Papers by Keyword: B-Spline

Abstract: This paper proposes a method of using the B-spline mathematical model to plan high smoothness curve trajectories with heading condition through given waypoints for autonomous underwater vehicles (AUVs) in particular and ships with rudder systems in general. In addition, this paper examines some of the physical limitations of this vehicle, which lead to some binding conditions of the trajectory. Besides, the paper applies B-spline approximation method to reduce the curvature of the trajectory, when waypoints are too close and we do not need to go through exactly these waypoints in the 2D plane. Finally, this paper also proposes the genetic algorithm application with some modifications to solve the optimization of B-spline path length with constraint in turning radius of the vehicle.

Abstract: The paper presents a method developed and used by the CCHAPT researchers for the graphic plotting of the index tests results for hydraulic turbines, the comparison of the efficiency curves resulted from testing to those obtained by the model transposition [1] i.e. the determination and comparison of the existing combinatory cam with that obtained from tests.The method presented in the paper was born from the need for processing and presenting the results of index tests within the shortest delay and eliminating the errors that might occur in the results plotting.

Abstract: A path planning algorithm based on B-spline interpolation techniques was constructed for automatic welding system.The system used a B-spline curve to reconstruct the weld,it was achieved by reversing the control points of B-spline curve through the prescribed date points. The weld posture model was then obtained from the osculating plane and normal plane of B-spline curve. By taking a series coordinate transformation to the weld posture model, the torch posture model based on control terminal was provided.Experiments show that the new algorithm can readily be used for various three-dimensional welding tasks.

Abstract: B-spline curve parameters were estimated by the least squares. The influences of the scanning angle to point cloud density was discussed. In the case of scanning distance and horizontal being sure, the smaller the vertical scanning angle, the larger point cloud density, it can reflect the characteristics of the surface cross-section more accurately. The factors affecting the accuracy of the surface reconstruction were discussed. The surface fitting order has a greater degree of influence on the accuracy of surface reconstruction. When the surface order was 4 and the number of control points was 75, the surface-point cloud deviation was 0.188mm by the segmented fitting.

Abstract: In order to realize the mechanized transplanting of rice pot seedling and ensure our food security, The pitch curve of non-circular gear is fitted based on cubic, non-uniform and rational B-spline curve. The planetary gear train transplanting mechanism has been invented for ride type, and kinematics mathematical model has been built through the kinematics analysis of transplanting mechanism. The computer aided analytical and optimized software has been developed by using software platform of Matlab. Through tuning the data points by man-machine interaction, pitch curve of non-circular gear is optimized and structural parameters are obtained, which can meet the demand of track and attitude in the transplanting process for rice pot seedling. In condition of the parameters, the correctness of the established model is verified by the virtual experiment by software of Adams.

Abstract: This paper is to provide literature review of the Non Uniform Rational B-Splines (NURBS) formulation in the curve and surface constructions. NURBS curves and surfaces have a wide application in Computer Aided Geometry Design (CAGD), Computer Aided Design (CAD), image processing and etc. The formulation of NURBS showing that NURBS curves and surfaces requires three important parameters in controlling the curve and also modifying the shape of the curves and surfaces. Yet, curves and surfaces fitting are still the major problems in the geometrical modeling. With this, the researches that have been conducted in optimizing the parameters in order to construct the intended curves and surfaces are highlighted in this paper.

Abstract: Using NURBS modeling method can construct more realistic and vivid 3D models, because it has a remarkable ability to control the curves of the model surface than the traditional grid modeling method. Therefore, on analysis of the primary principles of NURBS, this paper adopted professional modeling software 3DS MAX to explore the methods and techniques of NURBS in the complex surface building process. And then, the techniques and methods were applied to construct practical engineering machinery, completing the 3D model of bulldozer case.

Abstract: The purpose of this study is to design a redundant robot meeting the specific task and tracking the specific trajectory. Although the Moore-Penrose pseudo-inverse kinematic is commonly using to solve the inverse kinematic problem, it cannot solve the singularity exists in some situations that the rank of Jacobian matrix is not full rank for the redundant robot. Thus, a fuzzy motion planning algorithm is proposed to solve the inverse kinematic with singularity. Finally, we can obtain the position of five axes robot manipulator using fuzzy motion planning mapping method, and the errors of the singular points are approximately to zero. The results prove the fuzzy inverse kinematic mapping method can robust singular point when the tracking path with singular points.

Abstract: To improve the smoothness of SFS reconstructed surface, a new approach based on genetic algorithm is proposed. First, by analyzing the principle of SFS, the reason for surface error and distortion zone are found. By adding data according to the symmetric and asymmetric principle and B spline, data in the distortion zone are added and the reconstruction of SFS is realized. The evaluation criteria for smoothness of curves and surfaces are given according to the mathematical sense and concurrent level of CAGD/CAD/CAM. The surface smoothing process is turn into two curve smoothing process in two directions. The evaluation criteria is taken as the constraint condition and then the smoothing of SFS reconstructed surface based on the genetic algorithm is realized. The experimental results show that this new approach is an effective way to eliminate the bad points of SFS surface, improve the surface quality so it is a practical method for surface smoothing.

Abstract: Online viscosity and density estimation of ternary solution with sucrose and sodium chloride means a lot in osmotic dehydration of food. The authors proposed a novel B-spline method using recursive least squares to estimate viscosity and density using concentrations of the two solutes and temperature. A 4-D B-spline is established to estimate viscosity with the inputs of sucrose concentration, sodium chloride concentration and temperature. And a 3-D B-spline is also established to estimate density in a fixed temperature. The maximum relative errors of viscosity and density estimation are 0.0062% and 1.33×10^-5% respectively which show that the method is of high accuracy and suitable for practical application.