Search Results (29)

Soil compaction in grassland and agricultural soils reduces water infiltration, root growth and ecosystem services. Conventional deep tillage and coring can alleviate compaction but are energy intensive and strongly disturb the turf. This study proposes an earthworm-inspired subsurface robot as a low-disturbance loosening tool for compacted grassland soils. Design principles are abstracted from earthworm body segmentation, anchoring–propulsion peristaltic locomotion and corrugated body surface, and mapped onto a robotic body with anterior and posterior telescopic units, a flexible mid-body segment, a corrugated outer shell and a brace-wire steering mechanism. Kinematic simulations evaluate the peristaltic actuation mechanism and predict a forward displacement of approximately 15 mm/cycle. Using the finite element method and a Modified Cam–Clay soil model, different linkage layouts and outer-shell geometries are compared in terms of radial soil displacement and drag force in cohesive loam. The optimised corrugated outer shell combining circumferential and longitudinal waves lowers drag by up to 20.1% compared with a smooth cylinder. A 3D-printed prototype demonstrates peristaltic locomotion and steering in bench-top tests. The results indicate the potential of earthworm-inspired subsurface robots to provide low-disturbance loosening in conservation agriculture and grassland management, and highlight the need for field experiments to validate performance in real soils. Full article

The autonomous navigation of inspection robots in complex forest environments heavily relies on accurate trunk detection. However, existing detection models struggle to achieve both high accuracy and real-time performance on resource-constrained edge devices. To address this challenge, this study proposes a lightweight algorithm named YOLOv11-TrunkLight. The core innovations of the algorithm include (1) a novel StarNet_Trunk backbone network, which replaces traditional residual connections with element-wise multiplication and incorporates depthwise separable convolutions, significantly reducing computational complexity while maintaining a large receptive field; (2) the C2DA deformable attention module, which effectively handles the geometric deformation of tree trunks through dynamic relative position bias encoding; and (3) the EffiDet detection head, which improves detection speed and reduces the number of parameters through dual-path feature decoupling and a dynamic anchor mechanism. Experimental results demonstrate that compared to the baseline YOLOv11 model, our method improves detection speed by 13.5%, reduces the number of parameters by 34.6%, and decreases computational load (FLOPs) by 39.7%, while the average precision (mAP) is only marginally reduced by 0.1%. These advancements make the algorithm particularly suitable for deployment on resource-constrained edge devices of inspection robots, providing reliable technical support for intelligent forestry management. Full article

Currently, Japan’s fishing industry is facing a severe decline in its workforce. As a response, fishing mechanization using small underwater robots is promoted. These robots offer advantages due to their compact size, although their operating time is limited. A major source of this limited operating time is posture stabilization, which requires continuous thruster use and rapidly drains the battery. To reduce power consumption, anchoring the robot to the seabed with anchors is proposed. However, due to neutral buoyancy, the available thrust is limited, making penetration into the seabed difficult and reducing stability. To address this, we focus on composite-shaped anchors and vibration. The anchors combine a conical tip and a cylindrical shaft to achieve both penetrability and holding force. However, a trade-off exists between these functions depending on the tip angle; anchors with larger angles provide better holding capacity but lower penetrability. To overcome this limitation, vibration is applied to reduce soil resistance and facilitate anchor penetration. While vibration is known to aid penetration in saturated soft soils, the effect of tip angle under such conditions remains unclear. This study aims to clarify the optimal tip angle for achieving sufficient penetration and holding performance under vibratory conditions. Experiments in underwater saturated soft soil showed that vibration improves both penetration and holding. This effect was strong in anchors with tip angles optimized for holding force. These findings support the development of energy-efficient anchoring systems for autonomous underwater operations in soft seabed environments. Full article

As large language models (LLMs) and vision–language models (VLMs) become increasingly used in robotics area, a crucial question arises: to what extent do these models replicate human-like cognitive processes, particularly within socially interactive contexts? Whilst these models demonstrate impressive multimodal reasoning and perception capabilities, their cognitive plausibility remains underexplored. In this study, we address this gap by using human visual attention as a behavioural proxy for cognition in a naturalistic human-robot interaction (HRI) scenario. Eye-tracking data were previously collected from participants engaging in social human-human interactions, providing frame-level gaze fixations as a human attentional ground truth. We then prompted a state-of-the-art VLM (LLaVA) to generate scene descriptions, which were processed by four LLMs (DeepSeek-R1-Distill-Qwen-7B, Qwen1.5-7B-Chat, LLaMA-3.1-8b-instruct, and Gemma-7b-it) to infer saliency points. Critically, we evaluated each model in both stateless and memory-augmented (short-term memory, STM) modes to assess the influence of temporal context on saliency prediction. Our results presented that whilst stateless LLaVA most closely replicates human gaze patterns, STM confers measurable benefits only for DeepSeek, whose lexical anchoring mirrors human rehearsal mechanisms. Other models exhibited degraded performance with memory due to prompt interference or limited contextual integration. This work introduces a novel, empirically grounded framework for assessing cognitive plausibility in generative models and underscores the role of short-term memory in shaping human-like visual attention in robotic systems. Full article

