Dexterous manipulation using multi-finger hands requires precise control and adaptability. Previous studies of dexterous manipulation have been conducted via imitation learning (IL). However, training multi-finger robot hands to imitate expert-like behaviors still remains challenging, since traditional IL algorithms suffer from state occupancy mismatches. In this paper, we propose Generative Adversarial State Policy (GASP), a novel IL algorithm that employs a state-based discriminator to explicitly align the policy’s state distribution with demonstrations, thereby imitating expert-like behaviors. We also introduce a task environment that utilizes tactile sensors, which allows for more sensitive feedback on complex deformable objects. We validate our approach in tasks involving both rigid and deformable objects, including a challenging pour task, where it is shown that tactile feedback is critical for precise force control and dynamic adjustments of the multi-finger hand. Various experiments demonstrate that GASP outperforms state-of-the-art methods, with and without tactile sensors, in dexterous manipulation tasks.
Localized high-concentration electrolytes (LHCEs) are considered as promising electrolyte candidates to resolve technical issues of metal batteries owing to their unique interfacial properties and solvation structures. Herein, we propose a self-assembly chemical strategy into the LCHEs induced by ordered nanostructure of zwitterionic co-solutes for highly efficient and ultrastable zinc (Zn) metal batteries. Through the systematic screening of six zwitterionic compounds, 3-(decyldimethylammonio)propanesulfonate salt (C10) with the decyl chain and zwitterions was determined as an optimum to construct quasi-spherical aggregates with a periodic length of 3.77 nm, as confirmed by comprehensive synchronous small-angle X-ray scattering, Guinier, pair distance distribution function, Porod, and other spectroscopic characterizations and molecular dynamic simulation. In particularly, this self-assembled structure in electrolyte environments was attributed to increasing the proportion of both contact and aggregated ion pairs for the formation of LHCEs as well as to providing fast and selective Zn2+ conducting channels and uniform solid electrolyte interfaces for facilitated charge transfer kinetics. Moreover, the preferential adsorption of the self-assembled C10 on the Zn(002) surface modulated the electrical double layer to suppress hydrogen evolution and corrosion reactions. Consequently, the Zn||Zn symmetric cells in Zn(OTf)2/C10 electrolytes showed long-term plating/stripping behaviors over 2800 h at 1 mA cm-2 and 1 mAh cm-2 as well as over 1200 h even at 5 mA cm-2 and 5 mAh cm-2 with a very high depth of discharge of 42.7%. Furthermore, the Zn||VO2/CNT full cells in Zn(OTf)2/C10 electrolytes delivered a record-high capacity of 8.10 mAh cm-2 at an ultrahigh cathode mass loading of 50 mg cm-2 after 150 cycles.
Aqueous zinc-ion batteries (AZIBs) are gaining attention due to their safety and cost-effectiveness. However, zinc (Zn) anodes face persistent issues such as dendrite growth, side reactions, and the accumulation of inactive Zn, particularly when Zn powder is used due to its high surface area and corrosion susceptibility. In this work, we demonstrate a conformal SnO2 surface engineering strategy that enables slurry casted Zn powder anodes to operate stably under high depth of discharge (DOD) conditions in aqueous Zn-ion batteries. Electrochemical testing revealed enhanced cycling stability and plating/stripping reversibility in SnO2 coated Zn powder, accompanied by suppressed side reactions and reduced inactive Zn formation. Coulombic efficiency (CE) and DOD were significantly improved, while electrochemical impedance spectroscopy (EIS) confirmed reduced charge transfer resistance and improved interfacial characteristics. The coated Zn powder anodes also exhibited strong compatibility with zinc vanadium oxide (ZVO) and manganese dioxide (beta-MnO2) cathodes, demonstrating their versatility. Their performance in pouch cell configurations suggests practical scalability. Overall, this study highlights SnO2 surface coating as an effective strategy for improving Zn powder anode reversibility and durability, offering a feasible path toward high-performance and long-lasting AZIBs.
This article provides a comprehensive review of power management circuit techniques for millimeter-scale biomedical sensing systems, which operate under strict power and energy constraints. It begins by introducing a miniature sensing platform and outlining the key challenges associated with limited energy availability in such ultrasmall devices. The discussion then highlights advances in circuit design for efficient power conversion, battery management, ambient energy harvesting, and wireless power transfer. By examining these techniques, this article aims to clarify the major design challenges and emerging solutions that are driving the development of next-generation miniature biomedical electronics.
Time series classification (TSC) presents significant challenges in data analytics and plays an essential role in industries such as manufacturing, healthcare, and finance. Existing research has utilized sequence models, such as long short-term memory (LSTM) and transformers, to achieve high performance; however, these models fail to fully address the inherent challenges of capturing long-term dependencies in time series data. Recently, efforts have been made to overcome these limitations by processing time series data through conversion into images. One such method, the Gramian Angular Field (GAF), converts time series data into images that visually represent the relationship between time points, effectively capturing global trends that traditional sequence models often miss. However, single-modality approaches, which rely on only one representation of time series data, still face limitations in comprehensively capturing the diverse features of a time series. Additionally, existing multi-modality approaches may overlook the detailed features between time points. To address these issues, this article proposes a patch-level hybrid contrastive learning model that combines transformer with Vision Transformer (ViT). The proposed model converts a one-dimensional time series into GAF images and leverages contrastive learning to robustly learn fine-grained features. This enables the model to effectively capture both temporal relationships between time points and spatial relationships between patches, thereby enhancing its generalization capabilities. Experimental results demonstrate that the proposed model outperforms the existing methods on the UCR TSC dataset. This shows that the model can overcome the limitations of single-modality approaches by integrating the complex structural features of time series data through a multi-modality framework, ultimately leading to improved classification performance. This approach provides a new avenue for enhancing TSC and holds promise for applications across various industries.
