Position Of Point Research Articles

A data density-based measure of dexterity for continuum robots and its comparative study

Abstract Continuum robot-based surgical systems are becoming an effective tool for minimally invasive surgery. A flexible, dexterous, and compact robot structure is suitable for carrying out complex surgical operations. In this paper, we propose performance metrics for dexterity based on data density. Data density at a point in the workspace is higher if the number of reachable points is higher, with a unique configuration lying in a small square box around a point. The computation of these metrics is performed with forward kinematics using the Monte Carlo method and, hence, is computationally efficient. The data density at a particular point is a measure of dexterity at that point. In contrast, the dexterity distribution property index is a measure of how well dexterity is distributed across the workspace according to desired criteria. We compare the dexterity distribution property index across the workspace with the dexterity index based on the dexterous solid angle and manipulability-based approach. A comparative study reveals that the proposed method is simple and straightforward because it uses only the position of the reachable point as the input parameter. The method can quantify and compare the performance of different geometric designs of hyper-redundant and multisegment continuum robots based on dexterity.

Empirical Evaluation and Simulation of the Impact of Global Navigation Satellite System Solutions on Uncrewed Aircraft System–Structure from Motion for Shoreline Mapping and Charting

Uncrewed aircraft systems (UASs) and structure-from-motion/multi-view stereo (SfM/MVS) photogrammetry are efficient methods for mapping terrain at local geographic scales. Traditionally, indirect georeferencing using ground control points (GCPs) is used to georeference the UAS image locations before further processing in SfM software. However, this is a tedious practice and unsuitable for surveying remote or inaccessible areas. Direct georeferencing is a plausible alternative that requires no GCPs. It relies on global navigation satellite system (GNSS) technology to georeference the UAS image locations. This research combined field experiments and simulation to investigate GNSS-based post-processed kinematic (PPK) as a means to eliminate or reduce reliance on GCPs for shoreline mapping and charting. The study also conducted a brief comparison of real-time network (RTN) and precise point positioning (PPP) performances for the same purpose. Ancillary experiments evaluated the effects of PPK base station distance and GNSS sample rate on the accuracy of derived 3D point clouds and digital elevation models (DEMs). Vertical root mean square errors (RMSEz), scaled to the 95% confidence interval using an assumption of normally-distributed errors, were desired to be within 0.5 m to satisfy National Oceanic and Atmospheric Administration (NOAA) requirements for nautical charting. Simulations used a Monte Carlo approach and empirical tests to examine the influence of GNSS performance on the quality of derived 3D point clouds. RTN and PPK results consistently yielded RMSEz values within 10 cm, thus satisfying NOAA requirements for nautical charting. PPP did not meet the accuracy requirements but showed promising results that prompt further investigation. PPK experiments using higher GNSS sample rates did not always provide the best accuracies. GNSS performance and model accuracies were enhanced when using base stations located within 30 km of the survey site. Results without using GCPs observed a direct relationship between point cloud accuracy and GNSS performance, with R2 values reaching up to 0.97.

Entry point and screw path study of second sacral alar-iliac screw trajectory in spinal pelvic fixation under O-arm navigation

