Reliable and energy-efficient Bluetooth Low Energy (BLE) communication is crucial for Internet of Things (IoT) applications in dynamic environments. However, the Received Signal Strength Indicator (RSSI) and data throughput in BLE are highly susceptible to environmental variability, affecting communication performance. In this work, we systematically analyze the interdependence between RSSI, throughput, transmission power (TXP), and the peripheral device’s system power consumption under various real-world conditions, observing that adjusting the TXP effectively influences both the RSSI and throughput. We propose a robust closed-loop TXP control framework based on Proportional-Integral-Derivative (PID) controllers. Two initial control strategies were implemented: an RSSI-based approach and a throughput-based approach, each with distinct advantages and limitations. The RSSI-based method offers rapid responsiveness to signal fluctuations but lacks direct correlation with data throughput, while the throughput-based method provides more accurate feedback on transmitting throughput at the cost of slower response. To address these limitations, a hybrid RSSI–throughput control strategy was developed, combining the responsiveness of RSSI feedback with the accuracy of throughput measurements. This hybrid approach effectively maintains data throughput near the target level with minimal variance, even under rapidly changing environmental conditions.

Precise motion control of magnetic microrobots in complex and dynamic environments remains a critical challenge for enabling key applications such as targeted therapy and micromanipulation. Purely manual teleoperation is prone to operator fatigue and error, while fully autonomous systems often lack the robustness and adaptability to handle. Here, we propose a human-machine shared cascade control method for magnetically driven microrobots, which effectively integrates human cognitive intelligence with machine autonomy for collision-free navigation in dynamic environments. The outer-loop hybrid shared control unit smoothly modulates control authority in response to real-time collision risk, dynamically integrating the operator instructions and the autonomous navigation system output guided by the enhanced artificial potential field method to formulate the guidance law. For the inner-loop motion tracking, a data-driven adaptive orientation controller is designed, which integrates a nonlinear feedforward compensator leveraging a Gaussian process regression (GPR) model with a linear feedback controller whose parameters are optimized using the virtual reference feedback tuning (VRFT) method, ensuring fast and precise tracking of the desired motion. The effectiveness of the proposed method was validated through both simulation and physical experiments. In human-subject studies conducted on a physical magnetic actuation platform featuring both static and dynamic obstacle scenarios, quantitative results demonstrate that the shared control strategy significantly outperforms both purely manual and fully autonomous modes across all key metrics, including success rate, task completion time, stability, and safety ( $p \lt 0.001$ ). Furthermore, successful navigation within a complex gastric model demonstrates the potential of the shared control system for practical application in unstructured environments.

This article presents a software–hardware codesigned field-programmable gate array (FPGA)-based real-time electrocardiogram (ECG) classification system that combines methodological and practical innovations to achieve state-of-the-art performance with an ultracompact model. On the software side, we introduce a hardware-adaptive, configurable quantization-aware training (QAT) framework that enables layerwise precision assignment and flexible quantization, ensuring that the trained model is highly accurate and hardware-friendly even at ultralow bit widths. On the hardware side, we propose a resource-efficient FPGA accelerator featuring a streaming architecture and a cosine-approximated continuous wavelet transform (CWT) module, optimized for low-power and real-time inference. Implemented in an FPGA, we demonstrate that a six-layer Inception-ResNeXt (IRN) network can achieve 99.5% inference accuracy on the MIT-BIH ECG dataset with 200-mW dynamic power and 0.0767-mJ/inference energy efficiency.

In dynamic environments, Bluetooth Low Energy (BLE) data throughput often exhibits significant fluctuations. To counteract this instability, transmission power (TXP) is commonly configured at its maximum level—an approach that, while improving data throughput, substantially increases energy consumption. This study introduces a closed-loop TXP control strategy that utilizes a Proportional-Integral-Derivative (PID) controller, deployed at a central device to dynamically regulate the TXP of peripheral devices. The proposed method aims to optimize energy consumption while ensuring the required data throughput. The relationship between Received Signal Strength Indicator (RSSI) and throughput was thoroughly analyzed under varying environmental conditions, demonstrating that environmental factors critically influence throughput stability. Subsequently, comprehensive assessments of power consumption, throughput, and the central device’s RSSI were performed by varying the TXP setting. In distance-varying scenarios (0–50 meters), the PID-controlled system maintained a throughput of around 400 kbps while reducing peripheral power consumption by 4.4 times compared to fixed 8 dBm TXP operation. Furthermore, compared to a fixed -10 dBm TXP configuration, the PID controller sustained throughput above application requirements with the same power consumption level, achieving an optimal balance between energy efficiency and communication reliability.

