Motion coordination control of planar 5R parallel quadruped robot based on SCPL-CPG

The parallel leg of the quadruped robot has good structural stiffness, accurate movement, and strong bearing capacity, but it is complicated to control. To solve this problem, a series connection of parallel legs (SCPL) was proposed, as well as a control strategy combined with the central pattern generator (CPG). With the planar 5R parallel leg as the research object, the SCPL analysis method was used to analyze the leg structure. The topology of CPG network was built with the Hopf oscillator as the unit model, and the CPG was the core to model the robot control system. By continuously adjusting the parameters in the CPG control system and changing the connection weight, and the smooth transition between gaits was realized. The simulation results show that the SCPL analysis method can be effectively used in the analysis of parallel legs, and the control system can realize the smooth transition between gaits, which verifies the feasibility and effectiveness of the proposed control strategy.

Sensitivity analysis of design parameters of envelope enclosure performance in the dry-hot and dry-cold areas

In a bid to quantify the sensitivity of envelope enclosure’s design parameters in the dry-hot and dry-cold areas and to provide a reference for the local building performance design, this paper uses ANN modelling which combined with the improved Garson algorithm to calculate the connection weight sensitivity (CWS), the first-order sensitivity (RBD-S1 and DMIM-S1) and the global sensitivity (DMIM-delta) of the design parameters. These parameters were calculated by using different methods in SALib. Through the verification and analysis of the sensitive result, the applicability of the CWS and DMIM-delta was confirmed. Among the design parameters involved in this study, the sum of the sensitive values of S-D, S-N and S-A exceeds 60% in each performance label, and the sum of the sensitive values of WWR_S and WWR_N exceeds 20%. The performance design of envelope enclosure in this area requires applying reasonable shading components and appropriate optimisation of the North and South of WWR. After the sensitivity analysis process, the calculation efficiency of the model can be improved as far as possible without reducing the accuracy of the model in the later simplified calculation and multi-objective optimisation. The building performance simulation model has a high degree of non-linearity, and the interpretability of the model can be enhanced through the sensitivity analysis process. Although the internal calculation process is unknowable, the perception of the results caused by the input parameters is significantly enhanced.

Elucidating the neural mechanisms of Learning-to-Learn

Learning-to-learn, a progressive acceleration of learning while solving a series of similar problems, represents a core process of knowledge acquisition that draws attention in both neuroscience and artificial intelligence. To investigate its underlying brain mechanism, we trained a recurrent neural network model on arbitrary sensorimotor mappings. The network displayed an exponential speedup in learning. The neural substrate of a schema emerges within a low-dimensional subspace of population activity. Its reuse in new problems facilitates learning by limiting connection weight changes. Since the population trajectory of a recurrent network produces behavior, learning is determined by the vector field changes. We propose a novel analysis of weight-driven vector field changes, which showed that novel stimuli in new problems can distort the schema representation. Weight changes eliminate such distortions and improve the invariance of the reused representations in future learning. The accumulation of such weight changes across problems underlies the learning-to-learn dynamics.

Detection of Oil Spill Using SAR Imagery Based on AlexNet Model

Synthetic aperture radar (SAR) plays an irreplaceable role in the monitoring of marine oil spills. However, due to the limitation of its imaging characteristics, it is difficult to use traditional image processing methods to effectively extract oil spill information from SAR images with coherent speckle noise. In this paper, the convolutional neural network AlexNet model is used to extract the oil spill information from SAR images by taking advantage of its features of local connection, weight sharing, and learning for image representation. The existing remote sensing images of the oil spills in recent years in China are used to build a dataset. These images are enhanced by translation and flip of the dataset, and so on and then sent to the established deep convolutional neural network for training. The prediction model is obtained through optimization methods such as Adam. During the prediction, the predicted image is cut into several blocks, and the error information is removed by corrosion expansion and Gaussian filtering after the image is spliced again. Experiments based on actual oil spill SAR datasets demonstrate the effectiveness of the modified AlexNet model compared with other approaches.

