372results about How to "Reduce error" patented technology

This invention relates to a method and apparatus for characterizing composite materials, and in particular, to utilizing an artificial neural network for predicting an impact resistance of a composite material. A method for predicting an impact resistance of a composite material in accordance with the present invention includes the steps of designing an artificial neural network including a plurality of neurons, training the artificial neural network to predict the impact resistance by adjusting an output of the plurality of neurons according to sample data and known results of the sample data, inputting data of the composite material into the artificial neural network, and utilizing the artificial neural network to predict the impact resistance of the composite material.

A multi-modulation frequency laser range finder and a method for the same make use of a crystal oscillator to send out a modulation signal and a sampling signal. After the modulation signal is coupled with a light signal emitted by a laser diode, a laser light beam is generated and emitted to a target. The laser light signal is reflected by the target and received by a receiver to be converted to a reception signal, which is processed according to the frequency of the sampling signal. The processed result is finally sent to a data processor to calculate out the distance to the target. As compared to the prior art, less number of components are used, the precision is enhanced, and the circuit is much simplified, hence reducing the errors and the cost.

A novel method for adaptive Nonlinear Model Predictive Control (NMPC) of multiple input, multiple output (MIMO) systems, called Sampling Based Model Predictive Control (SBMPC) that has the ability to enforce hard constraints on the system inputs and states. However, unlike other NMPC methods, it does not rely on linearizing the system or gradient based optimization. Instead, it discretizes the input space to the model via pseudo-random sampling and feeds the sampled inputs through the nonlinear plant, hence producing a graph for which an optimal path can be found using an efficient graph search method.

The embodiment of the invention relates to the technical field of robot control, and discloses a method and a module for distinguishing a load of a six-axis robot. The method for distinguishing the load of the six-axis robot comprises the following steps: according to a preset parameter limiting value of each joint, generating an excitation track of the six-axis robot; when the six-axis robot carrying the load is controlled to run along the excitation track, collecting a parameter running value of each joint; according to the parameter running value of each joint and a preset robot dynamics model, calculating a torque predicting value of each joint; calculating a difference between the torque running value and the torque predicting value, the difference serving as torque produced by the load, and according to preset relationship of the torque produced by the load and a load inertia parameter, calculating the load inertia parameter. The embodiment of the invention also provides the module for distinguishing the load of the six-axis robot. The module for distinguishing the load of the six-axis robot disclosed by the embodiment has the advantage that compared with the prior art, the parameter running value of the robot is collected on line, and the load inertia parameter of the robot is distinguished; because a six-dimensional force sensor is not used, the cost is reduced.