Objective: To explore the feasibility of using the entry point and screw path parameters of sacroiliac (S2AI) screws inserted under O-arm 3D computer navigation as a reference for freehand screw insertion in patients with degenerative spinal deformities. Methods: A retrospective analysis was conducted on the clinical data of 66 patients with degenerative spinal deformities who received S2AI screw fixation assisted by the O-arm 3D computer navigation system at Nanjing Drum Tower Hospital Affiliated with Nanjing University Medical School from January 2017 to April 2022. The patients included 6 males and 60 females, with a mean age of (64.3±5.9) years. Preoperatively, the entry point was set as the intersection of 1 mm from the outer and lower edges of the S1 foramen, adjusted intraoperatively by navigation, and verified postoperatively by full spinal CT scans and 3D reconstructions. Using the S1 screw entry point as the origin, the position of the bilateral S2AI screw entry points, tailward angulation (SA), outward angulation (TA), and vertical distance from the entry point to the skin (SD) were measured on postoperative CT 3D images. The accuracy of S2AI screw placement and any intraoperative and postoperative complications were recorded. Differences in entry point coordinates and screw path parameters between the left and right sides were compared. The intraclass correlation coefficient (ICC) was used to assess intra-observer and inter-observer agreement. Results: The coordinates and screw path parameters for the S2AI screws placed using the O-arm navigation demonstrated good intra-observer and inter-observer consistency (ICC>0.75). The left entry point was located (8.08±1.39) mm laterally and (24.47±2.20) mm caudally from the S1 entry point, while the right entry point was (8.09±1.41) mm laterally and (24.40±2.54) mm caudally, with no significant difference between the left and right sides (both P>0.05). The left TA was 46.33°±3.44°, SA was 39.14°±6.12°, and SD was (60.38±13.37) mm; the right TA was 46.37°±3.41°, SA was 39.59°±5.89°, and SD was (60.30±12.24) mm, with no significant differences between the left and right parameters (all P>0.05). There was no significant neurovascular complications intraoperatively or two weeks postoperatively, with a screw insertion accuracy of 97.7% (129/132). Conclusions: In the fixation of the pelvis in degenerative spinal deformities, the use of O-arm navigation-assisted S2AI screw insertion has a high accuracy rate. It is recommended that for freehand screw insertion, the entry point should be located 24 mm caudally and 8 mm laterally from the outer and lower edges of the superior articular process of S1, with an TA of approximately 46° and a SA of approximately 39°.

Conjugate points of dynamic pairs and control systems

We study the geometry of dynamic pairs $(X,\VV)$ on a manifold $M$, where $X$ is a vector field and $\VV$ is a distribution on $M$, both satisfying a regularity condition. Special cases are pairs defined by systems of second order ODEs, geodesic sprays in Riemannian, Finslerian and Lagranian geometries, semi-Hamiltonian systems and control-affine systems. Analogs of conjugate points from the calculus of variations are defined for the pair $(X,\VV)$. The main results give estimates for the position of conjugate points in terms of a curvature operator, analogously to the Cartan--Hadamard and Bonet--Myers theorems. Contrary to classical cases, no metric is given a priori, the distribution $\VV$ may be nonintegrable and the curvature operator is defined in terms of $(X,\VV)$.

The Aerodynamic Assessment and Shot Point Position Accuracy of UAV Seismic Source

An unmanned aerial vehicle (UAV) seismic source, suitable for complex terrains due to its eco-friendly, cost-effective, and efficient nature, operates by hovering and releasing impact objects to generate seismic waves. The aerodynamic behavior of these impact objects is vital for shotpoint position accuracy and drag magnitude. Excessive drag will result in the loss of seismic source energy, whereas the deviations of the shotpoint position will disrupt the geometric structure of the observation system, significantly impacting the quality of seismic data acquisition. Through theoretical calculations, the influencing factors of air resistance and cross-wind deflection are analyzed. Air resistance increases with the increase in mass, cross-sectional area, speed, and drag coefficient. Increasing the mass and throwing height can increase the upper limit of the output, but after the air resistance and gravity are balanced, the output kinetic energy of the drone’s seismic source will no longer increase when the throwing height is increased. Using computational fluid dynamics simulations and field tests, this study examines the aerodynamics of four prevalent impact object designs. The results show that shuttle and spherical structures have the least drag during descent and are less influenced by crosswinds, thus being optimal for this application. Field measurement data are also presented for reference.