Classification of Bidikmisi Scholarship Acceptance using Neural Network Based on Hybrid Method of Genetic Algorithm

A Neural network is a series of algorithms that endeavours to recognize underlying relationships in a set of data through processes that mimic the way human brains operate. In the case of classification, this method can provide a fit model through various factors, such as the variety of the optimal number of hidden nodes, the variety of relevant input variables, and the selection of optimal connection weights. One popular method to achieve the optimal selection of connection weights is using a Genetic Algorithm (GA), the basic concept is to iterate over Darwin's evolution. This research presents the Neural Network method with the Backpropagation Neural Network (BPNN) and the combined method of BPNN with GA, where GA is used to initialize and optimize the connection weight of BPNN. Based on accuracy value, the BPNN method combined with GA provides better classification, which is 90.51%, in the case of Bidikmisi Scholarship classification in East Java.

Pruning Multilayered ELM Using Cholesky Factorization and Givens Rotation Transformation

Extreme learning machine is originally proposed for the learning of the single hidden layer feedforward neural network to overcome the challenges faced by the backpropagation (BP) learning algorithm and its variants. Recent studies show that ELM can be extended to the multilayered feedforward neural network in which the hidden node could be a subnetwork of nodes or a combination of other hidden nodes. Although the ELM algorithm with multiple hidden layers shows stronger nonlinear expression ability and stability in both theoretical and experimental results than the ELM algorithm with the single hidden layer, with the deepening of the network structure, the problem of parameter optimization is also highlighted, which usually requires more time for model selection and increases the computational complexity. This paper uses Cholesky factorization strategy and Givens rotation transformation to choose the hidden nodes of MELM and obtains the number of nodes more suitable for the network. First, the initial network has a large number of hidden nodes and then uses the idea of ridge regression to prune the nodes. Finally, a complete neural network can be obtained. Therefore, the ELM algorithm eliminates the need to manually set nodes and achieves complete automation. By using information from the previous generation’s connection weight matrix, it can be evitable to re-calculate the weight matrix in the network simplification process. As in the matrix factorization methods, the Cholesky factorization factor is calculated by Givens rotation transform to achieve the fast decreasing update of the current connection weight matrix, thus ensuring the numerical stability and high efficiency of the pruning process. Empirical studies on several commonly used classification benchmark problems and the real datasets collected from coal industry show that compared with the traditional ELM algorithm, the pruning multilayered ELM algorithm proposed in this paper can find the optimal number of hidden nodes automatically and has better generalization performance.

Dual mass MEMS gyroscope temperature drift compensation based on TFPF-MEA-BP algorithm

Purpose To reduce the influence of temperature on MEMS gyroscope, this paper aims to propose a temperature drift compensation method based on variational modal decomposition (VMD), time-frequency peak filter (TFPF), mind evolutionary algorithm (MEA) and BP neural network. Design/methodology/approach First, VMD decomposes gyro’s temperature drift sequence to obtain multiple intrinsic mode functions (IMF) with different center frequencies and then Sample entropy calculates, according to the complexity of the signals, they are divided into three categories, namely, noise signals, mixed signals and temperature drift signals. Then, TFPF denoises the mixed-signal, the noise signal is directly removed and the denoised sub-sequence is reconstructed, which is used as training data to train the MEA optimized BP to obtain a temperature drift compensation model. Finally, the gyro’s temperature characteristic sequence is processed by the trained model. Findings The experimental result proved the superiority of this method, the bias stability value of the compensation signal is 1.279 × 10–3°/h and the angular velocity random walk value is 2.132 × 10–5°/h/vHz, which is improved compared to the 3.361°/h and 1.673 × 10–2°/h/vHz of the original output signal of the gyro. Originality/value This study proposes a multi-dimensional processing method, which treats different noises separately, effectively protects the low-frequency characteristics and provides a high-precision training set for drift modeling. TFPF can be optimized by SEVMD parallel processing in reducing noise and retaining static characteristics, MEA algorithm can search for better threshold and connection weight of BP network and improve the model’s compensation effect.

