Multilayer feedforward networks are universal approximators
Neural Networks
Hierarchical mixtures of experts and the EM algorithm
Neural Computation
Computational Methods for Inverse Problems
Computational Methods for Inverse Problems
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Neural Networks - 2003 Special issue: Neural network analysis of complex scientific data: Astronomy and geosciences
Artificial Neural Networks and Inverse Problems of Optical Diagnostics
ISDA '06 Proceedings of the Sixth International Conference on Intelligent Systems Design and Applications - Volume 01
Neural network method to solve inverse problems for canopy radiative transfer models
Cybernetics and Systems Analysis
Inverse kinematics in robotics using neural networks
Information Sciences: an International Journal
Hierarchical Bayesian inference for Ill-posed problems via variational method
Journal of Computational Physics
| Hi-index | 0.07 |
Inverse problems appear in many areas ranging from microwave circuits to environmental studies and to robotics, just to mention a few. In this paper we propose a new approach to solving inverse problems based on decomposition of output space into cells, with the corresponding regions in the input space. Solutions are identified using a clustering method and the relationship between data in an output cell and the corresponding input region is modeled by a simple polynomial. It is shown that the proposed method achieves very high accuracy even with relatively high number of inputs and outputs. It is also extremely fast and is suitable for real-time control, where needed. The method is applied to a highly complex inverse problem in robot kinematics and its performance is demonstrated.