最新刊期
卷 65 ， 期 1 ， 2026
- 摘要：High-quality underwater visual image input prior information is crucial for underwater robots to accomplish various operational tasks. This paper first analyzes the causes of underwater image degradation，explaining different types of common image degradation from theoretical principles， and introduces two approaches to degradation recovery：underwater image enhancement and underwater camera calibration. Secondly， it systematically reviews the current research status of underwater image enhancement methods，existing underwater datasets，and underwater image quality evaluation systems. Subsequently，it summarizes underwater camera calibration methods， along with their advantages and limitations. Finally， the paper presents future research trends in underwater degradation recovery technology，including the robustness and generalization capability of enhancement algorithms， multi-sensor fusion， and integration with high-level vision tasks.关键词：underwater robot;underwater image enhancement;underwater camera calibration;deep learning478|308|0更新时间：2026-01-19
- 摘要：When the automatic ship paint-removal robot encounters external interference， existing algorithms suffering from performance degradation and insufficient real-time processing capability. To address these challenges，the Repvit-MobileNet block is integrated into the backbone network of YOLOV5 to enhance detection speed. Additionally，the positional attention mechanism has been incorporated after each stage of the backbone network， broadening the model's global receptive field and improving both target localization and interference resistance. Then， a convolutional block attention module（CBAM） is implemented in the neck network， and the feature extraction ability is enhanced by integrating the CBMA module to improve the detection performance of the network model. Lastly，a Refine-Loss loss function is proposed to optimize the geometric relationship between the predicted bounding box and the true bounding box which also balances weight and confidence information related to IOU，leading to improved accuracy in detecting the robot's target position.Subsequent experiments from ship robotic datasets show that the lightweight YOLOV5 network combining Repvit-MobileNet block and attention mechanism can reach 84.1% in the experiment with average precision， and the inference speed on the edge device reaches 26.6 f/s， which meets the need of industrial applications for object detection of ship paint-removal robots.关键词：paint-removal robot;light weight;attention mechanism;object detection336|628|0更新时间：2026-01-19
- 摘要：This paper proposes an improved method that integrates visual odometry to address the defect of traditional LIO-SLAM being affected by sparse laser features and dynamic occlusion in complex indoor environments，resulting in decreased positioning accuracy. Specifically，while maintaining the LIO-SLAM laser inertial tight coupling framework，ORB-SLAM is introduced as an independent visual odometry module to provide high-frequency and rich texture visual constraint information for the system. By using an adaptive weight fusion strategy， multi-source optimization of laser， inertial， and visual observations is achieved， enhancing robustness in environments with weak geometric constraints and rich textures but complex structures. The experiment was conducted in various typical indoor scenarios（corridors， open halls， and dynamic crowd environments）， and the results showed that the overall trajectory error was reduced to 70% of the original system， compared to the original LIO-SLAM. This study validates the feasibility and effectiveness of visual laser inertial multimodal fusion in complex indoor environments， providing new ideas for high-precision indoor autonomous positioning and map construction.关键词：indoor positioning;LIO-SLAM;ORB-SLAM;visual odometery;multi-sensor fusion307|409|0更新时间：2026-01-19
- 摘要：This paper designs a leg structure of a wheel-legged robot with strong adaptability and power switching capability based on planetary gear systems. The optimized modified cycloidal trajectory planning and wheel-legged switching foot-end trajectory planning are introduced. The position， speed and acceleration curves in different directions are continuous， smooth and without sudden changes， meeting the control requirements of the robot. Based on the virtual model control method， through WEBOTS platform simulation and single-leg physical prototype experiments， it is found that the deviation of the swing trajectory is concentrated at the end position of the swing phase and the highest point of the lifting leg， respectively， about 20 mm and 5 mm； the maximum deviation of the repeated positioning of the mechanism is 4 mm， which is only about 22.2% of the radius of the foot-end components （18 mm）； the average error of the wheel-legged switching foot-end trajectory in the X direction is approximately 2.4 mm， and of 1.26 mm in the Y direction， which can complete the posture switching task. The above conclusions provides better guidance for the design of wheel-legged composite robots.关键词：wheel-legged robots;power switching;trajectory planning;virtual model control491|456|0更新时间：2026-01-19
