Advanced intelligent systems (Weinheim an der Bergstrasse, Germany)最新文献

Leveraging the unique attributes of functional soft materials to generate force and deformation, significant advancements in soft actuators are driving the evolution of smart robotics. Liquid metals (LMs), known for their high deformability and tunable morphology, demonstrate remarkable actuating capabilities through controllable surface tension. Inspired by the predation method of chameleons, this work introduces a bioinspired LM actuator (BLMA) by modulating the morphology of LM. This BLMA enables high-strain (up to 170%) actuation by precisely directing LM droplets toward an electrode. Various parameters affecting the BLMA's actuating performance are explored. Notably, the application of a reductive voltage induces rapid solidification of supercooled LM, facilitating phase transition at room temperature. The solidified LM enhances its holding force of BLMA by over 1000 times. To underscore the superior capabilities of the BLMA, diverse applications, such as a complex two-dimensional plane actuator, a stepper motor with adjustable step intervals, a phase transition-controlled relay, and a laser code lock actuation gate set, are presented. It is anticipated that the exceptional characteristics of the BLMA will propel advancements in the realms of soft robotics and mechatronics.

Monolithic three-dimensional (M3D) integration advances integrated circuits by enhancing density and energy efficiency. Ferroelectric thin-film transistors (Fe-TFTs) attract attention for neuromorphic computing and back-end-of-the-line (BEOL) compatibility. However, M3D faces challenges like increased runtime temperatures due to limited heat dissipation, impacting system reliability. This work demonstrates the effect of temperature impact on single-gate (SG) Fe-TFT reliability. SG Fe-TFTs have limitations such as read-disturbance and small memory windows, constraining their use. To mitigate these, dual-gate (DG) Fe-TFTs are modeled using technology computer-aided design, comparing their performance. Compute-in-memory (CIM) architectures with SG and DG Fe-TFTs are investigated for deep neural networks (DNN) accelerators, revealing heat's detrimental effect on reliability and inference accuracy. DG Fe-TFTs exhibit about 4.6x higher throughput than SG Fe-TFTs. Additionally, thermal effects within the simulated M3D architecture are analyzed, noting reduced DNN accuracy to 81.11% and 67.85% for SG and DG Fe-TFTs, respectively. Furthermore, various cooling methods and their impact on CIM system temperature are demonstrated, offering insights for efficient thermal management strategies.

Reservoir computing (RC) possesses a simple architecture and high energy efficiency for time-series data analysis through machine learning algorithms. To date, RC has evolved into several innovative variants. The next generation reservoir computing (NGRC) variant, founded on nonlinear vector autoregression (NVAR) distinguishes itself due to its fewer hyperparameters and independence from physical random connection matrices, while yielding comparable results. However, NGRC networks struggle with massive Kronecker product calculations and matrix-vector multiplications within the read out layer, leading to substantial efficiency challenges for traditional von Neumann architectures. In this work, a hybrid digital-analog hardware system tailored for NGRC is developed. The digital part is a Kronecker product calculation unit with data filtering, which realizes transformation of nonlinear vector of the input linear vector. For matrix-vector multiplication, a computing-in-memory architecture based on resistive random access memory array offers an energy-efficient hardware solution, which markedly reduces data transfer and greatly improve computational parallelism and energy efficiency. The predictive capabilities of this hybrid NGRC system are validated through the Lorenz63 model, achieving a normalized root mean square error (NRMSE) of 0.00098 and an energy efficiency of 19.42TOPS W⁻¹.

The production industry is challenged to become more flexible and efficient while coping with a variety of disruptive events, such as natural disasters, infrastructure blockages, or economic crises. From the individual station on a production line to the global supply chain, everything is connected, making regulation and control a complex task. Biological molecular processes, such as the metabolism of living organisms or the cell cycle, are also extremely complex processes that can be compared to industrial production processes, both of which involve a series of intermediate steps and products. Thanks to (self-)regulatory mechanisms that have evolved over time, these biological mechanisms are very efficient and robust in the face of perturbations. This article proposes an explanatory representation of these complex processes, considering both biological and technical aspects. The aim is to facilitate biomimetic transfer of biological regulation mechanisms into the technical domain. It presents concepts for biomimetic regulation of production lines and sourcing strategies and introduces a workflow for generating digital twins. This workflow is inspired by the cell cycle checkpoints, which ensure that only perfect copies of DNA are passed on during cell replication. By leveraging this understanding, the production industry can potentially improve its own processes and efficiency.

Sensors based on everyday textiles are extremely promising for wearable applications. The present work focuses on high-performance textile-based capacitive strain sensors. Specifically, a conductive textile is obtained via vapor-phase polymerization of pyrrole, in which the usage of methanol co-vapor and the addition of imidazole to the iron chloride oxidant solution are shown to maximize conductivity. A technique to provide insulation and mechanical resistance using thermoplastic polyurethane and polystyrene-block-polyisoprene-block-polystyrene/barium titanate composite is developed. Such insulated conductive elastics are then used to fabricate highly sensitive twisted yarn capacitive sensors. A textile glove is subsequently embedded with such sensors. The wireless measurement and transmission system demonstrate efficacy in capturing capacitance variations upon strain and monitoring hand motions. A machine learning model to recognize 12 gestures is implemented—100% classification accuracy is obtained.