The preparation of the budget planning of the Rembang regency local government

The function of the Regional People's Representative Council (DPRD) is manifested in 3 (three) things, namely: the legislation function, the budget function, and the supervisory function. In particular, the function of the district/city DPRD is regulated in Article 149 paragraph (1) of Law Number 23 of 2014 concerning Regional Government (hereinafter referred to as the Regional Government Law) which states that the district/city DPRD has the function of forming district/city Regional Regulations, budgets, and supervision. Problem 1. What is the effectiveness of the main points of thought in the preparation of budget planning in Rembang Regency? 2. What are the factors that affect the effectiveness of the main points of thought in the preparation of budget planning in Rembang Regency? This research is categorized as an empirical (non-doctrinal) legal research that mainly examines the position of the main points of thought of the DPRD in the Rembang Regency RKPD by comparing the situation in the field with the laws and regulations (statute approach). The results of the study show that the effectiveness of the DPRD's Points of Thought in the preparation of budget planning in Rembang Regency is highly dependent on the synergy between the DPRD, the local government, and the community. As an instrument of representation of community aspirations conveyed through members of the DPRD, the Principles of the DPRD plays an important role in ensuring that the needs and priorities of the development of the Rembang community are accommodated in the Regional Revenue and Expenditure Budget (APBD), so that the Principles of the DPRD can function optimally, a mechanism that is well coordinated with the results of the Development Plan Deliberation (Musrenbang) is needed, so that there is no overlap in programs. 2. Factors that affect the effectiveness of Points of Thought (Pokir) in the preparation of budget planning in Rembang Regency are very diverse and interrelated. First, the quality of community participation is a key factor. The high participation of the community in the process of collecting aspirations through the Development Plan Deliberation (Musrenbang) will produce Pokir that is more relevant and in accordance with real needs in the field.

Towards Real-Time Integrated Water Vapor Estimates with Triple-Frequency Galileo Observations and CNES Products

Integrated water vapor (IWV) is a crucial parameter for tropospheric sounding and weather prediction applications. IWV is essentially calculated using observations from global navigation satellite systems (GNSS). Presently, the Galileo satellite system is further developed, including more visible satellites that transmit multi-frequency signals. This study aims to evaluate the accuracy of real-time IWV estimated from a triple-frequency Galileo-only precise point positioning (PPP) processing model utilizing E1, E5a, E5b, and E5 observations, which is not addressed by the previous studies. For this purpose, Galileo datasets from 10 global reference stations spanning various 4-week periods in the winter, spring, summer, and fall seasons are acquired. To process the acquired datasets, dual- and triple-frequency ionosphere-free PPP solutions are used, including E1E5a PPP, E1E5aE5b PPP, and E1E5E5b PPP solutions. The publicly available real-time products from the Centre National d’Etudes Spatiales (CNES) are utilized. The real-time IWV values are computed and then validated with the European Centre for Medium-Range Weather Forecasting (ECMWF) reanalysis products (ERA5) counterparts. The findings demonstrate that the root mean square error (RMSE) of the estimated IWV is less than 3.15 kg/m2 with respect to the ECMWF ERA5 counterparts. Furthermore, the E1E5aE5b PPP and E1E5E5b PPP models enhance the IWV’s accuracy by about 11% and 16%, respectively, compared with the E1E5a PPP model.

Absolute antenna group delay variations: estimation, single station repeatability and their influence on coordinate and float ambiguity estimation in baseline positioning

