receptive Field (HST-MRF) model that fuses patch information from different receptive fields to solve the problem of loss of feature information caused by patch segmentation. The heterogeneous Swin Transformer (HST) is the core module, which achieves the interaction of multi-receptive field patch information through heterogeneous attention and passes it to the next stage for progressive learning, thus complementing the patch structure information. We also designed a two-stage fusion module, multimodal bilinear pooling (MBP), to assist HST in further fusing multi-receptive field information and combining low-level and high-level semantic information for accurate localization of lesion regions. In addition, we developed adaptive patch embedding (APE) and soft channel attention (SCA) modules to retain more valuable information when acquiring patch embedding and filtering channel features, respectively, thereby improving model segmentation quality. We evaluated HST-MRF on multiple datasets for polyp, skin lesion and breast ultrasound segmentation tasks. Experimental results show that our proposed method outperforms state-of-the-art models and can achieve superior performance. Furthermore, we verified the effectiveness of each module and the benefits of multi-receptive field segmentation in reducing the loss of structural information through ablation experiments and qualitative analysis.

Research in medical visual question answering (MVQA) can contribute to the development of computer-aided diagnosis. MVQA is a task that aims to predict accurate and convincing answers based on given medical images and associated natural language questions. This task requires extracting medical knowledge-rich feature content and making fine-grained understandings of them. Therefore, constructing an effective feature extraction and understanding scheme are keys to modeling. Existing MVQA question extraction schemes mainly focus on word information, ignoring medical information in the text, such as medical concepts and domain-specific terms. Meanwhile, some visual and textual feature understanding schemes cannot effectively capture the correlation between regions and keywords for reasonable visual reasoning. In this study, a dual-attention learning network with word and sentence embedding (DALNet-WSE) is proposed. We design a module, transformer with sentence embedding (TSE), to extract a double embedding representation of questions containing keywords and medical information. A dual-attention learning (DAL) module consisting of self-attention and guided attention is proposed to model intensive intramodal and intermodal interactions. With multiple DAL modules (DALs), learning visual and textual co-attention can increase the granularity of understanding and improve visual reasoning. Experimental results on the ImageCLEF 2019 VQA-MED (VQA-MED 2019) and VQA-RAD datasets demonstrate that our proposed method outperforms previous state-of-the-art methods. According to the ablation studies and Grad-CAM maps, DALNet-WSE can extract rich textual information and has strong visual reasoning ability.

This article investigates the reliability of a special cyber-physical system (CPS) with an unreliable service and a complex boundary behavior. A flat semi-dormant multicontroller (FSDMC) model is constructed on a special CPS named arbitrated networked control system (ANCS) with dual channels. In this study, the dual-channel ANCS is considered as a Markov repairable system, which integrates the binary state of physical device failure and the multistate of information flow. The FSDMC is modeled as anN/(d,c)-M/M/c/K/SMWVqueuing system with an unreliable service. A dual rate matrix method is proposed to solve the stationary distribution of the queuing system and obtain the closed-form matrix solution of the distribution. Based on the queuing model, an optimization model is established to minimize the proposed cost performance rate function. A particle swarm optimization algorithm is used to solve the optimization model and obtain the optimal values of the system parameters under stable conditions. The closed-form expression of the instantaneous availability of the FSDMC on the physical failure rate and repair rate of the controller is yielded iteratively. The linear relationship between system instantaneous failure rate and task instantaneous failure rate is expressed. The sensitivity of task failure rate to system parameters is analyzed. Several reliability metrics are used to evaluate system reliability and task reliability. Experiments are conducted in real application scenarios to compare the task reliability using redundancy technology and real parallel applications. Experiments show that the proposed reliability model can more effectively guarantee system reliability goals compared with its counterparts.

As an essential feature of cloud computing, dynamic scalability enables the cloud system to dynamically expand or shrink resources according to user needs at runtime. Effectively predicting and optimizing the cost and performance of cloud computing platforms have become one of the key research challenges in the field of cloud computing. In this article, to quantitatively predict the cost and performance of cloud computing platforms, we propose a cloud computing resource analysis model considering both hot/cold startup and hot/cold shutdown of virtual machines (VMs), and use the M/M/N/oo queuing model to analyze cloud computing platform and acquire accurate performance indicators, such as elasticity indicators, cost indicators, performance indicators, cost-performance ratios, etc. In addition, we establish a multi-objective optimization model to optimize both performance and cost of cloud computing platform. Then the optimal stopping and cost-performance optimization algorithm are applied to obtain the optimal configurations, including the number of hot startup VMs, the system service rate, the hot/cold startup rate of VMs, and the hot/cold shutdown rate. By comparing with existing optimization methods, we demonstrate the superiority of our cost-performance ratio optimization method.

This paper investigates resource provisioning in cyber-physical systems (CPSs) by developing a new definition of anelasticity. A flat semi-dormant multicontroller (FSDMC) model is established on a special type of CPS platform named arbitrated networked control system with dual communication channels. A novel, quantitative, and formal definition of anelasticity for the FSDMC is proposed. A new finite capacity M/M/c queuing system with N-policy and asynchronous multiple working vacations of partial servers is established, and the FSDMC is modeled as a quasi-birth-and-death process to obtain the stationary probability distribution of the system. Based on the queueing model, we quantify various performance indices of the system to build a nonlinear cost-performance ratio (CPR) function. An optimization model is presented to minimize the CPR. A particle swarm optimization (PSO) algorithm is used to find the optimum solution of the optimization model and obtain the optimal configuration values of the system parameters under stability condition. By changing the system parameters, the sensitivity of the system performance indices and the CPR are analyzed, respectively. The unexpected workload varies randomly over time. Thus, an M/M/1/K queue is constructed in a Markovian environment by employing a three-state, irreducible Markov process. In this queue, the conditional average queue length and the probabilities of the three-state process are calculated. Then, the anelasticity value of the system is precisely determined. When the average arrival rate exceeds the average service rate in the queueing system, an optimal CPR unchanged adaptive algorithm based on PSO is designed to dynamically adjust the controller service rate. Extensive numerical results show the usefulness and effectiveness of the proposed techniques and exhibit that the system can maintain elastic invariance in adaptive adjustment parameters.

Recently, the modeling and design of distributed controllers in cyber-physical systems (CPSs), which suffer from messages lost, delay variation, and jitter, has gained lots of research attentions. A special CPS, arbitrated networked control system (ANCS), has been designed for scheduling or arbitrating networks in a control system. In this paper, we propose a novel ANCS with dual communication channels. The proposed ANCS uses a hierarchical flexible time-division multiple access (TDMA)/fixed priority scheduling policy that is based on the event trigger protocol. A flat semi-dormant multicontrollers (FSDMC) model is developed for the proposed ANCS. We then model the FSDMC as an N/(d,c)-M/M/c/K/SMWV queue, and obtain various performance indices. Based on the model, a multiobjective optimization problem is then formulated to minimize the nonlinear energy consumption function and the nominal delay function presented in this study. To resolve the multiobjective optimization problem, a scheduling algorithm based on the multiobjective particle swarm optimization algorithm is proposed to generate the Pareto front and the corresponding nondominated vector sets. An optimal stopping algorithm is also designed to obtain the optimal value of the number of semi-dormant controllers. The optimal values of various parameters of the control system are obtained by using the above nondominated vector sets, and are applied to the proposed ANCS. Extensive numerical results are provided to illustrate the usefulness of the proposed algorithms and the effects of the control system parameters on the optimal policy.