Search

At present, the packing method of "plastic bag–buffer packing–packing paper box" is adopted for bearing packaging. However, the common packing method has a poor packing effect and poor versatility. In this study, a new biomimetic cobweb cushion is proposed to solve the problem of insufficient cushioning capacity of high-precision bearing cushion packaging pads. First, according to the nature of cobweb form, the cobweb cushion structure configuration is determined. Next, based on the structure of the cushion and the relationship between the parameters of radial thread and spiral thread, a mechanical and target optimization model is established. The stress nephogram of bearing and the cobweb cushion are analyzed under three drop heights of 381, 610, and 700 mm, in the finite element simulation software to ensure that the maximum bearings stress is not beyond the material yield strength. Via the 3D printing technology, a cobweb cushion shell cast is made. Drop tests of the bearing were performed, and the results were verified with the finite element simulation analysis. This research can provide technical support for the protection of high-precision bearings from accidental drops during transportation. [ABSTRACT FROM AUTHOR]

During the development of the nervous system, neuronal cells extend axons and dendrites that form complex neuronal networks, which are essential for transmitting and processing information. Understanding the physical processes that underlie the formation of neuronal networks is essential for gaining a deeper insight into higher-order brain functions such as sensory processing, learning, and memory. In the process of creating networks, axons travel towards other recipient neurons, directed by a combination of internal and external cues that include genetic instructions, biochemical signals, as well as external mechanical and geometrical stimuli. Although there have been significant recent advances, the basic principles governing axonal growth, collective dynamics, and the development of neuronal networks remain poorly understood. In this paper, we present a detailed analysis of nonlinear dynamics for axonal growth on surfaces with periodic geometrical patterns. We show that axonal growth on these surfaces is described by nonlinear Langevin equations with speed-dependent deterministic terms and gaussian stochastic noise. This theoretical model yields a comprehensive description of axonal growth at both intermediate and long time scales (tens of hours after cell plating), and predicts key dynamical parameters, such as speed and angular correlation functions, axonal mean squared lengths, and diffusion (cell motility) coefficients. We use this model to perform simulations of axonal trajectories on the growth surfaces, in turn demonstrating very good agreement between simulated growth and the experimental results. These results provide important insights into the current understanding of the dynamical behavior of neurons, the self-wiring of the nervous system, as well as for designing innovative biomimetic neural network models. [ABSTRACT FROM AUTHOR]

Bio-inspired models based on the lobula giant movement detector (LGMD) in the locust's visual brain have received extensive attention and application for collision perception in various scenarios. These models offer advantages such as low power consumption and high computational efficiency in visual processing. However, current LGMD-based computational models, typically organized as four-layered neural networks, often encounter challenges related to noisy signals, particularly in complex dynamic environments. Biological studies have unveiled the intrinsic stochastic nature of synaptic transmission, which can aid neural computation in mitigating noise. In alignment with these biological findings, this paper introduces a probabilistic LGMD (Prob-LGMD) model that incorporates a probability into the synaptic connections between multiple layers, thereby capturing the uncertainty in signal transmission, interaction, and integration among neurons. Comparative testing of the proposed Prob-LGMD model and two conventional LGMD models was conducted using a range of visual stimuli, including indoor structured scenes and complex outdoor scenes, all subject to artificial noise. Additionally, the model's performance was compared to standard engineering noise-filtering methods. The results clearly demonstrate that the proposed model outperforms all comparative methods, exhibiting a significant improvement in noise tolerance. This study showcases a straightforward yet effective approach to enhance collision perception in noisy environments. [ABSTRACT FROM AUTHOR]

The features of the kernel extreme learning machine—efficient processing, improved performance, and less human parameter setting—have allowed it to be effectively used to batch multi-label classification tasks. These classic classification algorithms must at present contend with accuracy and space–time issues as a result of the vast and quick, multi-label, and concept drift features of the developing data streams in the practical application sector. The KELM training procedure still has a difficulty in that it has to be repeated numerous times independently in order to maximize the model's generalization performance or the number of nodes in the hidden layer. In this paper, a kernel extreme learning machine multi-label data classification method based on the butterfly algorithm optimized by particle swarm optimization is proposed. The proposed algorithm, which fully accounts for the optimization of the model generalization ability and the number of hidden layer nodes, can train multiple KELM hidden layer networks at once while maintaining the algorithm's current time complexity and avoiding a significant number of repeated calculations. The simulation results demonstrate that, in comparison to the PSO-KELM, BBA-KELM, and BOA-KELM algorithms, the PSOBOA-KELM algorithm proposed in this paper can more effectively search the kernel extreme learning machine parameters and more effectively balance the global and local performance, resulting in a KELM prediction model with a higher prediction accuracy. [ABSTRACT FROM AUTHOR]

