Path planning is one of the important components Unmanned Aerial Vehicle (UAV) mission, and it also key guarantee for successful completion UAV's mission. The traditional path algorithm has certain limitations deficiencies in complex dynamic environment. Aiming at obstacle environment, this paper proposes an improved TD3 algorithm, which enables UAV to complete autonomous through online learning continuous trial error. changes experience pool priority replay, so that agent can distinguish...

As an essential feature of cloud computing, dynamic scalability enables the system to dynamically expand or shrink resources according user needs at runtime. Effectively predicting and optimizing cost performance computing platforms have become one key research challenges in field computing. In this article, quantitatively predict platforms, we propose a resource analysis model considering both hot/cold startup shutdown virtual machines (VMs), use M/M/N/oo queuing analyze platform acquire...

The Transformer has been successfully used in medical image segmentation due to its excellent long-range modeling capabilities. However, patch is necessary when building a class model. This process ignores the tissue structure features within patch, resulting loss of shallow representation information. In this study, we propose Heterogeneous Swin with Multi-Receptive Field (HST-MRF) model that fuses information from different receptive fields solve problem feature caused by segmentation....

Research in medical visual question answering (MVQA) can contribute to the development of computer-aided diagnosis. MVQA is a task that aims predict accurate and convincing answers based on given images associated natural language questions. This requires extracting knowledge-rich feature content making fine-grained understandings them. Therefore, constructing an effective extraction understanding scheme are keys modeling. Existing schemes mainly focus word information, ignoring information...

Recently, the modeling and design of distributed controllers in cyber-physical systems (CPSs), which suffer from messages lost, delay variation, jitter, has gained lots research attentions. A special CPS, arbitrated networked control system (ANCS), been designed for scheduling or arbitrating networks a system. In this paper, we propose novel ANCS with dual communication channels. The proposed uses hierarchical flexible time-division multiple access (TDMA)/fixed priority policy that is based...

Face detection in the classroom environment is basis for student face recognition, sensorless attendance, and concentration analysis. Due to equipment, lighting, uncontrollability of students an unconstrained environment, images include many moving faces, occluded extremely small faces a environment. Since image sent detector will be resized smaller size, information extracted by very limited. This seriously affects accuracy detection. Therefore, this paper proposes adaptive fusion-based...

This article investigates the reliability of a special cyber-physical system (CPS) with an unreliable service and complex boundary behavior. A flat semi-dormant multicontroller (FSDMC) model is constructed on CPS named arbitrated networked control (ANCS) dual channels. In this study, dual-channel ANCS considered as Markov repairable system, which integrates binary state physical device failure multistate information flow. The FSDMC modeled <italic...

Table-based fact verification requests parsing table and statement structure performing numerical logical reasoning. Previous methods may select erroneous programs ignore the interpretability of table-based verification. Thus, we propose a double graph attention network reasoning method based on filtering program-like evidence (DGMFP). In detail, initially obtain tables forms to incorporate semantic symbolic information evidence. Then, construct an with statement–evidence pairs as nodes use...

In object-oriented software testing, a class is considered to be basic unit of testing. The state the objects may cause faults that cannot easily revealed with traditional testing techniques. this paper, we propose new technique for by using event-driven Petri nets (EDPN), which an extended version Nets, one techniques having ability analyze and test behavior interaction between data members member functions in class. We demonstrate how specify specification EDPNs given fault model mutant...

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 special type CPS platform named arbitrated networked control system with dual communication channels. novel, quantitative, and formal anelasticity for the FSDMC proposed. finite capacity M/M/c queuing N-policy asynchronous multiple working vacations partial servers established, modeled as...

A predicate that appears in specification or code can implement a condition. In this paper, we propose Predicate-Driven Petri Net(PDPN), which be applied to analyze and test the behavior for interaction between operators individuals variables expressions. With help of function, dynamic properties PDPN an iterative partition approach PDPNs were investigated. generalized BOR-MI strategy called as Meaning Match Search (MeMS) is developed generate minimal MeMS-constraint set. Using presented...

Boolean expression testing requires certain types of tests for each in program specification or implementation. Fault-based essentially uses a subset the exhaustive test set to detect special faults. A fault-based strategy called constraint true and false point (CTFTP) is proposed. The consists two case generation strategies, namely unique (UCTP) near (NCFP) strategy. An ordered binary label-driven Petri net model presented analyze interaction between transitions literals yield paths...

Research in medical visual question answering (MVQA) can contribute to the development of computeraided diagnosis. MVQA is a task that aims predict accurate and convincing answers based on given images associated natural language questions. This requires extracting knowledge-rich feature content making fine-grained understandings them. Therefore, constructing an effective extraction understanding scheme are keys modeling. Existing schemes mainly focus word information, ignoring information...

The Transformer has been successfully used in medical image segmentation due to its excellent long-range modeling capabilities. However, patch is necessary when building a class model. This process may disrupt the tissue structure images, resulting loss of relevant information. In this study, we proposed Heterogeneous Swin with Multi-Receptive Field (HST-MRF) model based on U-shaped networks for segmentation. main purpose solve problem structural information caused by using transformer...