Deep Learning Based Data Fusion for Sensor Fault Diagnosis and Tolerance in Autonomous Vehicles

Environmental perception is one of the key technologies to realize autonomous vehicles.Autonomous vehicles are often equipped with multiple sensors to form a multi-source environmental perception system.Those sensors are very sensitive to light or background conditions,which will introduce a variety of global and local fault signals that bring great safety risks to autonomous driving system during long-term running.In this paper,a real-time data fusion network with fault diagnosis and fault tolerance mechanism is designed.By introducing prior features to realize the lightweight network,the features of the input data can be extracted in real time.A new sensor reliability evaluation method is proposed by calculating the global and local confidence of sensors.Through the temporal and spatial correlation between sensor data,the sensor redundancy is utilized to diagnose the local and global confidence level of sensor data in real time,eliminate the fault data,and ensure the accuracy and reliability of data fusion.Experiments show that the network achieves state-of-the-art results in speed and accuracy,and can accurately detect the location of the target when some sensors are out of focus or out of order.The fusion framework proposed in this paper is proved to be effective for intelligent vehicles in terms of real-time performance and reliability.

Nonsynchronized state estimation of uncertain discrete-time piecewise affine systems

This paper investigates the problem of robust H-infinity state estimation for a class of uncertain discretetime piecewise affine systems where state space instead of measurable output space partitions are assumed so that the filter implementation may not be synchronized with plant state trajectory transitions.Based on a piecewise quadratic Lyapunov function combined with S-procedure and some matrix inequality convexifying techniques,two different approaches are developed to the robust filtering design for the underlying piecewise affine systems.It is shown that the filter gains can be obtained by solving a set of linear matrix inequalities (LMIs).Finally,a simulation example is provided to illustrate the effectiveness of the proposed approaches.

Robust fault detection for discrete-time Markovian jump systems with mode-dependent time-delays

这篇论文为类ofdiscrete时间 Markovian 调查一个差错察觉问题跳有围住标准的不确定性和模式依赖者 time-delays.Attention 的系统集中于基于它的参数矩阵也就是说依赖于系统模式的过滤器构造剩余生成器差错察觉过滤器是 aMarkovian 跳系统也。差错察觉过滤器的设计被使用H无穷控制理论归结为 H-infinityfiltering 问题，它能保证剩余和差错(更通常加权的差错)之间的差别在提高剩余的坚韧性到为错误建模的上下文尽可能小，控制输入和未知输入。为上述过滤器的存在的足够的状况借助于线性矩阵不平等被建立，它能被使用标准数字软件乐意地解决。一个数字例子被给说明建议方法的可行性。

Metabolic Profiling in Banana Pseudo-Stem Reveals a Diverse Set of Bioactive Compounds with Potential Nutritional and Industrial Applications

Banana(Musa spp.)is an ancient and popular fruit plant with highly nutritious fruit.The pseudo-stem of banana represents on average 75%of the total dry mass but its valorization as a nutritional and industrial by-product is limited.Recent advances in metabolomics have paved the way to understand and evaluate the presence of diverse sets of metabolites in different plant parts.This study aimed at exploring the diversity of primary and secondary metabolites in the banana pseudo-stem.Hereby,we identified and quantified 373 metabolites from a diverse range of classes including,alkaloids,flavonoids,lipids,phenolic acids,amino acids and its derivatives,nucleotide and its derivatives,organic acids,lignans and coumarins,tannins,and terpene using the widely-targeted metabolomics approach.Banana pseudo-stem is enriched in metabolites for utilization in the food industry(L-lysine and L-tryptophan,L-glutamic acid,Phenylalanine,Palmitoleic acid,α-Linolenic acid,and Lauric acid,and Adenine)and pharmaceutical industry(Guanosine and Cimidahurinine,Bergapten,Coumarins,Procyanidin A2,Procyanidin B1,Procyanidin B3,Procyanidin B2,and Procyanidin B4,Asiatic acid).The metabolome of banana pseudo-stem with integration across multiomics data may provide the opportunity to exploit the rich metabolome of banana pseudo-stem for industrial and nutritional applications.

Aldolase A promotes proliferation and G1/S transition via the EGFR/MAPK pathway in non-small cell lung cancer

Background:Our previous study demonstrated that aldolase A(ALDOA)is overexpressed in clinical human lung squamous cell carcinoma and that ALDOA promotes epithelial-mesenchymal transition and tumorigenesis.The pre-sent study aimed to explore the function of ALDOA in the modulation of non-small cell lung cancer(NSCLC)prolifera-tion and cell cycle progression and the potential mechanism.Methods:ALDOA was knocked down by short hairpin RNA in H520 and H1299 cells.ALDOA was overexpressed with vectors carrying the full-length ALDOA sequence in H1299 and H157 cells.The proliferation capacities were assessed with immunohistochemical staining,Cell Counting Kit-8 and colony formation assays.The cell cycle distribution was examined by flow cytometry,and molecular alterations were determined by western blotting.Cell synchronization was induced with nocodazole.The stability of cyclin D1 mRNA was tested.The pyruvate kinase M2 and ALDOA protein distributions were examined.Aerobic glycolysis was evaluated with Cell Titer-Glo assay,glucose colorimetric assay and lactate colorimetric assay.Results:ALDOA knockdown inhibited the proliferation and G1/S transition in H520 cells.Conversely,ALDOA over-expression promoted the proliferation and G1/S transition in H157 cells.The cell cycle synchronization assay showed that ALDOA expression increased in the G1 phase and G1/S transition.Furthermore,ALDOA knockdown reduced cyclin D1 expression by regulating epidermal growth factor receptor/mitogen-activated protein kinase(EGFR/MAPK)pathway.Similar results were found in H1299 and H157 cells.The inhibition of mitogen-activated protein kinase kinase 1/2 prompted the nuclear distribution of ALDOA.Additionally,ALDOA knockdown reduced nuclear distribution of PKM2,the extracellular lactate and intracellular adenosine triphosphate concentrations and elevated the extracellular glucose concentration.Conclusions:ALDOA contributed to activation of the EGFR/MAPK pathway,thus promoting cyclin D1 expression and enhancing proliferation and G1/S transition in NSCLC.Additionally,ALDOA facilitated NSCLC aerobic glycolysis.

Synchronising second-order multi-agent systems under dynamic topology via reference model-based algorithm

This paper revisits the synchronisation problem for second-order multi-agent systems(MASs)under dynamically changing communication topology.By employing the reference model-based synchronisation algorithm,it is finally shown that synchronisation for both the position and velocity states can be achieved if the union of the communication topologies has a directed spanning tree frequently enough.This extends the existing results obtained for second-order MASs which exploits mild communication topology condition guaranteeing the synchronisation to a more general case.Convergence analysis is successfully performed by exploiting the product properties of row-stochastic matrices,which can also provide us with an estimate the convergence rate towards the synchronisation.