Phase center offsets and variations (PCV) are common corrections applied to global navigation satellite system (GNSS) phase observations in the context of precise point positioning. Similar to PCV are group delay variations (GDV), which affect code observations. In this paper, absolute GDVs, which are independent of a reference antenna, are estimated as antenna- and frequency-specific for the frequencies 1575.42 MHz and 1176.45 MHz, where separation of receiving and transmitting antenna GDV is possible with non-rotating antennas. Nineteen receiving and transmitting antennas of GPS, Galileo, BeiDou, and QZSS satellites, using observations from static reference stations, are considered. The single station repeatability of the receiver antenna GDV i.e., the estimation of the GDV using only one station, is evaluated for all antenna-frequency combinations. The repeatability ranges from 1.1 to 7.6 cm at a zenith angle of 80°, showing significant differences between antennas and frequencies. The estimated GDV corrections are applied to multi-GNSS baseline positioning using a total of 116 baselines. Receiver antenna GDV corrections exhibit a significant non-zero mean effect on the code-based vertical coordinate estimate. Float ambiguities are estimated using observation periods of up to 20 min. The 0.95-quantile effect of the GDV corrections on narrow lane ambiguity estimates is 2.02 cycles at the first epoch and 0.54 cycles after 5 min using 30 s observation sampling. The effect on wide lane ambiguities is constant over the 20-min period considered, with the 0.95-quantile being around 0.1 cycles for the GNSS considered.

Vieta–Lucas polynomials-based collocation simulation to analyze the solvent fraction factor in active and passive control flow induced by torsional motion

The present research explores the Boger fluid flow past a stretching cylinder with torsional motion in the presence of the magnetic field. It is assumed that the cylinder rotates continuously around its axis and that the starting point's position along the axis correlates with the cylinder wall's expansion rate. Additionally, the consequence of active and passive control of nanoparticles, activation energy, thermophoresis, and Brownian motion effects are considered. Similarity variables transform the governing partial differential equations into non-dimensional ordinary differential equations (ODEs). Furthermore, the Vieta–Lucas polynomials-based collocation method (V-LPBCM) is employed to solve the resulting ODEs. The V-LPBCM outcomes of Nusselt and Sherwood numbers are compared with Runge–Kutta Fehlberg's fourth-fifth-order scheme for validation purposes. The impact of various dimensionless parameters on the different profiles is depicted in the graphical representation. The increase in values of the magnetic parameter, the ratio of relaxation time, and the Reynolds number decline the velocity profile. The velocity profile increases as the values of the solvent fraction parameter rise. The thermal profile increases as the heat source/sink, and thermophoretic parameters rise. The increase in values of activation energy parameter increases the thermal profile.

Influence of Ocean Current Features on the Performance of Machine Learning and Dynamic Tracking Methods in Predicting Marine Drifter Trajectories

Accurately and rapidly predicting marine drifter trajectories under conditions of information scarcity is critical for addressing maritime emergencies and conducting marine surveys with resource-limited unmanned vessels. Machine learning-based tracking methods, such as Long Short-Term Memory networks (LSTM), offer a promising approach for trajectory prediction in such scenarios. This study combines satellite observations and idealized simulations to compare the predictive performance of LSTM with a resource-dependent dynamic tracking method (DT). The results indicate that when driven solely by historical drifter paths, LSTM achieves better trajectory predictions when trained and tested on relative trajectory intervals rather than the absolute positions of individual trajectory points. In general, LSTM provides a more accurate geometric pattern of trajectories at the initial stages of forecasting, while DT offers superior accuracy in predicting specific trajectory positions. The velocity and curvature of ocean currents jointly influence the prediction quality of both methods. In regions characterized by active sub-mesoscale dynamics, such as the fast-flowing and meandering Kuroshio Current and Kuroshio Current Extension, DT predicts more reliable trajectory patterns but lacks precision in detailed position estimates compared to LSTM. However, in areas dominated by the fast but relatively straight North Equatorial Current, the performance of the two methods reverses. The two methods also demonstrate different tolerances for noise and sampling intervals. This study establishes a baseline for selecting machine learning methods for marine drifter prediction and highlights the limitations of AI-based predictions under data-scarce and resource-constrained conditions.