The paper presents an original approach to the synthesis of polycalciumorganyl silsesquioxanes through the reaction of polyorganyl silsesquioxanes [RSiO1.5]n (where R is an ethyl and phenyl radical) with sea urchin skeleton under the conditions of mechanochemical activation. The novelty and practical significance of the present study lies in the use of an available natural raw source as a source of calcium ions to initiate the reaction of calcium silicate formation and create a matrix for the formation of a porous inorganic composite framework. The thermal stability of the introduced silicates, i.e., the ability to maintain a porous structure at high temperatures, is key to the production of an ordered porous material. The reaction scheme was proposed to be based on the interaction of calcium carbonate with the siloxane bond. FTIR, XRD, GPC, and TGA were used to study the composition and structure of the obtained materials. The cross-sectional area of the polymer chain and the volumes of the coherent scattering regions of the polymers obtained were calculated from the XRD data. To prepare the composites, the sea urchin skeleton was further modified with polycalciumorganyl silsesquioxanes in a toluene solution. To remove the sea urchin skeleton, the obtained biomimetic composites were treated with hydrochloric acid. The results of the morphological and surface composition studies are reported. The method proposed in the paper could be of fundamental importance for the possibility of obtaining structured porous composite materials for a wide range of practical applications, including for the purpose of creating a composite that may be a promising carrier for targeted delivery of chemotherapy agents. [ABSTRACT FROM AUTHOR]

This paper introduces a novel method that enables robots to identify objects based on user gaze, tracked via eye-tracking glasses. This is achieved without prior knowledge of the objects' categories or their locations and without external markers. The method integrates a two-part system: a category-agnostic object shape and pose estimator using superquadrics and Siamese networks. The superquadrics-based component estimates the shapes and poses of all objects, while the Siamese network matches the object targeted by the user's gaze with the robot's viewpoint. Both components are effectively designed to function in scenarios with partial occlusions. A key feature of the system is the user's ability to move freely around the scenario, allowing dynamic object selection via gaze from any position. The system is capable of handling significant viewpoint differences between the user and the robot and adapts easily to new objects. In tests under partial occlusion conditions, the Siamese networks demonstrated an 85.2% accuracy in aligning the user-selected object with the robot's viewpoint. This gaze-based Human–Robot Interaction approach demonstrates its practicality and adaptability in real-world scenarios. [ABSTRACT FROM AUTHOR]

Inspired by the biological nervous system, deep neural networks (DNNs) are able to achieve remarkable performance in various tasks. However, they struggle to handle label noise, which can poison the memorization effects of DNNs. Co-teaching-based methods are popular in learning with noisy labels. These methods cross-train two DNNs based on the small-loss criterion and employ a strategy using either "disagreement" or "consistency" to obtain the divergence of the two networks. However, these methods are sample-inefficient for generalization in noisy scenarios. In this paper, we propose CoDC, a novel Co-teaching-basedmethod for accurate learning with label noise via both Disagreement and Consistency strategies. Specifically, CoDC maintains disagreement at the feature level and consistency at the prediction level using a balanced loss function. Additionally, a weighted cross-entropy loss is proposed based on information derived from the historical training process. Moreover, the valuable knowledge involved in "large-loss" samples is further developed and utilized by assigning pseudo-labels. Comprehensive experiments were conducted on both synthetic and real-world noise and under various noise types. CoDC achieved 72.81% accuracy on the Clothing1M dataset and 76.96% (Top1) accuracy on the WebVision1.0 dataset. These superior results demonstrate the effectiveness and robustness of learning with noisy labels. [ABSTRACT FROM AUTHOR]

