Search Results (12)

For large oil-immersed transformers, their metal-enclosed structure poses significant challenges for direct visual inspection of internal defects. To ensure the effective detection of internal insulation defects, this study employs a self-developed micro-robot for internal visual inspection. Given the substantial morphological and dimensional variations of target defects (e.g., carbon traces produced by surface discharge inside the transformer), the intelligent and efficient extraction of carbon trace features from complex backgrounds becomes critical for robotic inspection. To address these challenges, we propose the DCMC-UNet, a semantic segmentation model for carbon traces containing adaptive illumination enhancement and dynamic feature fusion. For blurred carbon trace images caused by unstable light reflection and illumination in transformer oil, an improved CLAHE algorithm is developed, incorporating learnable parameters to balance luminance and contrast while enhancing edge features of carbon traces. To handle the morphological diversity and edge complexity of carbon traces, a dynamic deformable encoder (DDE) was integrated into the encoder, leveraging deformable convolutional kernels to improve carbon trace feature extraction. An edge-aware decoder (EAD) was integrated into the decoder, which extracts edge details from predicted segmentation maps and fuses them with encoded features to enrich edge features. To mitigate the semantic gap between the encoder and the decoder, we replace the standard skip connection with a cross-level attention connection fusion layer (CLFC), enhancing the multi-scale fusion of morphological and edge features. Furthermore, a multi-scale atrous feature aggregation module (MAFA) is designed in the neck to enhance the integration of deep semantic and shallow visual features, improving multi-dimensional feature fusion. Experimental results demonstrate that DCMC-UNet outperforms U-Net, U-Net++, and other benchmarks in carbon trace segmentation. For the transformer carbon trace dataset, it achieves better segmentation than the baseline U-Net, with an improved mIoU of 14.04%, Dice of 10.87%, pixel accuracy (P) of 10.97%, and overall accuracy (Acc) of 5.77%. The proposed model provides reliable technical support for surface discharge intensity assessment and insulation condition evaluation in oil-immersed transformers. Full article

Artificial intelligence (AI) is one of the most promising technological advancements that have revolutionized the healthcare sector (medicine and dentistry). AI and its subsets, such as machine learning (ML), artificial neural networks (ANNs), and deep learning (DL), are being used in dentistry for data recording and management, patient education, radiographic interpretation, diagnosis, and treatment plans. AI and ML tools are commonly employed to improve oral hygiene and patient compliance. This narrative review paper discusses the innovations in AI-based plaque control aids (toothbrushes and interdental aids) that have improved overall health and patients’ hygiene compliance. We performed a literature search using different databases using the following keywords: “Artificial intelligence or machine learning or robots or robotics” AND “Toothbrush OR Smart toothbrush”. We included all the studies evaluating the use of any smart toothbrush, AI, or robotics for oral hygiene, plaque control, and patient education. AI-based smart toothbrushes helped patients to brush effectively by indicating the amount of pressure and the time taken for brushing, along with providing feedback on their brushing performance. Many microrobots can even recognize and automatically remove biofilm. Some AI-based smart toothbrushes are beneficial for children, patients with disabilities lack of manual dexterity, and neurological disorders. However, dental professionals choose AI-based smart toothbrushes for patients with poor oral hygiene and poor compliance for more effective control of oral diseases and to provide better health. Full article

In recent years, the field of microrobots has exploded, yielding many exciting new functions and applications, from object grasping and release to in vivo drug transport. Smart responsive materials have had a profound impact on the field of microrobots and have given them unique functions and structures. We analyze three aspects of microrobots, in which the future development of microrobots requires more efforts to be invested, and in which smart materials play a significant role in the development of microrobots. These three aspects are smart materials for building microrobots, manufacturing methods, and the functions and applications they achieve. In this review, we discuss the deformation mechanism of materials in response to external stimuli, starting from smart materials, and discuss fabrication methods to realize microrobots, laying the theoretical foundation for future smart material-based microrobots to realize their intelligence and programmability. Full article

Microrobotics has opened new horizons for various applications, especially in medicine. However, it also witnessed challenges in achieving maximum optimal performance. One key challenge is the intelligent, autonomous, and precise navigation control of microrobots in fluid environments. The intelligence and autonomy in microrobot control, without the need for prior knowledge of the entire system, can offer significant opportunities in scenarios where their models are unavailable. In this study, two control systems based on model-free deep reinforcement learning were implemented to control the movement of a disk-shaped magnetic microrobot in a real-world environment. The training and results of an off-policy SAC algorithm and an on-policy TRPO algorithm revealed that the microrobot successfully learned the optimal path to reach random target positions. During training, the TRPO exhibited a higher sample efficiency and greater stability. The TRPO and SAC showed 100% and 97.5% success rates in reaching the targets in the evaluation phase, respectively. These findings offer basic insights into achieving intelligent and autonomous navigation control for microrobots to advance their capabilities for various applications. Full article