Affordable Real-Time PPP—Combining Low-Cost GNSS Receivers with Galileo HAS Corrections in Static, Pseudo-Kinematic, and UAV Experiments

The Galileo High Accuracy Service (HAS) is a free of charge Global Navigation Satellite System (GNSS) augmentation service provided by the European Union. It is designed to enable real-time Precise Point Positioning (PPP) with a target accuracy (at the 95% confidence level) of 20 cm and 40 cm in the horizontal and vertical components, respectively, to be achieved within 300 s. The performance of the service has been confirmed with geodetic-grade receivers. However, mass market applications require low-cost GNSS receivers connected to low-cost antennae. This paper focuses on the performance of the real-time static and kinematic positioning achieved with Galileo HAS and low-cost GNSS receivers. The study is limited to GPS+Galileo dual-frequency positioning, thus exploiting the full potential of Galileo HAS SL1. We demonstrate that the target accuracy of Galileo HAS SL1 is reached with both geodetic-grade and low-cost receivers in dual-frequency static and kinematic applications in open-sky conditions. Precision of a few centimeters is reached for static positioning, while kinematic positioning results in subdecimeter precision. Vertical accuracy is limited by missing phase center offset models for low-cost antennas. In general, the performance of low-cost hardware using Galileo HAS for real-time PPP is comparable to that of geodetic-grade hardware. Therefore, combining low-cost GNSS receivers with Galileo HAS is feasible and justified.

Chaos and attraction domain of fractional Φ6‐van der Pol with time delay velocity

This article investigates the chaotic analysis and attractive domain of a fractional‐order ‐van der Pol with time delay velocity under harmonic excitation. Firstly, eight different types of bifurcation states of the system under different parameters are calculated by using the undisturbed system. Secondly, the Melnikov method is used to explore the effect of time delay velocity on the threshold of chaos in the Smale horseshoe sense under the double‐well potential and three‐well potential of the system. Finally, through numerical analysis of the phase diagram, bifurcation diagram, and maximum Lyapunov exponent, the influence of time delay velocity on system chaos is studied. The results indicate that an increase in the delay velocity coefficient will lead to the system transitioning from a chaotic state to a periodic state, while an increase in the delay velocity term will lead to the system transitioning from a periodic state to a chaotic state. In the study of system bifurcation, it is found that the position of the equilibrium points of the system changes during periodic motion. Therefore, cell mapping is used to draw the attractive domain of the system is studying the influence of initial conditions on the equilibrium point of the system and the results show that there is a close relationship between the attraction domain and the process of chaos occurrence.

Research on the role of multi-sensor system information fusion in improving hardware control accuracy of intelligent system

Abstract Multi-sensor management and control technology generally constructs a reasonable objective function to solve the optimal control command set to control a limited number of sensors to obtain higher quality measurement information, thus obtaining better target tracking performance. In the process of multi-sensor information fusion, there is not only the problem of information redundancy but also obvious time delay. A sensor fusion algorithm combined with global optimization algorithm is innovatively proposed. According to the key frames saved in the previous steps, feature points in local maps, sensor information, and loop information, a global optimization algorithm based on graph optimization model is constructed to optimize the position and pose of intelligent hardware system and the position of spatial feature points. Moreover, this work studies and experiments on multi-sensor fusion simultaneous localization and mapping (SLAM) comprehensively and systematically, and the experimental results show that the algorithm proposed in this work is superior to common open-source SLAM algorithm in positioning accuracy and mapping effect under special circumstances. Therefore, the method proposed in this work can be applied to intelligent driving of vehicles, vision-assisted movement of robots and intelligent control of unmanned aerial vehicles, thus effectively improving the hardware control accuracy of intelligent systems.