Depth estimation is an ill-posed problem; objects of different shapes or dimensions, even if at different distances, may project to the same image on the retina. Our brain uses several cues for depth estimation, including monocular cues such as motion parallax and binocular cues such as diplopia. However, it remains unclear how the computations required for depth estimation are implemented in biologically plausible ways. State-of-the-art approaches to depth estimation based on deep neural networks implicitly describe the brain as a hierarchical feature detector. Instead, in this paper we propose an alternative approach that casts depth estimation as a problem of active inference. We show that depth can be inferred by inverting a hierarchical generative model that simultaneously predicts the eyes' projections from a 2D belief over an object. Model inversion consists of a series of biologically plausible homogeneous transformations based on Predictive Coding principles. Under the plausible assumption of a nonuniform fovea resolution, depth estimation favors an active vision strategy that fixates the object with the eyes, rendering the depth belief more accurate. This strategy is not realized by first fixating on a target and then estimating the depth; instead, it combines the two processes through action–perception cycles, with a similar mechanism of the saccades during object recognition. The proposed approach requires only local (top-down and bottom-up) message passing, which can be implemented in biologically plausible neural circuits. [ABSTRACT FROM AUTHOR]

This document is a special issue of the journal Biomimetics, focusing on the design and control of bio-inspired robots. It explores various aspects of bionics in robotics, including robot design, perception, control, and decision-making, as well as incorporating neuroscience and brain science. The issue covers a wide range of topics, such as stiffness adjustment for continuum robots, biomimetic motor control, stroke rehabilitation, reinforcement learning for quadruped robots, improved spiking neural networks, energy-efficient image segmentation, kinematics analysis, synthetic nervous systems for robotic control, online running-gait generation, and bio-inspired perception and navigation for service robots. The document also discusses specific papers within the special issue that address challenges in robotic perception and navigation, legged robot control, and motion control of continuum robots and robotic arms. It concludes by announcing plans for a second special issue on related topics. [Extracted from the article]

The rapid progress of artificial neural networks (ANN) is largely attributed to the development of the rectified linear unit (ReLU) activation function. However, the implementation of software-based ANNs, such as convolutional neural networks (CNN), within the von Neumann architecture faces limitations due to its sequential processing mechanism. To overcome this challenge, research on hardware neuromorphic systems based on spiking neural networks (SNN) has gained significant interest. Artificial synapse, a crucial building block in these systems, has predominantly utilized resistive memory-based memristors. However, the two-terminal structure of memristors presents difficulties in processing feedback signals from the post-synaptic neuron, and without an additional rectifying device it is challenging to prevent sneak current paths. In this paper, we propose a four-terminal synaptic transistor with an asymmetric dual-gate structure as a solution to the limitations of two-terminal memristors. Similar to biological synapses, the proposed device multiplies the presynaptic input signal with stored synaptic weight information and transmits the result to the postsynaptic neuron. Weight modulation is explored through both hot carrier injection (HCI) and Fowler–Nordheim (FN) tunneling. Moreover, we investigate the incorporation of short-term memory properties by adopting polysilicon grain boundaries as temporary storage. It is anticipated that the devised synaptic devices, possessing both short-term and long-term memory characteristics, will enable the implementation of various novel ANN algorithms. [ABSTRACT FROM AUTHOR]

The lack of intuitive controllability remains a primary challenge in enabling transhumeral amputees to control a prosthesis for arm reaching with residual limb kinematics. Recent advancements in prosthetic arm control have focused on leveraging the predictive capabilities of artificial neural networks (ANNs) to automate elbow joint motion and wrist pronation–supination during target reaching tasks. However, large quantities of human motion data collected from different subjects for various activities of daily living (ADL) tasks are required to train these ANNs. For example, the reaching motion can be altered when the height of the desk is changed; however, it is cumbersome to conduct human experiments for all conditions. This paper proposes a framework for cloning motion datasets using deep reinforcement learning (DRL) to cater to training data requirements. DRL algorithms have been demonstrated to create human-like synergistic motion in humanoid agents to handle redundancy and optimize movements. In our study, we collected real motion data from six individuals performing multi-directional arm reaching tasks in the horizontal plane. We generated synthetic motion data that mimicked similar arm reaching tasks by utilizing a physics simulation and DRL-based arm manipulation. We then trained a CNN-LSTM network with different configurations of training motion data, including DRL, real, and hybrid datasets, to test the efficacy of the cloned motion data. The results of our evaluation showcase the effectiveness of the cloned motion data in training the ANN to predict natural elbow motion accurately across multiple subjects. Furthermore, motion data augmentation through combining real and cloned motion datasets has demonstrated the enhanced robustness of the ANN by supplementing and diversifying the limited training data. These findings have significant implications for creating synthetic dataset resources for various arm movements and fostering strategies for automatized prosthetic elbow motion. [ABSTRACT FROM AUTHOR]