Micro- and nanorobots are highly intelligent and efficient. They can perform various complex tasks as per the external stimuli. These robots can adapt to the required functional form, depending on the different stimuli, thus being able to meet the requirements of various application scenarios. So far, microrobots have been widely used in the fields of targeted therapy, drug delivery, tissue engineering, environmental remediation and so on. Although microbots are promising in some fields, few reviews have yet focused on them. It is therefore necessary to outline the current status of these microbots’ development to provide some new insights into the further evolution of this field. This paper critically assesses the research progress of microbots with respect to their preparation methods, stimulus–response mechanisms and applications. It highlights the suitability of different preparation methods and stimulus types, while outlining the challenges experienced by microbots. Viable solutions are also proposed for the promotion of their practical use. Full article

To meet the increasing needs of point-of-care testing in clinical diagnosis and daily health monitoring, numerous cutting-edge techniques have emerged to upgrade current portable biosensors with higher sensitivity, smaller size, and better intelligence. In particular, due to the controlled locomotion characteristics in the micro/nano scale, microrobots can effectively enhance the sensitivity of biosensors by disrupting conventional passive diffusion into an active enrichment during the test. In addition, microrobots are ideal to create biosensors with functions of on-demand delivery, transportation, and multi-objective detections with the capability of actively controlled motion. In this review, five types of portable biosensors and their integration with microrobots are critically introduced. Microrobots can enhance the detection signal in fluorescence intensity and surface-enhanced Raman scattering detection via the active enrichment. The existence and quantity of detection substances also affect the motion state of microrobots for the locomotion-based detection. In addition, microrobots realize the indirect detection of the bio-molecules by functionalizing their surfaces in the electrochemical current and electrochemical impedance spectroscopy detections. We pay a special focus on the roles of microrobots with active locomotion to enhance the detection performance of portable sensors. At last, perspectives and future trends of microrobots in biosensing are also discussed. Full article

With the continuous integration of material science and bionic technology, as well as increasing requirements for the operation of robots in complex environments, researchers continue to develop bionic intelligent microrobots, the development of which will cause a great revolution in daily life and productivity. In this study, we propose a bionic flower based on the PNIPAM–PEGDA bilayer structure. PNIPAM is temperature-responsive and solvent-responsive, thus acting as an active layer, while PEGDA does not change significantly in response to a change in temperature and solvent, thus acting as a rigid layer. The bilayer flower is closed in cold water and gradually opens under laser illumination. In addition, the flower gradually opens after injecting ethanol into the water. When the volume of ethanol exceeds the volume of water, the flower opens completely. In addition, we propose a bionic Venus flytrap soft microrobot with a bilayer structure. The robot is temperature-responsive and can reversibly transform from a 2D sheet to a 3D tubular structure. It is normally in a closed state in both cold (T < 32 °C) and hot water (T > 32 °C), and can be used to load and transport objects to the target position (magnetic field strength < 1 T). Full article

Due to their fascinating solitary and collective behavior, photochemical microrobots have attracted extensive attention from researchers and have obtained a series of outstanding research progress in recent years. However, due to the limitation of using a single light source, the realization of reconfigurable and controllable motion behaviors of the photochemical microrobot is still facing a series of challenges. To release these restrictions, we reported a multi-light-field-coupling-based method for driving the photochemical microrobot or its swarm in a regulatable manner. Here, we first designed a control system for coupling multiple light sources to realize the programmable application of four light sources in different directions. Then a $Ti{O}_2$-based photochemical microrobot was prepared, with its surface electric field distribution under different lighting conditions estimated by modeling-based simulation, where the feasibility of regulating the microrobot’s motion behavior via the proposed setup was verified. Furthermore, our experimental results show that under the action of the compound light fields, we can not only robustly control the motion behavior of a single $Ti{O}_2$ microrobot but also reconfigure its collective behaviors. For example, we realized the free switching of the single $Ti{O}_2$ microrobots’ movement direction, and the controllable diffusion, aggregation, the locomotion and merging of $Ti{O}_2$ microrobot swarms. Our discovery would provide potential means to realize the leap-forward control and application of photochemical microrobots from individuals to swarms, as well as the creation of active materials and intelligent synthetic systems. Full article

