Search

Experience replay can reuse past experience to update target policy and improve the utilization of samples,which has become an important component of deep reinforcement learning.Prioritized experience replay performs selective sampling based on experience replay to use samples more efficiently.Nevertheless,the current prioritized experience replay methods will reduce the diversity of samples sampled from the experience buffer,causing the neural network to converge to the local optimum.To tackle the above issue,a novel method named exploration and exploitation balanced experience replay (E3R) is proposed to ba-lances exploration and utilization.This method can comprehensively consider the exploration utility and utilization utility of the samples,and sample according to the weighted sum of two similarities.One of them is the similarity between the behavior strategy and the target strategy in the same state of action,and the other is the similarity between the current state and the past state.Besides,the E3R is combined with the policy gradient algorithm soft actor-critic and the value function algorithm deep Q lear-ning,and experiments are carried out on the suite of OpenAI gym tasks.Experimental results show that,compared to traditional random sampling and sequential differential priority sampling,E3R can achieve faster convergence speed and higher cumulative return.

Using glucose as the reducing agent, the nano-silver-graphene composite material (Ag/GR) was synthesized in situ by electroless plating, the structural morphology of the material was characterized and analyzed by X-ray diffraction (XRD), X-ray energy dispersive spectroscopy (EDS), X-ray photoelectron spectroscopy (XPS), scanning electron microscope (SEM) and Fourier infrared spectrometry (FTIR). The results show that the loading morphology of silver on the graphene surface is the single state that meets the expectation. The average particle size of AgNPs is about 21 nm. At the same time, the electrochemical response of ascorbic acid (AA) on Ag/GR/GCE electrochemical sensors was studied using the electrochemical test methods of cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), linear sweep voltammetry (LSV) and differential pulse voltammetry (DPV). The electrochemical test results show that the Ag/GR composite has the highest response electrochemical signal of 212.9 μA and the lowest resistance of charge transfer of 90.5 Ω. The peak current is about twice that of the graphene electrode (110 μA) and about five times that of the glassy carbon electrode (42.5 μA). Due to the good synergistic effect of AgNPs and graphene, it has obvious electrocatalytic activity for AA. And the anode peak current and concentration show positive correlation linear change trend in the range of 5-120 μmol/L. However, the anode peak current of AA is highly correlated with the natural logarithm of the concentration range from 50 μmol/L to 120 μmol/L. The low limits of detection is 0.06 μmol/L.

Conductive hydrogel fibers exhibit great potential in soft robots, bioelectronics, and human–machine interfaces due to the unique combination of electrical conductivity, high water content, tissue‐like mechanical properties, and 1D structure. Despite significant advances in hydrogel technologies, the typical conductive hydrogel fibers show low conductivity (<10 S cm−1), weak mechanical properties, and water stability, which makes it challenging to satisfy the requirements of practical applications. Here, a facile strategy is proposed to construct hydrogel fibers with ultrahigh conductivity and toughness by exploiting the synergistic effects of freezing‐thawing, salting‐out, and drying‐annealing. The continuous phase separation induced by the combined processes results in hierarchical structures, promoting the formation of interconnected conductive networks and increasing the fiber's crystallinity and crystal domain size. The prepared conductive hydrogel fibers exhibited ultrahigh conductivity (≈958 S cm−1), excellent mechanical properties (strength (≈6.2 MPa), stretchability (>300%), and toughness (≈10 MJ m−2)), high water content (≈75%), outstanding water stability, and fatigue resistance properties. In addition, the processibility of conductive hydrogel yarns and fabrics are demonstrated and their potential application in bioelectronics. Overall, this work presents a preparation strategy for conductive hydrogel fibers, which will facilitate the advancement of soft electronics and may inspire structural construction in other polymers. [ABSTRACT FROM AUTHOR]

Composite tubes are widely adopted in engineering protections owning to desirable passive energy absorption. With the aim of enhancing energy absorption and making full use of structural materials, a novel tube inspired by the structure of cattle horn has been designed. It consisted of a shell composed of nested tubes and a core composed of evenly distributed tubes, which are connected by beams. The energy absorption properties of bionic tube, six-cell tube, four-cell tube and circular tube were compared by numerical models verified by axial compression. The factors affecting mechanical responses and energy absorption capacities of bionic tube were analysed. Compared to the six-cell tube, the bionic tube exhibited a 104.91% increase in peak load, a 239.51% increase in mean load and an 83.39% increase in specific energy absorption. Under the premise of ensuring the energy absorption, mean load and crushing force efficiency, the mass and peak load can be controlled by limiting the wall thickness and height of core. The proposed bionic tube had the potential to be applied in the automobile and aerospace industry. [ABSTRACT FROM AUTHOR]