Variable Speed Control Method of Freeway Mainline based on IGA-DRFNN

The flight body attitude parameters recorder that is designed by combining the geomagnetic/gyro inertia measurement, can complete the test data acquisition, real time calculating of attitude angle and information storage. According to the output characteristic of the magnetic sensor and gyro, the mathematical model of attitude calculation algorithm is established, the kalman filter with the information fusion ability based on the characteristics of the sensor is designed. Through simulation, the correctness of the algorithm is verified. And the experimental results show that kalman filter can effectively restrain the gyro error accumulation while calculating, the system can accurately real-time calculate the flight parameters. Speed control of freeway mainline is very necessary to guarantee stable traffic flow and reduce traffic accident. In this paper, a variable speed control method of freeway mainline based on dynamic recurrent fuzzy neural network (DRFNN) with immune genetic algorithm (IGA) optimization is proposed. This method fully considers the traffic flow condition, vehicle type composition, road condition as well as weather condition. The IGA optimization make the DRFNN has a certain learning and self-adaptation capability, which improves the anti-interference and robustness and leads to quick convergence speed and low training error. Analysis and numerical results from application to realistic scenario demonstrate that this method has higher sensitivity under complex external influence conditions compared with relative methods and is more suitable for conducting efficient real-time speed control of freeway mainline.