The self-propulsion of bodies floating in water is of great interest for developing new robotic and intelligent systems at different scales, and whenever possible, Marangoni propulsion is an attractive candidate for the locomotion of untethered micro-robots. Significant cases have been shown using liquid and solid surfactants that allow an effective propulsion for bodies floating on water to be achieved. Here, we show for the first time a strategy for activating a twofold functionality where the self-propulsion of a floating body is combined with the formation of a polymer thin film at the water surface. In fact, we demonstrate that by using polymer droplets with an appropriate concentration of solvent and delivering such drops at specific locations onto freely floating objects, it is possible to form “on-the-fly” thin polymer films at the free water surface. By exploiting self-propulsion, a polymer thin film can be formed that could cover quite extensive areas with different shapes depending on the motion of the floating object. This intriguing twice-functionality activated though a single phenomenon, i.e., film formation and related locomotion, could be used in perspective to perform complex operations at water surfaces, such as dynamic liquid packaging, cleaning, and moving away floating particles, monolayer films, or macro-sized objects, as discussed in the text. Full article

Magnetic microrobots are vital tools for targeted therapy, drug delivery, and micromanipulation on cells in the biomedical field. In this paper, we report an automated control and path planning method of magnetic microrobots based on computer vision. Spherical microrobots can be driven in the rotating magnetic field generated by electromagnetic coils. Under microscopic visual navigation, robust target tracking is achieved using PID–based closed–loop control combined with the Kalman filter, and intelligent obstacle avoidance control can be achieved based on the dynamic window algorithm (DWA) implementation strategy. To improve the performance of magnetic microrobots in trajectory tracking and movement in complicated environments, the magnetic microrobot motion in the flow field at different velocities and different distribution obstacles was investigated. The experimental results showed that the vision-based controller had an excellent performance in a complex environment and that magnetic microrobots could be controlled to move to the target position smoothly and accurately. We envision that the proposed method is a promising opportunity for targeted drug delivery in biological research. Full article

Deep neural networks are widely used in the field of image processing for micromachines, such as in 3D shape detection in microelectronic high-speed dispensing and object detection in microrobots. It is already known that hyperparameters and their interactions impact neural network model performance. Taking advantage of the mathematical correlations between hyperparameters and the corresponding deep learning model to adjust hyperparameters intelligently is the key to obtaining an optimal solution from a deep neural network model. Leveraging these correlations is also significant for unlocking the “black box” of deep learning by revealing the mechanism of its mathematical principle. However, there is no complete system for studying the combination of mathematical derivation and experimental verification methods to quantify the impacts of hyperparameters on the performances of deep learning models. Therefore, in this paper, the authors analyzed the mathematical relationships among four hyperparameters: the learning rate, batch size, dropout rate, and convolution kernel size. A generalized multiparameter mathematical correlation model was also established, which showed that the interaction between these hyperparameters played an important role in the neural network’s performance. Different experiments were verified by running convolutional neural network algorithms to validate the proposal on the MNIST dataset. Notably, this research can help establish a universal multiparameter mathematical correlation model to guide the deep learning parameter adjustment process. Full article

The next generation of intelligent robotic systems has been envisioned as micro-scale mobile and externally controllable robots. Visualization of such small size microrobots to track their motion in nontransparent medium such as human tissue remains a major challenge, limiting translation into clinical applications. Herein, we present a novel, non-invasive, real-time imaging method by integrating ultrasound (US) and photoacoustic (PA) imaging modalities for tracking and detecting the motion of a single microrobot in deep biological tissue. We developed and evaluated a prototyped PA-guided magnetic microrobot tracking system. The microrobots are fabricated using photoresist mixed with nickel (Ni) particles. The microrobot motion was controlled using an externally applied magnetic field. Our experimental results evaluated the capabilities of PA imaging in visualizing and tracking microrobots in opaque tissue and tissue-mimicking phantoms. The results also demonstrate the ability of PA imaging in detecting a microrobot with the sizes less than the minimum detectable size by US imaging (down to 50 µm). The spectroscopic PA imaging studies determined an optimal wavelength (700 nm) for imaging microrobots with embedded Ni particles in oxygenated (fresh) human blood. In addition, we examined the ability of PA imaging to detect the microrobots through a nontransparent tissue mimic and at a depth of 25 mm, where conventional optical methods are unable to be used in tracking the objects. These initial results demonstrate the feasibility of an integrated US and PA imaging method to push the boundaries of microrobot applications into translational applications. Full article