Compensation of Nonlinearities Using Neural Networks Implemented on Inexpensive Microcontrollers

This paper describes a method of linearizing the nonlinear characteristics of many sensors and devices using an embedded neural network. The neuron-by-neuron process was developed in assembly language to allow the fastest and shortest code on the embedded system. The embedded neural network also requires an accurate approximation for hyperbolic tangent to be used as the neuron activation function. The proposed method allows for complex neural networks with very powerful architectures to be embedded on an inexpensive 8-b microcontroller. This process was then demonstrated on several examples, including a robotic arm kinematics problem.

[1]  Shuzhi Sam Ge,et al.  Adaptive neural control of uncertain MIMO nonlinear systems , 2004, IEEE Transactions on Neural Networks.

[2]  G. L. Dempsey,et al.  Control sensor linearization using a microcontroller-based neural network , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[3]  Bimal K. Bose,et al.  Neural Network Applications in Power Electronics and Motor Drives—An Introduction and Perspective , 2007, IEEE Transactions on Industrial Electronics.

[4]  Bogdan M. Wilamowski,et al.  Challenges in applications of computational intelligence in industrial electronics , 2010, 2010 IEEE International Symposium on Industrial Electronics.

[5]  Gérard-André Capolino,et al.  Fuzzy logic applied to speed control of a stepping motor drive , 2000, IEEE Trans. Ind. Electron..

[6]  Ching Chuen Chan,et al.  Adaptive neuro-fuzzy modeling of battery residual capacity for electric vehicles , 2002, IEEE Trans. Ind. Electron..

[7]  Marios M. Polycarpou,et al.  High-order neural network structures for identification of dynamical systems , 1995, IEEE Trans. Neural Networks.

[8]  Nicolás J. Medrano-Marqués,et al.  Sensor linearization with neural networks , 2001 .

[9]  Bogdan M. Wilamowski,et al.  Solving parity-N problems with feedforward neural networks , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..

[10]  B.M. Wilamowski,et al.  A Neural Network Implementation on an Inexpensive Eight Bit Microcontroller , 2008, 2008 International Conference on Intelligent Engineering Systems.

[11]  Nicholas Cotton,et al.  A Neural Network Implementation on Embedded Systems , 2010 .

[12]  B.M. Wilamowski,et al.  Neural network architectures and learning algorithms , 2009, IEEE Industrial Electronics Magazine.

[13]  Muhammad Usman Asad,et al.  A low cost microcontroller implementation of neural network based hurdle avoidance controller for a car-like robot , 2010, 2010 The 2nd International Conference on Computer and Automation Engineering (ICCAE).

[14]  Hao Yu,et al.  Improved Computation for Levenberg–Marquardt Training , 2010, IEEE Transactions on Neural Networks.

[15]  Bogdan M. Wilamowski,et al.  Compensation of Sensors Nonlinearity with Neural Networks , 2010, 2010 24th IEEE International Conference on Advanced Information Networking and Applications.

[16]  Hao Yu,et al.  Neural Network Learning Without Backpropagation , 2010, IEEE Transactions on Neural Networks.

[17]  Gérard-André Capolino,et al.  Determination of Scaling Factors for Fuzzy Logic Control Using the Sliding-Mode Approach: Application to Control of a DC Machine Drive , 2007, IEEE Transactions on Industrial Electronics.

[18]  R. Krishnan,et al.  Sensorless control of single switch based switched reluctance motor drive using neural network , 2004, 30th Annual Conference of IEEE Industrial Electronics Society, 2004. IECON 2004.

[19]  Okyay Kaynak,et al.  Computing Gradient Vector and Jacobian Matrix in Arbitrarily Connected Neural Networks , 2008, IEEE Transactions on Industrial Electronics.

[20]  Marco Liserre,et al.  Implementation issues of a fuzzy-logic-based three-phase active rectifier employing only Voltage sensors , 2005, IEEE Transactions on Industrial Electronics.

[21]  Frank L. Lewis,et al.  Multilayer neural-net robot controller with guaranteed tracking performance , 1996, IEEE Trans. Neural Networks.