The precise control of soft and continuum robots requires knowledge of their shape, which has, in contrast to classical rigid robots, infinite degrees of freedom. To partially reconstruct the shape, proprioceptive techniques use built-in sensors, resulting in inaccurate results and increased fabrication complexity. Exteroceptive methods so far rely on expensive tracking systems with reflective markers placed on all components, which are infeasible for deformable robots interacting with the environment due to marker occlusion and damage. Here, a regression approach is presented for three-dimensional key point estimation using a convolutional neural network. The proposed approach uses data-driven supervised learning and is capable of online markerless estimation during inference. Two images of a robotic system are captured simultaneously at 25 Hz from different perspectives and fed to the network, which returns for each pair the parameterized key point or piecewise constant curvature shape representations. The proposed approach outperforms markerless state-of-the-art methods by a maximum of 4.5% in estimation accuracy while being more robust and requiring no prior knowledge of the shape. Online evaluations on two types of soft robotic arms and a soft robotic fish demonstrate the method's accuracy and versatility on highly deformable systems.

Microwave (MW)-driven strategies could potentially offer transmissivity through obstacles, selective energization, wave-focusing, and nonmechanical steering via phased-array technologies, which have unique application prospects in confined environments. Implementing multidegree of freedom (MDOF) control is the key to developing motion strategies for advanced MW-driven robots. However, the use of MWs for MDOF control of robots has so far remained a challenge. This article first presents examples of wireless driving and controlling millimeter-scale peristaltic pipeline robots (diameter, 4 mm; length, 30 mm; weight, 0.39 g) utilizing multifrequency MWs in complex nonmetal channel environments. Herein, shape memory alloy springs combined with passive wires, whose length depends on the frequencies of MWs, are adopted to form selectively controlled robot components, and silicon rubber replica technology is utilized to achieve miniaturization of the robot. The monopole antenna model is developed for the actuator, and its structure is refined using antenna theory. This allows the successfully achievement of MDOF control of the robot at several MW frequencies (2.4, 4, and 5.9 GHz). Powered by MWs from outside the pipe, the pipeline robot achieves horizontal, vertical, and curved motions in a pipe with a diameter of 5 mm.

In this transformative study, machine learning (ML) and t-distributed stochastic neighbor embedding (t-SNE) are employed to interpret intricate patterns in colorimetric images of cold atmospheric plasma (CAP)-treated water. The focus is on CAP's therapeutic potential, particularly its ability to generate reactive oxygen and nitrogen species (RONS) that play a crucial role in antimicrobial activity. RGB, HSV, LAB, YCrCb, and grayscale color spaces are extracted from the colorimetric expression of oxidative stress induced by RONS, and these features are used for unsupervised ML, employing density-based spatial clustering of applications with noise (DBSCAN). The DBSCAN model's performance is evaluated using homogeneity, completeness, and adjusted rand index with a predictive data distribution graph. The best results are achieved with 3,3′,5,5′-tetramethylbenzidine–potassium iodide colorimetric assay solution immediately after plasma treatment, with values of 0.894, 0.996, and 0.826. t-SNE is further conducted for the best-case scenario to evaluate the clustering efficacy and find the best combination of features to better present the results. Correspondingly, t-SNE enhances clustering efficacy and adeptly handles challenging points. The approach pioneers dynamic and comprehensive solutions, showcasing ML's precision and t-SNE's transformative visualization. Through this innovative fusion, complex relationships are unraveled, marking a paradigm shift in biomedical analytical methodologies.

Additive manufacturing technologies increasingly revolutionize current production techniques for object manufacturing. Particularly, fused deposition modeling (FDM) strongly impacts production processes by enabling the cost-effective and efficient creation of structures with complex designs and innovative geometries. The use of conductive filaments in FDM printing is paving the way for the advancement of entirely printed sensors and circuits, although this domain is still in its early stages. In this article, the design and production of bilayer deformable pressure sensors fabricated using conductive thermoplastic polyurethane are investigated. The potential to vary the mechanical and electrical characteristics of FDM-printed components by adjusting printing parameters is explored. The influence of different levels of material infill (20%, 50%, and 100%) and different contact geometries between layers (domes, pyramids, and cylinders) is studied. Electromechanical tests are carried out to characterize the sensor, applying pressures up to 22 kPa. The 3D-printed pressure sensors demonstrate tunable mechanical and electrical sensitivities at different infill values, with the highest value of −6.3 kPa⁻¹ achieved by using a pyramid layer at 100% infill. Sensor outputs registered during cyclic tests show reproducible responses with a wide range of sensitivity, paving the way for applicability in recording both static and periodic pressure changes.

Dynamic and agile locomotion in legged robots enables them to overcome obstacles and navigate complex and unstructured terrain. However, the leg mechanisms and actuators needed for versatile locomotion are much more challenging to manufacture and integrate in sub-gram scale robots. Herein, Picotaur, a 15.4 mg hexapedal robot with legs that enable various locomotion tasks such as turning, climbing 3D-printed stairs, and pushing loads for the first time at these size scales, is presented. 3D printing with two-photon polymerization enables the manufacture of electrostatically driven 2 degrees of freedom legs on a robot body made from a flexible printed circuit board. Based on simple control inputs, Picotaur can achieve alternating tripod gaits, reaching speeds up to 57 mm (7.2 body lengths) per second, as well as pronking gaits to tackle a wider variety of terrain. This approach to manufacturing and controlling legged robots at smaller scales provides a path forward toward robots that can be used for practical applications ranging from inspection to exploration and rival the performance of insects at similar size scales.