Search

Abstract: Plenty of applications have to take into account dynamic factors. However traditional machine learning paradigm lacks adaptivity to deal with sudden change of the unknown underlying target function because its common assumption: training and test samples are drawn from the same feather space (input space) and the same distribution (target function). The difficulty is that it is short of new training data when change happens. A novel method, which creates virtual examples based on semi-empirical formula, is proposed to make up the loss of new samples when the target function alters suddenly. This method is tested and verified to perform fairly well on the robotic belt grinding platform, which suffers for dynamic factors when modeling the material removal rate. The experiments also prove that virtual examples as a kind of incorporation of prior knowledge are able to mind out the information contained in the semi-empirical formula for the learner. [Copyright &y& Elsevier]

Traditional precise orbit determination (POD) for low Earth orbit (LEO) satellites relies on observations from ground stations and onboard receivers. Although the accuracy can reach centimeter level, there are still problems such as insufficient autonomous operation capability. The inter-satellite link (ISL) is a link used for communication between satellites and has a function of dual-way ranging. Numerous studies have shown that observational data using ISLs can be adopted for POD of navigation satellites. In this contribution, we mainly focus on LEO satellites POD with ISLs. First, we design LEO constellations with different numbers of satellites and ISL measurements, based on which the constellations are simulated. Then rough tests of POD using different link topologies are carried out. The results show that in the 60-LEO constellation the average 3-dimensional (3D) orbital errors are 0.112 m using "4-connected" link topology with constant 4 links per satellite and 0.069 m using "all-connected" link topology with theoretically maximum numbers of links. After that, we carry out refined POD experiments with several sets of satellite numbers and different observation accuracy. The results show the higher link ranging accuracy and the more numbers of links bring higher POD precision. POD with ISLs gets bad performance in the case of center of gravity reference when link ranging accuracy is poor and numbers of links are small. When the link accuracy is 40 cm, average 3D orbital errors of 60-LEO constellation are 0.358 m, which can only meet the demand of autonomous navigation. With the constraint of the right ascension of the ascending node (RAAN), POD using ISLs reaches an extremely high precision when adopting a spatial reference provided by navigation satellites. For 120-LEO constellation, the average 3D orbital errors are 0.010 m; for 192-LEO constellation, the errors are 0.006 m. [ABSTRACT FROM AUTHOR]

In Xinjiang, China, esophageal carcinoma has a high incidence in Kazak and Uighur populations. MicroRNA (miR)-21 and miR-375 are related to esophageal carcinoma. This study thus investigated their potencials in early diagnosis and prognosis in Kazak and Uighur populations, to provide evidences for serum markers of esophageal cancer. A total of 126 Kazak or Uighur esophageal cancer patients were enrolled as the disease group, along with 86 local Han patients as disease control cohort, and 80 healthy Kazak or Uighur individuals. MiRNA expression was detected by in situ hybridization in tissues and by qRT-PCR in serum. ROC approach was used to evaluate the diagnostic value of miRNA on esophageal carcinoma. Cox analysis was performed to screen factors governing prognosis. MiR-21 level was significantly elevated in both tissue and serum samples of esophageal cancer patients, while miR-375 was down-regulated. Such difference was more potent in disease group compared to disease control group. MiR expression was correlated with infiltration depth, TNM stage, vascular invasion, and lymph node metastasis. Elevated expression of miR-21 reduced the sensitivity of radio-therapy, and increased recurrence frequency. The diagnostic value of single assay for miR-21 or miR-375 was lower than the combined assay (AUC=0.812 or 0.739 vs. 0.858). They also affected patient prognosis (OR=1.53 or 0.652). MiR-21 and miR-375 presented abnormal expression in Kazak or Uighur esophageal carcinoma patients and were independent factors affecting prognosis. The combined assay of miR-21 and miR-375 may help to make early diagnosis of esophageal cancer.