Today, most humanoid mechanical fingers use an underactuated mechanism driven by linkages or tendons, with only a single and fixed grasping trajectory. This paper proposes a new multi-mode humanoid finger mechanism based on linkage and tendon fusion transmission, which is embedded with an adjustable-length tendon mechanism to achieve three types of grasping mode. The structural parameters of the mechanism are optimized according to the kinematic and static models. Furthermore, a discussion was conducted on how to set the speed ratio of the linkage driving motor and the tendon driving motor to adjust the length and tension of the tendon, in order to achieve the switching of the shape-adaptive, coupled-adaptive, and variable coupling-adaptive grasping modes. Finally, the multi-mode functionality of the proposed finger mechanism was verified through multiple grasping experiments. [ABSTRACT FROM AUTHOR]

The rapidly evolving field of Virtual Reality (VR)-based Human–Computer Interaction (HCI) presents a significant demand for robust and accurate hand tracking solutions. Current technologies, predominantly based on single-sensing modalities, fall short in providing comprehensive information capture due to susceptibility to occlusions and environmental factors. In this paper, we introduce a novel sensor fusion approach combined with a Long Short-Term Memory (LSTM)-based algorithm for enhanced hand tracking in VR-based HCI. Our system employs six Leap Motion controllers, two RealSense depth cameras, and two Myo armbands to yield a multi-modal data capture. This rich data set is then processed using LSTM, ensuring the accurate real-time tracking of complex hand movements. The proposed system provides a powerful tool for intuitive and immersive interactions in VR environments. [ABSTRACT FROM AUTHOR]

The traditional propeller-based propulsion of underwater robots is inefficient and poorly adapted to practice. By contrast, underwater biomimetic robots show better stability and maneuverability in harsh marine environments. This is particularly true of undulating propulsion biomimetic robots. This paper classifies the existing underwater biomimetic robots and outlines their main contributions to the field. The propulsion mechanisms of underwater biomimetic undulating robots are summarized based on theoretical, numerical and experimental studies. Future perspectives on underwater biomimetic undulating robots are also presented, filling the gaps in the existing literature. [ABSTRACT FROM AUTHOR]

The study of the human visual system not only helps to understand the mechanism of the visual system but also helps to develop visual aid systems to help the visually impaired. As the systematic study of neural signal processing mechanisms in early biological vision continues, the hierarchical structure of the visual system is gradually being dissected, bringing the possibility of building brain-like computational models from a bionic perspective. In this paper, we follow the objective facts of neurobiology and propose a parallel distributed processing computational model of primary visual cortex orientation selection with reference to the complex process of visual signal processing and transmission between the retina to the primary visual cortex, the hierarchical receptive field structure between cells in each layer, and the very fine-grained parallel distributed characteristics of cortical visual computation, which allow for high speed and efficiency. We approach the design from a brain-like chip perspective, map our network model on the field programmable gate array (FPGA), and perform simulation experiments. The results verify the possibility of implementing our proposed model with programmable devices, which can be applied to small wearable devices with low power consumption and low latency. [ABSTRACT FROM AUTHOR]

This paper presents a hydrodynamics study that examines the comparison and collaboration of two swimming modes relevant to the universality of dolphins. This study utilizes a three-dimensional virtual swimmer model resembling a dolphin, which comprises a body and/or caudal fin (BCF) module, as well as a medium and/or paired fin (MPF) module, each equipped with predetermined kinematics. The manipulation of the dolphin to simulate various swimming modes is achieved through the application of overlapping grids in conjunction with the parallel hole cutting technique. The findings demonstrate that the swimming velocity and thrust attained through the single BCF mode consistently surpass those achieved through the single MPF mode and collaborative mode. Interestingly, the involvement of the MPF mode does not necessarily contribute to performance enhancement. Nevertheless, it is encouraging to note that adjusting the phase difference between the two modes can partially mitigate the limitations associated with the MPF mode. To further investigate the potential advantages of dual-mode collaboration, we conducted experiments by increasing the MPF frequency while keeping the BCF frequency constant, thus introducing the concept of frequency ratio (β). In comparison to the single BCF mode, the collaborative mode with a high β exhibits superior swimming velocity and thrust. Although its efficiency experiences a slight decrease, it tends to stabilize. The corresponding flow structure indirectly verifies the favorable impact of collaboration. [ABSTRACT FROM AUTHOR]