Magnetic helical microrobots have great potential in biomedical applications due to their ability to access confined and enclosed environments via remote manipulation by magnetic fields. However, achieving collision-free navigation for microrobots in complex and unstructured environments, particularly in highly dynamic settings, remains a challenge. In this paper, we present a novel deep reinforcement learning-based control framework for magnetic helical microrobots, focusing on the tasks of goal-reaching and dynamic obstacle avoidance. To streamline data collection, a specialized training environment capturing essential aspects of navigation for magnetic helical microrobots is devised. The robustness and adaptability of the trained policy are supported using a randomization technique within the training environment. To facilitate seamless integration with real-world magnetic actuation systems, a visual processing algorithm based on OpenCV is devised and incorporated to collect policy observations. Simulations and experiments in various scenarios validate the high robustness and adaptability of the method. The performance assessment revealed a success rate of 99% in navigating the microrobot around 4 dynamic obstacles of comparable speeds and a success rate of 90% in environments with 14 dynamic obstacles. The results indicate the potential for future applications of our method in unstructured, confined, and dynamic living environments.Note to Practitioners—The motivation of this work is to develop a robust and effective control scheme for collision-free navigation of magnetic helical microrobots in dynamic environments. The conventional navigation strategies in dynamic environments mainly include global path planning and local path replanning; thus, highly dynamic environments require frequent updates to the planned path, making it difficult to apply in highly dynamic environments. In this work, a deep reinforcement learning-based control framework is proposed that can guide microrobots through many dynamic obstacles to a series of locations without collisions. The simulation and experimental results validate the efficacy of the proposed control framework and the robustness and adaptability of the trained policy. The proposed control scheme enables better understanding of advanced motion control methods for magnetic microrobots.

Advancements in wearable technology and edge computing have transformed cardiovascular monitoring, driving the demand for private, secure, and real-time diagnostic solutions. This paper presents an edge-wearable ECG monitoring system that integrates personalized continuous wavelet transform (CWT) preprocessing, a DeepFool-FGSM adversarial defense, and an optimized parallel PoolFormer architecture for resource-constrained FPGA deployment. The personalized CWT captures individual-specific ECG features and mitigates model-inversion privacy risks. The defense approach balances robustness and computational efficiency and reduces hardware complexity and energy via quantization-aware training (QAT). Evaluations on field programmable gate array (FPGA) confirm high diagnostic accuracy (98.93%), real-time inference (latency <1.7 ms), and improved robustness against adversarial perturbations, with 0.055 W FPGA-core power. Together, the system delivers confidentiality, integrity, and availability for cybersecure, personalized ECG monitoring at the edge.

