Distributed Periodic Steady State Kalman Filter

In this paper a distributed implementation for the periodic steady state Kalman filter is proposed. The distributed algorithm has parallel structure and can be implemented using processors in parallel without idle time. The number of processors is equal to the model period. The resulting speedup is also derived. The Finite Impulse Response (FIR) form of the periodic steady state Kalman filter is derived.

Entity recognition of Chinese medical text based on multi-head self-attention combined with BILSTM-CRF

<abstract> <p>Named entities are the main carriers of relevant medical knowledge in Electronic Medical Records (EMR). Clinical electronic medical records lead to problems such as word segmentation ambiguity and polysemy due to the specificity of Chinese language structure, so a Clinical Named Entity Recognition (CNER) model based on multi-head self-attention combined with BILSTM neural network and Conditional Random Fields is proposed. Firstly, the pre-trained language model organically combines char vectors and word vectors for the text sequences of the original dataset. The sequences are then fed into the parallel structure of the multi-head self-attention module and the BILSTM neural network module, respectively. By splicing the output of the neural network module to obtain multi-level information such as contextual information and feature association weights. Finally, entity annotation is performed by CRF. The results of the multiple comparison experiments show that the structure of the proposed model is very reasonable and robust, and it can effectively improve the Chinese CNER model. The model can extract multi-level and more comprehensive text features, compensate for the defect of long-distance dependency loss, with better applicability and recognition performance.</p> </abstract>

Features of driving of the steering wheel driving cars

Existing methods of diagnosing steering can be characterized by low efficiency. For various reasons, both declarative and actual (supported by the equipment) methods, as a rule, have low accuracy and inability to localize faults. The car's built-in diagnostics cannot affect the situation due to the small number of sensors in the steering system. The reasons for the low accuracy of the methods include design features, low availability of components (low maintainability). Difficulties in localization of malfunctions are caused by the structural scheme which is characterized by parallel - consecutive construction. The parameters of diagnostic methods are analyzed, the proposed method is based on the structure of the steering, in the implementation of which test effects are applied to the steered wheels. In total it is necessary to carry out three measurements of backlashes and as a result of mathematical processing of results it becomes possible to localize malfunction in three links of consecutive elements of the steering mechanism or a steering drive. In accordance with this approach, steering is considered as a set of three structures - parallel and two sequential. Rack and pinion steering was used as a model. Here, the parallel structure includes elements of the steering linkage: swing arm, left and right; steering rod, left and right; steering rack - left and right hinges. The sequential structure - left, includes a swing arm, left; steering rod, left; steering rack hinge, left; steering gear, steering shaft, steering wheel. Accordingly, the sequential structure of the right includes similar elements with the attribute "right". The structure of the steering play is considered in a similar way. As a result, it becomes possible to obtain a transformed system of three algebraic equations connecting clearances in three groups of mates and backlashes in parallel and two sequential steering structures. To measure the backlash, the turntables of the BOSCH FWA 4410 stand were used; in another version, the wheels were hung out. As a result of tests carried out on VW GOLF, VW PASSAT and RENAULT 25 vehicles with significant mileage, data was obtained indicating the need for technical interventions on localized groups of interfaces.

Development and Analysis of a Parallel Structure Mechanism with Slotted Arcuate Guides and a Constant Entry Point

The development and study of new mechanisms of a parallel structure ensuring the constancy of the point of instrument entry into the working area is an urgent task of surgery and scientific medicine. A structural synthesis of a parallel structure mechanism designed for robotic laparoscopic surgery has been performed. The proposed mechanism is equipped with arcuate guides with slots installed on the base perpendicular to each other. At the intersection of the slots, there is a bushing through which a straight shaft connected to the output link moves linearly. This arrangement provides a constant entry point, which allows the developed mechanism to be used in laparoscopic surgery and studies of plasma properties. For this mechanism, the inverse problem of positions has been solved, the working area has been determined, and a 3D model has been developed.

Energy Management Strategy Based on a Novel Speed Prediction Method

Vehicle speed prediction can obtain the future driving status of a vehicle in advance, which helps to make better decisions for energy management strategies. We propose a novel deep learning neural network architecture for vehicle speed prediction, called VSNet, by combining convolutional neural network (CNN) and long-short term memory network (LSTM). VSNet adopts a fake image composed of 15 vehicle signals in the past 15 s as model input to predict the vehicle speed in the next 5 s. Different from the traditional series or parallel structure, VSNet is structured with CNN and LSTM in series and then in parallel with two other CNNs of different convolutional kernel sizes. The unique architecture allows for better fitting of highly nonlinear relationships. The prediction performance of VSNet is first examined. The prediction results show a RMSE range of 0.519–2.681 and a R2 range of 0.997–0.929 for the future 5 s. Finally, an energy management strategy combined with VSNet and model predictive control (MPC) is simulated. The equivalent fuel consumption of the simulation increases by only 4.74% compared with DP-based energy management strategy and decreased by 2.82% compared with the speed prediction method with low accuracy.

Characterization of a G-quadruplex from hepatitis B virus and its stabilization by binding TMPyP4, BRACO19 and PhenDC3

Specific guanine rich nucleic acid sequences can form non-canonical structures, like the four stranded G-quadruplex (GQ). We studied the GQ-forming sequence (named HepB) found in the genome of the hepatitis B virus. Fluorescence-, infrared- and CD-spectroscopy were used. HepB shows a hybrid form in presence of K+, but Na+, Li+, and Rb+ induce parallel structure. Higher concentrations of metal ions increase the unfolding temperature, which was explained by a short thermodynamic calculation. Temperature stability of the GQ structure was determined for all these ions. Na+ has stronger stabilizing effect on HepB than K+, which is highly unusual. The transition temperatures were 56.6, 53.8, 58.5 and 54.4 °C for Na+, K+, Li+, and Rb+ respectively. Binding constants for Na+ and K+ were 10.2 mM and 7.1 mM respectively. Study of three ligands designed in cancer research for GQ targeting (TMPyP4, BRACO19 and PhenDC3) showed unequivocally their binding to HepB. Binding was proven by the increased stability of the bound form. The stabilization was higher than 20 °C for TMPyP4 and PhenDC3, while it was considerably lower for BRACO19. These results might have medical importance in the fight against the hepatitis B virus.

A position control-based approach to haptic rendering of stiff objects using piece-wise linear model

In traditional force rendering approaches, it is quite popular to model a virtual stiff wall as a spring-damper system to compute the interaction force, which can easily lead to unstable behavior. In this paper, we present an approach to ensure no penetration into the wall by position control. The approach approximates the nonlinear model of a 6-DOF parallel-structure haptic device by a piece-wise linear model to improve the performance compared with a controller designed from a one-point linearized model in haptic rendering. A simulation-based performance comparison study shows that the new controller can render higher stiffness than the previous solution.

On the Parallel Kinematics of the FET Stretching Press in the Stretch Forming Operations in the Manufacture of Parts with Complex Spatial Geometry

The kinematics of a stretching press mechanism of a parallel structure are investigated. The kinematic dependences aimed at setting the spatial position of the working elements of a cross-stretching press of FET type, which allow positioning its jaws at any time during the technological operations, are obtained. The forward and inverse control kinematics problems of the jaw mechanism stretching press are solved