Advancement of Finite Element Method Solver Used in Dam Safety Monitoring System by Interpolation of Pore Pressure and Temperature Values

In this paper, we focused on the advancement of Dam Monitoring Software that incorporates the Finite Element Method (FEM), as these large infrastructure constructions are crucial for ensuring a dependable water supply, irrigation, flood control, renewable electric energy generation, and safe operation, which is of utmost importance to any country. However, the material properties and geotechnical environments of dams can change (deteriorate) over time, while the standards and legal norms that govern them become more and more rigorous, so in order to accurately assess the state of a dam and detect any concerning behavior, the software must be updated as well. The custom-developed FEM solver, unlike many commercial alternatives, is adaptable and can be reconfigured to function within a Dam Monitoring System. In this paper, we present the procedure for interpolating numerical values at measurement points, when the position of the measurement point does not align with the node of the element, allowing for additional instrument locations to be added to the monitored system without the need for remeshing the numerical model. This procedure is used to compare the actual pore pressures and temperature values of the concrete dam structure with the prediction of the numerical model, and the agreement is much greater with the new interpolation algorithm in comparison to the nearest nodal values, with the average relative difference for pore pressure reduced from 8.89% to 8.10%, justifying this implementation.

Numerical analysis of vortex-induced vibration of deepwater drilling riser based on Van der Pol wake oscillator model

Coupled equations of the dynamic model and the Van der Pol wake oscillator model are solved by the central finite difference method for the deepwater drilling riser. The vortex-induced vibration (VIV) response and fatigue life of the riser are calculated and the model validation is performed by comparing with the published experimental and simulation results. The effects of the top tension, current flow velocity, flow velocity profile, internal flow velocity as well as the installation of buoyancy modules on the VIV are discussed. Dynamic response and fatigue life of the riser under In-Line (IL) and Cross-Flow (CF) VIV are preliminarily analyzed. The results show that the VIV of the riser has multiple modes and exhibits a mixed behavior of standing waves and traveling waves. With the increase of top tension and flow velocity profile coefficient, the order of the VIV dominant mode and the peak values of the root mean square (RMS) of VIV displacement decrease, and the fatigue life of the riser is extended. With the increase of current flow velocity, the order of the VIV dominant mode increases and the riser fatigue life decreases. The effect of internal flow velocity on the riser VIV is neglectable. The installation of buoyancy modules can improve the riser stress state and extend the fatigue life. Compared with the CF VIV model, the calculated minimum fatigue life of the riser is extended under the IL and CF coupled VIV model due to the decrease of bending moment and the changing position of fatigue weak point.

How the Aliphatic Glycol Chain Length Determines the Pseudoeutectic Composition in Biodegradable Isodimorphic poly(alkylene succinate-ran-caprolactone) Random Copolyesters.

We synthesize four series of novel biodegradable poly(alkylene succinate-ran-caprolactone) random copolyesters using a two-step ring-opening/transesterification and polycondensation process with ε-caprolactone (PCL) as a common comonomer. The second comonomers are succinic acid derivatives, with variations in the number of methylene groups (nCH2) in the glycol segment, nCH2 = 2, 4, 8, and 12. The obtained copolyesters were poly(ethylene succinate-ran-PCL) (ESxCLy), poly(butylene succinate-ran-PCL) (BSxCLy), poly(octamethylene succinate-ran-PCL) (OSxCLy), and poly(dodecylene succinate-ran-PCL) (DSxCLy). We discovered a new mixed isodimorphic/comonomer exclusion crystallization in ESxCLy copolymers. The BSxCLy, OSxCLy, and DSxCLy copolymers display isodimorphic behavior. Our findings revealed a significant variation in the pseudoeutectic point position, from mixed isodimorphism/comonomer exclusion crystallization to isodimorphism with pseudoeutectic point variation from 54% to up to 90%. Moreover, we established a link between the melting temperature depression slope variation and the comonomer inclusion/exclusion balance, providing valuable insights into the complex topic of isodimorphic random copolymers.