Magnetic helical microrobots have attracted considerable attention in navigation control. However, the performance of microrobots is negatively affected by time-varying uncertain perturbations and obstacles, at the microscale. In this study, we present a navigation control scheme for accurately guiding the helical microrobot to targeted positions in dynamically changing environments. To efficiently plan smooth paths, a search-based algorithm with pruning rules is implemented to quickly find collision-free waypoints and design an optimal method with spatial and dynamic constraints for obtaining smooth paths globally. Velocity gain and potential fields are integrated to develop an emergency local motion replanning method for addressing random obstacles that suddenly appear in the preset path. In order to attain microrobot system dynamic linearization and achieve precise path following of a helical microrobot, a robust control strategy that integrates geometric and model-free controllers in a complementary manner is presented. The geometric controller as a feedforward controller, responsible for managing path information and generating guidance laws. In contrast, the model-free controller operates as a feedback controller, specifically designed to rapidly address position deviation. Meanwhile, we employ an observer to compensate for disturbances. Experimental results of precise motion control in both static and dynamic environments demonstrate the effectiveness of this navigation control scheme, which is promising for moving with high accuracy in cluttered and dynamic living enclosed environments.Note to Practitioners—This paper was motivated by the problem of the navigation control of magnetic microrobots in dynamic environment. The existing navigation control methods of microrobots mainly focus on the static environment, which is challenging to meet the emergency obstacle avoidance requirements in the cluttered environment with low Reynolds number. In addition, the conventional path following control always ignores the nonlinearity of the microrobot system, resulting in insufficient following accuracy. In this work, a novel navigation control method for microrobots is proposed, which can guide microrobots to accurately follow dynamically planned paths in cluttered environments without collision. Simulations and experiments validate the performance of the proposed navigation control method using helical microrobots. The proposed navigation control method paves the way for a better understanding of advanced navigation control method for magnetic microrobots.

Manipulating heterogeneous microtargets based on optoelectronic tweezers (OETs) to construct micropatterns with specific distribution and ordered arrangement enables recapitulating the spatial architecture of cells in native tissues, and has significant potential in tissue regeneration, medical diagnostics, and cell behavior research. However, the uncertain disturbances in liquid environment, collision risk, and electrokinetic interference in OETs system can cause microtargets to deviate from the safe and controlled state, especially for manipulation tasks with heterogeneous microtargets. Here, we propose an improved hierarchical value iteration network (IHVIN)-GAT-based path planning method for parallel manipulation of heterogeneous microtargets with independent control, integrating goal assignment, feature extraction, and decentralized decision-making. The Kuhn-Munkres-based goal assignment model periodically modifies the matching relationship between microtargets and goal positions to reduce the task complexity. High-order features involving path planning are extracted by an IHVIN model, and then selectively aggregated and convolved through GAT to yield real-time locomotion strategies for all microtargets. For the issues of constraint variability and system heterogeneity, discrete locomotion constraints are developed through analysis of escape mechanism, then embedded into modeling procedures and converted to heterogeneous edge weights in graph domain. The simulation and experimental results demonstrate the desired performance of the IHVIN-GAT model in high timeliness, high strategy quality, and compatibility for microtarget number, where up to 20 microtargets from three categories are parallel manipulated within 16 s to form arbitrary micropatterns recapitulating microscale architecture of cells in native tissues. We anticipate that our method will contribute to construct more biomimetic microstructures with heterogeneous cells for biomedical applications in the future.

Recent failures in real-world self-driving tests have suggested a paradigm shift from directly learning in real-world roads to building a high-fidelity driving simulator as an alternative, effective, and safe tool to handle intricate traffic environments in urban areas. To date, traffic simulation can construct virtual urban environments with various weather conditions, day and night, and traffic control for autonomous vehicle testing. However, mutual interactions between autonomous vehicles and pedestrians are rarely modeled in existing simulators. Besides vehicles and pedestrians, the usage of personal mobility devices is increasing in congested cities as an alternative to the traditional transport system. A simulator that considers all potential road-users in a realistic urban environment is urgently desired. In this work, we propose a novel, extensible, and microscopic method to build heterogenous traffic simulation using the force-based concept. This force-based approach can accurately replicate the sophisticated behaviors of various road users and their interactions through a simple and unified way. Furthermore, we validate our approach through simulation experiments and comparisons to the popular simulators currently used for research and development of autonomous vehicles.

Shape shifting soft microrobots are generated from self-folding hydrogel bilayer structures. The folding conditions are analyzed to develop an optimal strategy for producing desired three-dimensional shapes. We present two different methods for programming magnetization in these microrobots that are variant and invariant to folding. The microrobots can be navigated through user-defined trajectories using rotating magnetic fields, and the morphing in response to temperature changes can be tuned for adaptive behavior. On-demand modulation of the mobility of individual microrobots is demonstrated by morphing their shape using selective near infrared light (NIR) exposure.