$H_\infty$ State Estimation of Delayed Recurrent Memristive Neural Networks: continuous-time case and discrete-time case

This paper investigates the problem of $H_\infty$ state estimation of delayed recurrent memristive neural networks (DRMNNs) with both continuous-time and discrete-time cases. By utilizing Lyapunov-Krasovskii functional (LKF) and linear matrix inequalities (LMIs), two criterions are provided to guarantee the asymptotically stable of the estimation error systems with a $H_\infty$ performance. The connection weight parameters of DRMNNs are dealed with logical switching signals, which greatly reduces the computational complexity. The given conditions can be easily checked by solving LMIs, the obtained theoretical results are supported demonstrated by two numerical examples.

Adaptive Rewiring in Weighted Networks Shows Specificity, Robustness, and Flexibility

Brain network connections rewire adaptively in response to neural activity. Adaptive rewiring may be understood as a process which, at its every step, is aimed at optimizing the efficiency of signal diffusion. In evolving model networks, this amounts to creating shortcut connections in regions with high diffusion and pruning where diffusion is low. Adaptive rewiring leads over time to topologies akin to brain anatomy: small worlds with rich club and modular or centralized structures. We continue our investigation of adaptive rewiring by focusing on three desiderata: specificity of evolving model network architectures, robustness of dynamically maintained architectures, and flexibility of network evolution to stochastically deviate from specificity and robustness. Our adaptive rewiring model simulations show that specificity and robustness characterize alternative modes of network operation, controlled by a single parameter, the rewiring interval. Small control parameter shifts across a critical transition zone allow switching between the two modes. Adaptive rewiring exhibits greater flexibility for skewed, lognormal connection weight distributions than for normally distributed ones. The results qualify adaptive rewiring as a key principle of self-organized complexity in network architectures, in particular of those that characterize the variety of functional architectures in the brain.

Predicción basada en redes neuronales de la resistencia a compresión a los 7 días de los UHPC incorporando SCM

RESUMEN El hormigón de ultra alto rendimiento (UHPC) es un tipo de hormigón de alta tecnología que exhibe excelentes propiedades mecánicas y de durabilidad. En los últimos años, el uso de materiales cementantes suplementarios (SCM) como sustitución parcial del humo de cemento y sílice ha sido objeto de gran interés por parte de la comunidad científica para reducir los altos costos y la huella de carbono del UHPC. Algunas sus aplicaciones, como el refuerzo sísmico de estructuras existentes no dúctiles, requieren del desarrollo de resistencias tempranas. Sin embargo, la sustitución del cemento y humo de sílice puede modificar de algunas propiedades, como las resistencias iniciales de UHPC. Por otro lado, el uso de SCM produce un material altamente complejo, siendo más difícil comprender el efecto de cada componente y sus interacciones en el desarrollo de resistencias tempranas en el hormigón. Este estudio tiene como objetivo desarrollar un modelo de redes neuronales artificiales (ANN) para predecir la resistencia a la compresión a los siete días del UHPC, pudiendo incorporar varios SCM como el humo de sílice, ceniza volante, escoria granulada de alto horno, polvo de vidrio reciclado, ceniza de cascarilla de arroz, residuo de catalizador de craqueo catalítico fluido, metacaolín, carbonato cálcico pulverizado, además de filler mineral como el polvo de cuarzo. Para el desarrollo del modelo de una sola capa oculta se usaron 523 datos de investigaciones publicadas. Además, el modelo también fue validado mediante el uso de trabajos experimentales. Finalmente, el algoritmo Connection-Weight-Approach (CWA) se utilizó para analizar las relaciones entre los componentes del UHPC y la resistencia a la compresión a los siete días. Los resultados señalaron que el modelo ANN es un modelo eficiente para predecir la resistencia a la compresión a los 7 días del UHPC incluso cuando se incorporan SCM.