This paper presents an explicit formula based on reparameterization technique for progressively computing a simple root of a smooth function,which may have wide applications in robotics,geomagnetic navigation,geometri...This paper presents an explicit formula based on reparameterization technique for progressively computing a simple root of a smooth function,which may have wide applications in robotics,geomagnetic navigation,geometric processing and computer graphics.Comparing with Newton-like method,it can achieve convergence rate 2 by adding one more functional evaluation,improve the computational stability and ensure the convergence,and also obtain higher convergence rate and higher efficiency index.Compared with clipping methods for polynomials,it doesn't need to bound the polynomials,directly bound the roots and can also work well for non-polynomial functions with much higher computational efficiency.Comparing with previous progressive methods,it achieves a much higher computational efficiency and is extended to solve bivariate equation system.Numerical examples show its much better performance on approximation error,computational efficiency and computational stability.展开更多
基金Supported by the National Natural Science Foundation of China (61972120)。
文摘This paper presents an explicit formula based on reparameterization technique for progressively computing a simple root of a smooth function,which may have wide applications in robotics,geomagnetic navigation,geometric processing and computer graphics.Comparing with Newton-like method,it can achieve convergence rate 2 by adding one more functional evaluation,improve the computational stability and ensure the convergence,and also obtain higher convergence rate and higher efficiency index.Compared with clipping methods for polynomials,it doesn't need to bound the polynomials,directly bound the roots and can also work well for non-polynomial functions with much higher computational efficiency.Comparing with previous progressive methods,it achieves a much higher computational efficiency and is extended to solve bivariate equation system.Numerical examples show its much better performance on approximation error,computational efficiency and computational stability.