The increasing presence of high-rise buildings with curved and convex facades poses significant challenges for facade-cleaning robots, particularly in terms of mobility and anchoring. To address this, we propose a rope-riding mobile anchor (RMA) system capable of repositioning the anchor point of a cleaning robot on convex building surfaces. The RMA travels horizontally along a roof-mounted nylon rope using caterpillar tracks with U-shaped grooves, and employs a four-bar linkage mechanism to fix its position securely by increasing rope contact friction. This structural principle was selected for its simplicity, stability under heavy loads, and efficient actuation. Experimental results show that the RMA can support a payload of 50.5 kg without slippage under tensions up to 495.24 N, and contributes to reducing the power consumption of the cleaning robot during operation. These findings demonstrate the RMA’s effectiveness in extending the robot’s working range and enhancing safety and stability in facade-cleaning tasks on complex curved surfaces. Full article

Kirigami-inspired geometries offer a lightweight, bioinspired strategy for friction enhancement and anchoring in soft robotics. This study presents a bioinspired kirigami structure designed to enhance the anchoring performance of soft robotic systems through systematic geometric and actuation parameter optimization. Drawing inspiration from the anisotropic friction mechanisms observed in reptilian scales, we integrated linear, triangular, trapezoidal, and hybrid kirigami cuts onto flexible plastic sheets. A compact 12 V linear actuator enabled cyclic actuation via a custom firmware loop, generating controlled buckling and directional friction for effective locomotion. Through experimental trials, we quantified anchoring efficiency using crawling distance and stride metrics across multiple cut densities and actuation conditions. Among the tested configurations, the triangular kirigami with a 4 × 20 unit density on 100 µm PET exhibited the most effective performance, achieving a stride efficiency of approximately 63% and an average crawling speed of ~47 cm/min under optimized autonomous operation. A theoretical framework combining buckling mechanics and directional friction validated the observed trends. This study establishes a compact, tunable anchoring mechanism for soft robotics, offering strong potential for autonomous exploration in constrained environments. Full article

Small object detection in traffic scenes presents unique challenges for mobile robots operating under constrained computational resources and highly dynamic environments. Unlike general object detection, small targets often suffer from low resolution, weak semantic cues, and frequent occlusion, especially in complex outdoor scenarios. This study systematically analyses the challenges, technical advances, and deployment strategies for small object detection tailored to mobile robotic platforms. We categorise existing approaches into three main strategies: feature enhancement (e.g., multi-scale fusion, attention mechanisms), network architecture optimisation (e.g., lightweight backbones, anchor-free heads), and data-driven techniques (e.g., augmentation, simulation, transfer learning). Furthermore, we examine deployment techniques on embedded devices such as Jetson Nano and Raspberry Pi, and we highlight multi-modal sensor fusion using Light Detection and Ranging (LiDAR), cameras, and Inertial Measurement Units (IMUs) for enhanced environmental perception. A comparative study of public datasets and evaluation metrics is provided to identify current limitations in real-world benchmarking. Finally, we discuss future directions, including robust detection under extreme conditions and human-in-the-loop incremental learning frameworks. This research aims to offer a comprehensive technical reference for researchers and practitioners developing small object detection systems for real-world robotic applications. Full article

Anchor–bolt support operations are lengthy and conducted under harsh conditions, restricting the efficiency and safety of roadway excavation. To address these challenges, we developed an integrated solution combining mechanical structure optimization with control algorithms. Specifically, we designed a novel automated drilling system equipped with a robotic manipulator and an anchor–bolt magazine to handle modular hollow self-drilling anchor bolts, enabling automated support operations. To achieve precise docking in unmanned conditions, we employed an inner-loop fractional-order proportional–integral–derivative (FOPID) controller optimized by an improved particle swarm optimization (ILPSO) algorithm. Additionally, robust control based on H∞ control theory was introduced to ensure reliable system performance under disturbances and model uncertainties. Simulation results indicate that the ILPSO-tuned FOPID controller significantly outperforms conventional controllers in dynamic response accuracy; frequency–domain analysis further confirms that the H∞ control approach enhances system stability. Collectively, these results provide a theoretical basis for advancing automated mining technologies. Full article