- 摘要：This paper presents a hybrid control strategy combining feedforward compensation and fuzzy control. The approach leverages an accurate dynamic model for feedforward control and incorporates fuzzy logic to enhance overall controller performance. Key vibration sources during robot operation are analyzed， and trajectory accuracy， as defined by ISO9283， is adopted as the evaluation metric. The proposed fuzzy feedforward controller is compared with traditional PD control and feedforward-compensated PD control. Simulation results demonstrate that the proposed method significantly improves trajectory tracking accuracy and vibration suppression， validating its effectiveness.关键词：articulated robot;dynamic model;feedforward control;fuzzy control;vibration suppression230|331|0更新时间：2026-01-19
Column of vision perception and control of robot(Contributing editor:YUAN Xiaofang )
- 摘要：This study introduces a front-mounted thruster as an auxiliary propulsion mechanism，multiple front thruster configurations were designed and systematically analyzed to evaluate their impact on the vehicle’s propulsion performance. A propeller body-force method combined with an overset grid approach was employed to perform two-degree-of-freedom self-propulsion simulations of the vehicle equipped with these thrusters. Based on this， a comprehensive investigation was conducted into the effects of thruster installation position and angle on the vehicle’s hydrodynamic performance. The analysis encompassed the entire operational process from start-up acceleration to steady cruising， focusing on velocity， lift coefficient， drag coefficient， lift-to-drag ratio， and pitching moment coefficient， as well as flow field， pressure distribution， and vortex structures during steady-state operation. The results demonstrate that optimizing the placement and orientation of front thrusters can significantly improve the vehicle’s sailing efficiency. Notably， adjusting the thruster installation angle effectively enhances the lift-to-drag ratio and stability of the underwater vehicle..关键词：Manta ray-inspired underwater vehicle;front-mounted thruster;body-force method;sailing efficiency75|89|0更新时间：2026-01-19
- 摘要：This paper proposes an Improved Grey Wolf Optimization-based Strong Tracking Cubature Kalman Particle Filter algorithm（IGWO-STCPF）.The proposed method first employs a Strong Tracking Cubature Kalman Filter（STCKF） to incorporate measurement information for dynamically adjusting the particle mean and covariance， thereby enhancing the effectiveness of importance sampling. Then， an entropy-weighted GWO is introduced into the resampling stage to mitigate particle degeneration and improve estimation accuracy. Simulation results demonstrate that， compared with STCKF，PF，PSO-PF，and PSO-CPF algorithms， the proposed IGWO-STCPF improves trajectory estimation accuracy by 13.41%，18.58%，21.86%，and 21.33%，respectively. These results confirm the robustness and effectiveness of the proposed method in complex underwater scenarios.关键词：autonomous underwater vehicle;PF;STCKF;GWO;information entropy47|72|0更新时间：2026-01-19
- 摘要：This study proposes a semantic information-based aerial image stitching methodology. In the first stage， a coarse registration method using projection transformation based on 3D vector rotation is adopted. Through geographic coordinate conversion， view frustum modeling， plane intersection calculation， and projection matrix estimation， efficient coarse registration of aerial images onto a unified elevation plane is achieved. In the second stage， a semantic segmentation and LightGlue feature based optimization method is utilized. Each row of images after coarse registration is treated as an independent optimization unit： road topological structures are obtained via semantic segmentation， and high-precision registration is realized by combining a lightweight feature matching network with geometric constraint optimization. Experimental results show that the proposed two-stage optimization model effectively overcomes the dependence of camera intrinsics. Finally， the accurate disaster-stricken locations are visualized on the geographic system interface， enabling comprehensive and systematic recording of disaster characteristics.关键词：fire rescue;unmanned aerial vehicle;orthophoto;semantic segmentation76|110|0更新时间：2026-01-19