The reptile search algorithm is an effective optimization method based on the natural laws of the biological world. By restoring and simulating the hunting process of reptiles, good optimization results can be achieved. However, due to the limitations of natural laws, it is easy to fall into local optima during the exploration phase. Inspired by the different search fields of biological organisms with varying flight heights, this paper proposes a reptile search algorithm considering different flight heights. In the exploration phase, introducing the different flight altitude abilities of two animals, the northern goshawk and the African vulture, enables reptiles to have better search horizons, improve their global search ability, and reduce the probability of falling into local optima during the exploration phase. A novel dynamic factor (DF) is proposed in the exploitation phase to improve the algorithm's convergence speed and optimization accuracy. To verify the effectiveness of the proposed algorithm, the test results were compared with ten state-of-the-art (SOTA) algorithms on thirty-three famous test functions. The experimental results show that the proposed algorithm has good performance. In addition, the proposed algorithm and ten SOTA algorithms were applied to three micromachine practical engineering problems, and the experimental results show that the proposed algorithm has good problem-solving ability. [ABSTRACT FROM AUTHOR]

The unique drive principle and strong manipulation ability of spider legs have led to several bionic robot designs. However, some parameters of bionic actuators still need to be improved, such as torque. Inspired by the hydraulic drive principle of spider legs, this paper describes the design of a bionic actuator characterized by the use of air pressure on each surface and its transmittance in the direction of movement, achieving a torque amplification effect. The produced torque is as high as 4.78 N m. In addition, its torque characteristics during folding motions are similar to those during unfolding motions, showing that the bionic actuator has stable bidirectional drive capability. [ABSTRACT FROM AUTHOR]

Biological fish can always sense the state of water flow and regulate the angle of attack in time, so as to maintain the highest movement efficiency during periodic flapping. The biological adjustment of the caudal fin's angle of attack (AoA) depends on the contraction/relaxation of the tail muscles, accompanying the variation in tail stiffness. During an interaction with external fluid, it helps to maintain the optimal angle of attack during movement, to improve the propulsion performance. Inspired by this, this paper proposes a tail joint motion control scheme based on AoA feedback for the high-speed swimming of bionic dolphins. Firstly, the kinematic characteristics of the designed robot dolphin are analyzed, and the hardware basis is clarified. Second, aiming at the deficiency of the tail motor, which cannot effectively cooperate with the waist joint motor during high-frequency movement, a compensation model for the friction force and latex skin-restoring force is designed, and a joint angle control algorithm based on fuzzy inference is proposed to realize the tracking of the desired joint angle for the tail joint in torque mode. In addition, a tail joint closed-loop control scheme based on angle of attack feedback is proposed to improve the motion performance. Finally, experiments verify the effectiveness of the proposed motion control scheme. [ABSTRACT FROM AUTHOR]

The application of artificial intelligence in everyday life is becoming all-pervasive and unavoidable. Within that vast field, a special place belongs to biomimetic/bio-inspired algorithms for multiparameter optimization, which find their use in a large number of areas. Novel methods and advances are being published at an accelerated pace. Because of that, in spite of the fact that there are a lot of surveys and reviews in the field, they quickly become dated. Thus, it is of importance to keep pace with the current developments. In this review, we first consider a possible classification of bio-inspired multiparameter optimization methods because papers dedicated to that area are relatively scarce and often contradictory. We proceed by describing in some detail some more prominent approaches, as well as those most recently published. Finally, we consider the use of biomimetic algorithms in two related wide fields, namely microelectronics (including circuit design optimization) and nanophotonics (including inverse design of structures such as photonic crystals, nanoplasmonic configurations and metamaterials). We attempted to keep this broad survey self-contained so it can be of use not only to scholars in the related fields, but also to all those interested in the latest developments in this attractive area. [ABSTRACT FROM AUTHOR]