To address the challenge of precise picking point localization in morphologically diverse safflower plants, this study proposes PointSafNet—a novel three-stage 3D point cloud analysis framework with distinct architectural and methodological innovations. In Stage I, we introduce a multi-view reconstruction pipeline integrating Structure from Motion (SfM) and Multi-View Stereo (MVS) to generate high-fidelity 3D plant point clouds. Stage II develops a dual-branch architecture employing Star modules for multi-scale hierarchical geometric feature extraction at the organ level (filaments and frui balls), complemented by a Context-Anchored Attention (CAA) mechanism to capture long-range contextual information. This synergistic feature learning approach addresses morphological variations, achieving 86.83% segmentation accuracy (surpassing PointNet++ by 7.37%) and outperforming conventional point cloud models. Stage III proposes an optimized geometric analysis pipeline combining dual-centroid spatial vectorization with Oriented Bounding Box (OBB)-based proximity analysis, resolving picking coordinate localization across diverse plants with 90% positioning accuracy and 68.82% mean IoU (13.71% improvement). The experiments demonstrate that PointSafNet systematically integrates 3D reconstruction, hierarchical feature learning, and geometric reasoning to provide visual guidance for robotic harvesting systems in complex plant canopies. The framework’s dual emphasis on architectural innovation and geometric modeling offers a generalizable solution for precision agriculture tasks involving morphologically diverse safflowers. Full article

At present, China’s coal mine permanent tunneling support commonly uses mechanized drilling and anchoring equipment; there are low support efficiency, labor intensity, and other issues. In order to further improve the support efficiency and liberate productivity, this paper further researches the trajectory planning of the drilling and anchoring materials of the robotic arm for the drilling machine “grasping–carrying–loading–unloading” on the basis of the drilling and anchoring robotic system designed by the team in the previous stage. Firstly, the kinematic model of the robotic arm with material was established by improving the D-H parameter method. Then, the working space of the robotic arm with the material was analyzed using the Monte Carlo method. The singular bit-shaped region of the robotic arm was restricted, and obstacles were removed from the working space. The inverse kinematics was utilized to solve the feasible domain of the robotic arm with material. Secondly, in order to avoid blind searching, the guidance of the Bi-RRT algorithm was improved by adding the target guidance factor, and the two-way tree connection strategy for determining the feasible domain was combined with the Bi-RRT algorithm’s feasible domain judgment bi-directional tree connection strategy to improve the convergence speed of the Bi-RRT algorithm. Then, in order to adapt to the dynamic environment and avoid the global planning algorithm from falling into the local minima, on the basis of the above planning methods, an improved Bi-RRT trajectory planning algorithm incorporating the artificial potential field was proposed, which takes the planned paths as the guiding potential field of the artificial potential field to make full use of the global information and avoid falling into the local minimization. Finally, a simulation environment was built in a ROS environment to compare and analyze the planning effect of different algorithms. The simulation results showed that the improved Bi-RRT trajectory planning algorithm incorporating the artificial potential field improved the optimization speed by 69.8% and shortened the trajectory length by 46.6% compared with the traditional RRT algorithm. Full article

As a full tension structural system, the spoke-type single-layer cable net structure has a light graceful shape and superior mechanical properties. During construction, the structure will gradually be tensioned from the flexible unstressed state to the formed state with stiffness, and the structural configuration changes greatly, making construction difficult. This study focused on the spoke-type single-layer cable net structure of the Linyi Olympic Sports Center Stadium. The structural finite element model was established in ANSYS, and the construction scheme was selected and simulated using the nonlinear dynamic finite element method (NDFEM). Most of the existing structural automatic measuring systems are suitable for measuring points with gentle deformation. However, there is the lack of a stable and reliable automatic configuration monitoring system for the construction of single-layer cable net structures. Based on the Lecia TS16 robotic total station (RTS), the configuration automatic monitoring system (CAMS) was developed to obtain the coordinate data of key nodes on the ring cable and compression ring during the construction process. The original finite element model of clamps was refined to obtain the corresponding data in ANSYS. The results indicate that the selected construction scheme has a rational mechanical response according to the finite element simulation. The radial cable force when anchoring the traction cables is smaller than or equal to that in the formed state, which proves that the construction method of anchoring in batches is safe. The results of the ANSYS simulation is basically consistent with those obtained by CAMS, proving that the simulation method is credible. CAMS has good stability and measurement accuracy and can achieve the automatic monitoring of large structural deformation. The research findings provide valuable guidance for practical construction and other similar projects. Full article

