Reinforcement learning (RL) is a promising direction in automated parking systems (APSs), as integrating planning and tracking control using RL can potentially maximize the overall performance. ...However, commonly used model-free RL requires many interactions to achieve acceptable performance, and model-based RL in APS cannot continuously learn. In this paper, a data-efficient RL method is constructed to learn from data by use of a model-based method. The proposed method uses a truncated Monte Carlo tree search to evaluate parking states and select moves. Two artificial neural networks are trained to provide the search probability of each tree branch and the final reward for each state using self-trained data. The data efficiency is enhanced by weighting exploration with parking trajectory returns, an adaptive exploration scheme, and experience augmentation with imaginary rollouts. Without human demonstrations, a novel training pipeline is also used to train the initial action guidance network and the state value network. Compared with path planning and path-following methods, the proposed integrated method can flexibly co-ordinate the longitudinal and lateral motion to park a smaller parking space in one maneuver. Its adaptability to changes in the vehicle model is verified by joint Carsim and MATLAB simulation, demonstrating that the algorithm converges within a few iterations. Finally, experiments using a real vehicle platform are used to further verify the effectiveness of the proposed method. Compared with obtaining rewards using simulation, the proposed method achieves a better final parking attitude and success rate.
In automatic parking motion planning, multi-objective optimization including safety, comfort, parking efficiency, and final parking performance should be considered. Most of the current research ...relies on the parking data from expert drivers or prior knowledge of humans. However, it is challenging to obtain a large amount of high-quality expert drivers' data. Furthermore, expert drivers' data or prior knowledge of humans does not guarantee an optimal multi-objective parking performance. In this paper, we propose a model-based reinforcement learning method that learns parking policy of the data, by executing the data generation, data evaluation, and training network, iteratively. The trained network is used to guide the data generation cycle in the subsequent iteration. Based on this proposed method, we can get rid of human experience largely and learn parking strategies autonomously and quickly. The learned strategies ensure the multi-objective optimality of above requirements in the parking process. First, an environment model that approximates the actual environment is established, and the learning efficiency is accelerated through the simulated interaction between the agent and the environment model. To make the system independent of expert data or prior knowledge, a data generation algorithm combining Monte Carlo Tree Search (MCTS) and longitudinal and lateral policies is proposed. Then, to meet the multi-objective optimal demands mentioned above, a reward function is constructed to evaluate and filter the parking data. Finally, a neural network is used to learn the parking strategy from the filtered data. From the real vehicle test benchmarked with a mass-produced parking system, the proposed method is found to achieve better parking efficiency and lower requirements for start parking posture, thereby verifying the algorithm's superiority.
The acoustic imaging (AI) technique could map the position and the strength of the sound source via the signal processing of the microphone array. Conventional methods, including far-field ...beamforming (BF) and near-field acoustic holography (NAH), are limited to the frequency range of measured objects. A method called Bregman iteration based acoustic imaging (BI-AI) is proposed to enhance the performance of the two-dimensional acoustic imaging in the far-field and near-field measurements. For the large-scale ℓ1 norm problem, Bregman iteration (BI) acquires the sparse solution; the fast iterative shrinkage-thresholding algorithm (FISTA) solves each sub-problem. The interpolating wavelet method extracts the information about sources and refines the computational grid to underpin BI-AI in the low-frequency range. The capabilities of the proposed method were validated by the comparison between some tried-and-tested methods processing simulated and experimental data. The results showed that BI-AI separates the coherent sources well in the low-frequency range compared with wideband acoustical holography (WBH); BI-AI estimates better strength and reduces the width of main lobe compared with ℓ1 generalized inverse beamforming (ℓ1-GIB).
Gut bacteria consists of 150 times more genes than humans that are vital for health. Several studies revealed that gut bacteria are associated with disease status and influence human behavior and ...mentality. Whether human brain injury alters the gut bacteria is yet unclear, we tested 20 fecal samples from patients with cerebral intraparenchymal hemorrhage and corresponding healthy controls through metagenomic shotgun sequencing. The composition of patients' gut bacteria changed significantly at the phylum level; Verrucomicrobiota was the specific phylum colonized in the patients' gut. The functional alteration was observed in the patients' gut bacteria, including high metabolic activity for nutrients or neuroactive compounds, strong antibiotic resistance, and less virulence factor diversity. The changes in the transcription and metabolism of differential species were more evident than those of the non-differential species between groups, which is the primary factor contributing to the functional alteration of patients with cerebral intraparenchymal hemorrhage.
The clinical significance of immune-related antigen CD58 in gliomas remains uncertain. The aim of this study was to examine the clinical importance and possible core related genes of CD58 in gliomas.
...Pan-cancer analysis was to observe the association between CD58 and different tumors, glioma RNA sequencing data and clinical sample analyses were used to observe the relationship between CD58 and glioma, shRNA interference models were to observe the impact of CD58 on glioma cell function, and four glioma datasets and two online analysis platforms were used to explore the core related genes affecting the correlation between CD58 and glioma.
High CD58 expression was associated with worse prognosis in various tumors and higher malignancy in glioma. Down regulation of CD58 expression was linked to decreased proliferation, increased apoptosis, and reduced metastasis in glioma cells. The pathways involved in CD58-related effects were enriched for immune cell adhesion and immune factor activation, and the core genes were CASP1, CCL2, IL18, MYD88, PTPRC, and TLR2. The signature of CD58 and its core-related genes showed superior predictive power for glioma prognosis.
High CD58 expression is correlated with more malignant glioma types, and also an independent risk factor for mortality in glioma. CD58 and its core-related genes may serve as novel biomarkers for diagnosing and treating glioma.