Column of unmanned system (Contributing editor: WANG Hairong)
- 摘要：This paper proposes a vehicle risk analysis intelligent agent framework integrating knowledge graph technology and large language model（LLM） guided by chain-of-thought（CoT）. we construct a risk knowledge graph tailored for critical vehicle safety supervision to unify fragmented data including risky driving behaviors， traffic accidents， and road network information into a cohesive semantic network. And a hierarchical CoT reasoning architecture grounded in prompt engineering is designed to steer the LLM through a traceable inference pipeline comprising query comprehension，data retrieval，result validation and query refinement，thereby enhancing the accuracy and robustness of question-answering systems. Experimental validation confirms that the proposed intelligent agent of vehicle risk analysis effectively facilitates natural language-driven risk assessment and association mining across diverse complex query scenarios，delivering an efficient，intelligent，and interpretable decision-support tool for traffic management with demonstrable superiority in operational efficacy and analytical transparency.关键词：vehicle;risk analysis;knowledge graph;LLM;chain-of-thought;intelligent agent53|71|0更新时间：2026-01-19
- 摘要：The numerical blow-up caused by mass non-conservation in existing free-interface capturing algorithms stems from asynchronous density-flow solving. To overcome this problem，the numerical relationship is derived between local mass non-conservation and the diagonally dominance. Based on this derivation， a dynamic relaxation solution strategy is constructed， where the relaxation coefficient is dynamically updated in mass-non-conserved cells，to ensure the diagonal dominance of the momentum equation.Combined with the selection of an implicit high-resolution face-centered density interpolation method，a refined free-surface capturing algorithm is proposed.The performance of the refined algorithm is validated through classical dam-break free-surface flow test cases.The results demonstrate that the dynamic relaxation strategy effectively enhances numerical robustness and computational efficiency，while the implicit face-centered density interpolation method significantly reduces numerical diffusion，with particularly notable performance improvements in scenarios involving large density ratios and Courant numbers.关键词：free interface;volume of fluid;relaxation factor;high resolution interpolation75|97|0更新时间：2026-01-19
- 摘要：To address the spatial-temporal prediction of urban electric vehicle charging load under multi-factor influences，a fusion model combining spatial-temporal graph convolutional and Squeeze-and-Excitation network（STGC-SENet） was proposed. A periodic temporal segment extraction block was constructed to acquire features across recent，daily and weekly scales. Temporal dependencies and spatial topological correlations were synchronously modeled through spatio-temporal graph convolution. A Squeeze-and-Excitation channel attention layer was introduced to recalibrate feature channels， dynamically enhancing key feature. Experiments based on real-world urban charging datasets demonstrate that： compared to baseline model（LTSM， MSTGCN， ASTGCN），STGC-SENet reduced the mean absolute error by 3.36， 0.56 and 1.10， respectively. Regarding factor sensitivity， incorporating the daily periodicity factor demonstrates the most significant prediction improvement compared to recent and weekly temporal segments. Under historical charging load inputing， integrating charging pile occupancy rate and real-time electricity price yields optimal prediction performance，with MAE decreasing to 5.08；conversely， incorporating the number of charging piles and meteorological factors did not yield significant prediction improvement.关键词：electric vehicle;charging load;multi-factor fusion;spatial-temporal correlation;graph convolutional network;multi-channel attention184|265|0更新时间：2026-01-19
- 摘要：This paper investigates the chemical constituents of Yunnan black tea trichomes from Fengqing county. Using various chromatographic techniques including silica gel， Sephadex LH-20， C18 reverse-phase， D101 macroporous adsorption resin column chromatography， and semi-preparative HPLC， ten compounds were isolated from the ethyl acetate fraction of a 95% ethanol extract of Yunnan black tea trichomes. Their structures were identified via spectroscopic methods （NMR， MS， and ECD） as： Caffeine （1）， gallic acid （2）， kaempferol （3）， epicatechin （4）， （+）-catechin （5）， isoquercitrin （6）， （-）-epicatechin-3-gallate （7）， pyrrolidinone H （8）， tiliroside （9）， kaempferol-3-O-［2，6-di-O-（E）-p-coumaroyl］-β-D-glucopyranoside （10）， respectively. Preliminary content analysis indicated a relatively high caffeine content （0.64% of dry weight）， while tea polyphenols accounted for 0.03% of dry weight， suggesting Yunnan black tea trichomes warrants further development.关键词：Yunnan black tea;tea trichome;chemical component;caffeine;tea polyphenols108|224|0更新时间：2026-01-19