Neuronal networks are complex systems of interconnected neurons responsible for transmitting and processing information throughout the nervous system. The building blocks of neuronal networks consist of individual neurons, specialized cells that receive, process, and transmit electrical and chemical signals throughout the body. The formation of neuronal networks in the developing nervous system is a process of fundamental importance for understanding brain activity, including perception, memory, and cognition. To form networks, neuronal cells extend long processes called axons, which navigate toward other target neurons guided by both intrinsic and extrinsic factors, including genetic programming, chemical signaling, intercellular interactions, and mechanical and geometrical cues. Despite important recent advances, the basic mechanisms underlying collective neuron behavior and the formation of functional neuronal networks are not entirely understood. In this paper, we present a combined experimental and theoretical analysis of neuronal growth on surfaces with micropatterned periodic geometrical features. We demonstrate that the extension of axons on these surfaces is described by a biased random walk model, in which the surface geometry imparts a constant drift term to the axon, and the stochastic cues produce a random walk around the average growth direction. We show that the model predicts key parameters that describe axonal dynamics: diffusion (cell motility) coefficient, average growth velocity, and axonal mean squared length, and we compare these parameters with the results of experimental measurements. Our findings indicate that neuronal growth is governed by a contact-guidance mechanism, in which the axons respond to external geometrical cues by aligning their motion along the surface micropatterns. These results have a significant impact on developing novel neural network models, as well as biomimetic substrates, to stimulate nerve regeneration and repair after injury. [ABSTRACT FROM AUTHOR]

This paper presents a novel facial expression recognition network, called Distract your Attention Network (DAN). Our method is based on two key observations in biological visual perception. Firstly, multiple facial expression classes share inherently similar underlying facial appearance, and their differences could be subtle. Secondly, facial expressions simultaneously exhibit themselves through multiple facial regions, and for recognition, a holistic approach by encoding high-order interactions among local features is required. To address these issues, this work proposes DAN with three key components: Feature Clustering Network (FCN), Multi-head Attention Network (MAN), and Attention Fusion Network (AFN). Specifically, FCN extracts robust features by adopting a large-margin learning objective to maximize class separability. In addition, MAN instantiates a number of attention heads to simultaneously attend to multiple facial areas and build attention maps on these regions. Further, AFN distracts these attentions to multiple locations before fusing the feature maps to a comprehensive one. Extensive experiments on three public datasets (including AffectNet, RAF-DB, and SFEW 2.0) verified that the proposed method consistently achieves state-of-the-art facial expression recognition performance. The DAN code is publicly available. [ABSTRACT FROM AUTHOR]

Inspired by rodents' ability to navigate freely in a given space, bionavigation systems provide alternatives to traditional probabilistic solutions. This paper proposed a bionic path planning method based on RatSLAM to provide a novel viewpoint for robots to make a more flexible and intelligent navigation scheme. A neural network fusing historic episodic memory was proposed to improve the connectivity of the episodic cognitive map. It is biomimetically important to generate an episodic cognitive map and establish a one-to-one correspondence between the events generated by episodic memory and the visual template of RatSLAM. The episodic cognitive map can be improved by imitating the rodents' behavior of memory fusion to produce better path planning results. The experimental results of different scenarios illustrate that the proposed method identified the connectivity between way points, optimized the result of path planning, and improved the flexibility of the system. [ABSTRACT FROM AUTHOR]

Deep Convolutional Neural Networks (CNNs) represent the state-of-the-art artificially intelligent computing models for image classification. The advanced cognition and pattern recognition abilities possessed by humans are ascribed to the intricate and complex neurological connection in human brains. CNNs are inspired by the neurological structure of the human brain and show performance at par with humans in image recognition and classification tasks. On the lower extreme of the neurological complexity spectrum lie small organisms such as insects and worms, with simple brain structures and limited cognition abilities, pattern recognition, and intelligent decision-making abilities. However, billions of years of evolution guided by natural selection have imparted basic survival instincts, which appear as an "intelligent behavior". In this paper, we put forward the evidence that a simple algorithm inspired by the behavior of a beetle (an insect) can fool CNNs in image classification tasks by just perturbing a single pixel. The proposed algorithm accomplishes this in a computationally efficient manner as compared to the other adversarial attacking algorithms proposed in the literature. The novel feature of the proposed algorithm as compared to other metaheuristics approaches for fooling a neural network, is that it mimics the behavior of a single beetle and requires fewer search particles. On the contrary, other metaheuristic algorithms rely on the social or swarming behavior of the organisms, requiring a large population of search particles. We evaluated the performance of the proposed algorithm on LeNet-5 and ResNet architecture using the CIFAR-10 dataset. The results show a high success rate for the proposed algorithms. The proposed strategy raises a concern about the robustness and security aspects of artificially intelligent learning systems. [ABSTRACT FROM AUTHOR]