In response to the challenges of multiple personnel, heavy support tasks, and high labor intensity in coal mine tunnel drilling and anchoring operations, this study proposes a novel tracked drilling and anchoring robot. The robot is required to maintain alignment with the centerline of the tunnel during operation. However, owing to the effects of skidding and slipping between the track mechanism and the floor, the precise control of a drilling and anchoring robot in tunnel environments is difficult to achieve. Through an analysis of the body and track mechanisms of the drilling and anchoring robot, a kinematic model reflecting the pose, steering radius, steering curvature, and angular velocity of the drive wheel of the drilling and anchoring robot was established. This facilitated the determination of speed control requirements for the track mechanism under varying driving conditions. Mathematical models were developed to describe the relationships between a tracked drilling and anchoring robot and several key factors in tunnel environments, including the minimum steering space required by the robot, the minimum relative steering radius, the steering angle, and the lateral distance to the sidewalls. Based on these models, deviation-correction control strategies were formulated for the robot, and deviation-correction path planning was completed. In addition, a PID motion controller was developed for the robot, and trajectory-tracking control simulation experiments were conducted. The experimental results indicate that the tracked drilling and anchoring robot achieves precise control of trajectory tracking, with a tracking error of less than 0.004 m in the x-direction from the tunnel centerline and less than 0.001 m in the y-direction. Considering the influence of skidding, the deviation correction control performance test experiments of the tracked drilling and anchoring robot at dy = 0.5 m away from the tunnel centerline were completed. In the experiments, the tracked drilling and anchoring robot exhibited a significant difference in speed between the two sides of the tracks with a track skid rate of 0.22. Although the real-time tracking maximum error in the y-direction from the tunnel centerline was 0.13 m, the final error was 0.003 m, meeting the requirements for position deviation control of the drilling and anchoring robot in tunnel environments. These research findings provide a theoretical basis and technical support for the intelligent control of tracked mobile devices in coal mine tunnels, with significant theoretical and engineering implications. Full article

A drill-anchor robot is an essential means of efficient drilling and anchoring in coal-mine roadways. It is significant to calculate the position of the drill-anchor robot based on the positioning information of the supported anchor rod to improve tunneling efficiency. Therefore, identifying and positioning the supported anchor rod has become a critical problem that needs to be solved urgently. Aiming at the problem that the target in the image is blurred and cannot be accurately identified due to the low and uneven illumination environment, we proposed an improved YOLOv7 (the seventh version of the You Only Look Once) model based on the fusion of image enhancement and multiattention mechanism, and the self-made dataset is used for testing and training. Aiming at the problem that the traditional positioning method cannot guarantee accuracy and efficiency simultaneously, an anchor-rod positioning method using depth image and RGB image alignment combined with least squares linear fitting is proposed, and the positioning accuracy is improved by processing the depth map. The results show that the improved model improves the mAP by 5.7% compared with YOLOv7 and can accurately identify the target. Through the positioning method proposed in this paper, the error between the positioning coordinate and the measurement coordinate of the target point on each axis does not exceed 11 mm, which has high positioning accuracy and improves the positioning accuracy and robustness of the anchor rod in the coal-mine roadway. Full article

Lotus pods in unstructured environments often present multi-scale characteristics in the captured images. As a result, it makes their automatic identification difficult and prone to missed and false detections. This study proposed a lightweight multi-scale lotus pod identification model, MLP-YOLOv5, to deal with this difficulty. The model adjusted the multi-scale detection layer and optimized the anchor box parameters to enhance the small object detection accuracy. The C3 module with transformer encoder (C3-TR) and the shuffle attention (SA) mechanism were introduced to improve the feature extraction ability and detection quality of the model. GSConv and VoVGSCSP modules were adopted to build a lightweight neck, thereby reducing model parameters and size. In addition, SIoU was utilized as the loss function of bounding box regression to achieve better accuracy and faster convergence. The experimental results on the multi-scale lotus pod test set showed that MLP-YOLOv5 achieved a mAP of 94.9%, 3% higher than the baseline. In particular, the model’s precision and recall for small-scale objects were improved by 5.5% and 7.4%, respectively. Compared with other mainstream algorithms, MLP-YOLOv5 showed more significant advantages in detection accuracy, parameters, speed, and model size. The test results verified that MLP-YOLOv5 can quickly and accurately identify multi-scale lotus pod objects in complex environments. It could effectively support the harvesting robot by accurately and automatically picking lotus pods. Full article

Ultra-wideband (UWB) systems promise centimetre-level accuracy for indoor positioning, yet they remain susceptible to non-line-of-sight (NLOS) errors due to complex indoor environments. A fusion mechanism that integrates the UWB with an odometer sensor is introduced to address this challenge and achieve a high positioning accuracy. A sliding window method is applied to identify NLOS anchors effectively. The modified UWB-only positioning has an average error under 13 cm with an RMSE of 16 cm. Then, a loosely coupled approach named Dynamic Dimension Fusion (DDF) is designed to mitigate the odometer’s cumulative errors that achieve a remarkable average error and RMSE below 5 cm, notably superior to established unscented Kalman filter (UKF) fusion techniques. DDF utilises UWB data to correct the one-dimensional heading error of the odometer when the robot moves in a straight line and to correct both heading and mileage in two dimensions when the robot is turning. Comprehensive real-world experimental evaluations underscore the efficacy and robustness of this novel approach. Full article