•The highlights of this study are:•This paper observed the clinical significance of CD58 in glioma from three different aspects, including bioinformatics analysis, clinical tissue sample analysis, and in vitro cell experiments. And the results were consistent.•This paper constructed a network of CD58-related genes.•This paper also developed a signature of CD58 & core related genes and confirmed its ability to predict glioma mortality risk.
The use of biocompatible ionic liquids (bio-ILs) as a "green" solvent with uniquely tunable cations and anions, specific selectivity, and many other desirable physicochemical properties in ...biocatalysis has shown great promises to address many challenges. A plethora of studies have demonstrated ILs possess a range of exceptional capabilities in coordinating biocatalysis such as stabilizing enzyme structure, enhancing enzyme-substrate interaction, increasing substrate solubility and product separation/removal, and improving enzyme immobilization/recycling. Therefore, bio-ILs have been explored for biocatalysis in a wide range of homogenous as well as multiphasic systems including biphasic, multiphasic, microemulsion, and whole-cell systems. This review provides an overview of IL-enabled biocatalysis including enzymatic and whole-cell reactions in homogeneous and multiphasic systems. First, the properties of IL concentration, specific ion effect, pH, viscosity, alkyl chain length, hydrophobicity, and polarity on the structures and activities of enzymes are discussed together with characterization techniques to elucidate the mechanisms of IL-enzyme interactions. Second, while homogenous biocatalysis systems containing ILs are briefly introduced and summarized, particular attention is given to multiphasic systems. Furthermore, recent progress in using bio-ILs to modify emerging new enzyme carrier materials including carbonaceous and magnetic nanomaterials, gel microspheres, and membranes as well as their effect on increasing enzyme stability and recycling are highlighted. Finally, we provide our expert opinion for more innovative research in the field and future outlook for the industrial application of ILs in biocatalysis.
The 2D porous protonated NiSe/g-C3N4 nanosheets are synthesized via a chemical-calcination-etching-acidification-hydrothermal method. The obtained 2D NiSe/g-C3N4 nanosheets (NiSe/HCN-3) exhibit ...signally enhanced HER (∼1184.74 μmol g−1 h−1)/degradation (∼0.04645 min−1) than the monomer g-C3N4 (∼150/5-folds), including a decent stability (average HER ∼1152.71 μmol g−1 h−1). It mainly attributes to the NiSe/g-C3N4 interface owns appropriate potential gradient can promote the interface carrier driving to optimize efficiency, including increasing carrier transport, extending carrier lifetime, and reducing recombination, as well as the porous nanosheets and surface protonation obtain numerous active sites for decreasing overpotential and promoting electron solid/liquid interfacial diffusion. Additionally, the 2D porous nanosheets can improve the solar efficiency (multiple internal reflections) and shorten the carrier pathway.
Display omitted
•NiSe co-catalyst optimize carrier efficiency by appropriate Fermi level gradient.•Surface protonation increase active sites to promote electron interfacial diffusion.•Porous nanosheets increase active sites/solar efficiency and shorten carrier path.
The photocatalyst would be a promising method for easing energy crisis and improving ecological environment. However, the high photo-generated carrier recombination and photocorrosion would be the ...core issue for restricting its actual applications, and can be ameliorated by the surface heterojunction modification. The Bi2WO6/CdS hollow core-shell S-scheme nano-heterojunction is synthesized via chemical-hydrothermal method. Bi2WO6/CdS nano-heterojunction (Bi2WO6/CdS-3) exhibits signally enhanced visible light HER (∼1250.33 μmol g−1·h−1)/degradation ciprofloxacin (∼85.0%/24 min, ∼0.07580 min−1) than monomer CdS (∼36/5.9-folds) and monomer Bi2WO6 (∼50/12.7-folds), including decent stability (average HER ∼1225.28 μmol∙g−1 h−1 and photodegradation ∼83.3%). It mainly attributes to the Bi2WO6/CdS S-scheme heterojunction with appropriate potential gradient can promote interface carrier driving to optimize efficiency and provide strong stronger reduction/oxidation, including increasing carrier transport, extending carrier lifetime, and reducing recombination. Additionally, the hollow core-shell structure can increase numerous active sites for decreasing overpotential and promoting electron diffusion, and improve solar efficiency (multiple reflections), including shorting carrier pathway for photocorrosion optimization, while enhancing structural stability.
Display omitted
•S-scheme heterojuction optimize carrier efficiency via greater Fermi levels gradient.•S-scheme heterojuction provide strong stronger reduction/oxidation.•Hollow core-shell structure increase active sites to promote electron diffusion.•Template auxiliary can regulate size accurately and form hollow structure easily.
Hollow cubic nanoheterojunction in SnS2/CdS is obtained via in situ sulfuration-hydrothermal-chemical deposition. The SnS2/CdS nanoheterojunction (SnS2/CdS-3) exhibits remarkable enhanced ...photocatalytic HER (∼1424.56 μmol g−1 h−1) and photodegradation than normal CdS (∼72/∼28 folds) and SnS2 (∼148/∼29 folds). In this structure, the SnS2/CdS heterojunction with greater Fermi levels gradient can promote interface carrier drive to optimize carrier efficiency, the hollow cubic structure can provide numerous active sites to decreased overpotential and promote electron diffusion, and increase light reflections, including inhibiting photocorrosion and maintaining structural stability.
Display omitted
•Heterojuction with greater Fermi levels gradient can optimize carrier efficiency.•Hollow cubic structure can increase active sites to promote electron diffusion.•Hollow cubic structure can shorten carrier pathway to inhibit photocorrosion.