Abstract:
A probabilistic neural network, comprising a hidden layer of neurons, each computing respective membership matrix elements for an input vector of the neural network according to a respective radial basis function defined by a respective spread factor and according to the distance of the input vector from a respective constant vector, wherein said hidden layer comprises at least two neurons having different spread factors (S). A method of training the novel probabilistic neural network is also disclosed.
Abstract:
A control system (10) for an electromechanical-braking system (1) provided with actuator elements (2, 5) configured to actuate braking elements (6, 8) for exerting a braking action has a control stage (16) for controlling the braking action on the basis of a braking reference signal (w). The control stage comprises a model-based predictive control block (16), in particular of a generalized predictive self-adaptive control (GPC) type, operating on the basis of a control quantity (F b ) representing the braking action. The control system further has: a model-identification stage (14), which determines parameters (a, b, d) identifying a transfer function (G(z)) of the electromechanical-braking system (1); and a regulation stage (15), which determines an optimal value of endogenous parameters of the control system on the basis of the value of the identifying parameters.