A track prediction algorithm is proposed based on the Genetic Algorithm (GA) and Back Propagation neural network (BPNN), in which the GA is used to optimize the initial weights and thresholds of hidden layers of the network. Firstly, a 5-5-4 topology of the BPNN for track prediction is established according to the characteristics of track prediction. Its inputs are historical tracks, and the output just is the predicted track. Secondly, the weights and thresholds of the BPNN are regarded as genes of individual in GA to construct chromosomes, individuals, and populations. Then, the optimal solution of the GA is used as the weights and thresholds of the BPNN for network training to get a better output. In the experiment, the track data are derived from Automatic Dependent Surveillance-Broadcast, and experimental results show that the proposed algorithm can improve both the accuracy and time efficiency of the BPNN. In addition, we use the predicted track of the proposed method to detect flight conflict by constructing conflict scenarios. The detecting process and results are visualized in the Geographic Information System, and the correct results of detecting again show that the proposed method is reliable.
Track prediction algorithm based on GA-BPNN
2021-10-20
1490260 byte
Conference paper
Electronic Resource
English
Steady-state characteristics prediction of marine towing cable with BPNN
Taylor & Francis Verlag | 2024
|Short-Term Load Forecasting Model Based on Multi-label and BPNN
Springer Verlag | 2017
|