- 摘要：This study aimed to investigate the structural characteristics of the plankton community in Huangpi Lake and their relationships with environmental factors.This study conducted quarterly plankton surveys at 15 sampling sites in a lake wetland area during 2024.The correlations between plankton species composition and environmental variables were examined using redundancy analysis （RDA） and Pearson correlation analysis to gain deeper insight into the community-environment interactions.Across the four seasons，a total of 200 phytoplankton species from 8 phyla and 84 genera were identified.Diatoms represented the most abundant phylum （33.5%），followed by green algae （26.0%）.Phytoplankton density exhibited seasonal variation in the order：summer>spring>autumn>winter.The dominant phytoplankton species included Navicula simplex，Cyclotella comensis，and Phormidium tenue. A total of 59 zooplankton species belonging to 17 families and 34 genera were recorded， with rotifers comprising the dominant group （62.7%）. Zooplankton density was highest in summer， followed by winter， spring， and autumn. The predominant zooplankton species were Polyarthra trigla，Microcyclops varicans，and Bosmina fatalis.The mean Shannon-Wiener diversity index for phytoplankton was 2.06，the Pielou evenness index was 0.71， and the Margalef richness index was 3.45.Integrated assessment indicated that Huangpi Lake exhibits high species richness，a complex community structure， and generally light to moderate pollution levels，reflecting an intermediate water quality status. RDA and correlation analyses revealed that the phytoplankton community was primarily influenced by water temperature， chemical oxygen demand and total phosphorus.The zooplankton community showed strong correlations with nutrient concentrations，dissolved oxygen，and water temperature.关键词：Huangpi Lake;plankton;community structure;environmental factor;correlation analysis130|426|0更新时间：2026-01-19
- 摘要：A predator-prey system with fear effect and prey refuge is proposed and studied. First， we prove the boundedness of the solution by comparison principle. Then， we obtain the existence and stability conditions of the equilibrium points. Then， the conditions for the system to undergo transcritical bifurcation and Hopf bifurcation at the boundary equilibrium point and internal equilibrium point are analyzed， respectively. Finally， the previous theoretical results are verified by numerical simulation.关键词：predator-prey system;fear effect;prey refuge;transcritical bifurcation;Hopf bifurcation90|88|0更新时间：2026-01-19
- 摘要：A neighbor full sum distinguishing total coloring of a graph
- 摘要：The existence， uniqueness， and finite-time stability of solutions are investigated for a class of fractional hybrid （p，q）-integral-difference systems. Firstly， sufficient conditions for both the uniqueness and existence of solutions are derived using Banach's contraction mapping principle and Krasnoselskii’s fixed point theorem， respectively. Subsequently， the finite-time stability of the system solutions is rigorously verified under specified conditions. To demonstrate the theoretical results， a numerical example is presented along with computational results to validate the feasibility and practical applicability of the main theoretical findings.关键词：fractional （p，q）-integral;fractional （p，q）-difference;initial value problem;fixed point theorem102|57|0更新时间：2026-01-19
- 摘要：We investigate a class of non-integrable two-particle Calogero-Moser systems modulated by a power-law external potential. The local well-posedness of the Cauchy problem is established under the strict initial separation condition for the particles. For suitably prepared initial configurations, local solutions can be extended globally via energy conservation; conversely, negative energy conditions induce (in)finite-time blowup. The linear (in)stability of stationary solutions is analyzed, with their energy serving as a threshold. Numerical investigations employ a fourth-order Runge-Kutta scheme with adaptive step-size control. Simulations demonstrate that the trajectories either converge to steady states or exhibit blowup, depending on the power exponent
Research articles
- Postal code：510275
- Tel：020-84112585，84113223 Email：xuebaozr@mail.sysu.edu.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0