Rebar binding point location method based on improved YOLOv5 and thinning algorithm

In order to solve the problem that the location of rebar binding points is unstable and easy to be affected by environmental noise, this paper proposes a rebar binding point location method based on object depth classification and thinning algorithm. Firstly, YOLOv5 is used to locate the target frame of the rebar binding point, and the Biformer module suitable for small target detection is added to YOLOv5. Secondly, the continuous depth data in the target frame is classified by the depth-first algorithm, and the top level rebar is classified and stored as a binary image, and the Zhang-Suen thinning algorithm is used to obtain the reinforced skeleton. Finally, the skeleton intersection point recognition algorithm is used to obtain the position of the rebar binding point. From the training results, 95 % of the target frames for rebar binding points were recognized. Experiments show that the method can accurately obtain the location of the binding points of the rebar, and can accurately classify the noise, non-target binding points and target binding points, and has high stability for the location of binding points, which is better than YOLOv8-POSE.

Design and analysis of folding mechanism for automotive seats based on space maximization

To solve the problem of maximizing space utilization of car folding seats within limited spatial constraints, a design method for space-maximizing four-bar folding seat hinges is proposed. First, we will investigate the existing types of car folding seats, identify the mainstream configurations, and determine the optimal position of the four-bar linkage hinge point based on the constraints present within the car using theoretical analysis methods. Secondly, kinematic and dynamic models of the four-bar linkage mechanism are established, followed by the formulation of kinematic and dynamic equations. Finally, numerical methods are employed to solve these equations and obtain the kinematic and dynamic characteristics. The results indicate that the design method proposed in the article, which aims to maximize the space of the hinge four-bar folding seat, is efficient and feasible. This approach optimizes the utilization of interior space and enhances passenger comfort while accommodating the folding function of the seat within the existing vehicle environment. This method holds significant engineering implications for the design of folding seats in similar types of vehicles.

A Virtual Reality Tool for Accuracy Assessment of 3D Models in an Immersive Virtual Environment

Abstract. Accurate validation and assessment techniques are essential for ensuring the reliability of spatial reconstructions derived from photogrammetry, enabling well-informed decision-making across diverse domains. This study presents a Virtual Reality (VR) based accuracy assessment tool tailored for evaluating the accuracy and quality of 3D models generated by Unmanned Aerial Vehicles (UAVs). Leveraging the Unity game engine platform, our workflow entails three key steps: aligning real-world coordinates with an arbitrary Unity coordinate system, transforming the positions of Ground Control Points (GCPs) from field survey to the arbitrary system using a reference GCP, and marking observed points on the 3D models. Absolute and Relative Root Mean Square Errors (RMSE), Mean Errors (ME), and Standard Deviation of errors (SD) are computed within the virtual environment via the game object transform properties. The error distributions around each GCP are visually depicted using Unity game engine components for enhanced interaction and comprehension. The efficacy of the tool is validated through experimentation on four 3D models generated from varying camera angles during UAV data capture. The tool provides the opportunity to directly interact with the 3D models and visualize the errors, which is quite distinct from traditional methods. Using the developed tool, results were obtained to indicate that configurations employing camera angles of 60° + 75º exhibit notable performance in terms of relative and absolute accuracy.

Research on a Feature Point Detection Algorithm for Weld Images Based on Deep Learning

Laser vision seam tracking enhances robotic welding by enabling external information acquisition, thus improving the overall intelligence of the welding process. However, camera images captured during welding often suffer from distortion due to strong noises, including arcs, splashes, and smoke, which adversely affect the accuracy and robustness of feature point detection. To mitigate these issues, we propose a feature point extraction algorithm tailored for weld images, utilizing an improved Deeplabv3+ semantic segmentation network combined with EfficientDet. By replacing Deeplabv3+’s backbone with MobileNetV2, we enhance prediction efficiency. The DenseASPP structure and attention mechanism are implemented to focus on laser stripe edge extraction, resulting in cleaner laser stripe images and minimizing noise interference. Subsequently, EfficientDet extracts feature point positions from these cleaned images. Experimental results demonstrate that, across four typical weld types, the average feature point extraction error is maintained below 1 pixel, with over 99% of errors falling below 3 pixels, indicating both high detection accuracy and reliability.