|
Abstract: |
Using knowledge-driven trajectory prediction is difficult to describe complex reasoning processes, and the use of expert prior knowledge can add too many subjective factors to the judgment process, which makes it difficult to adapt to increasingly complex environments. Therefore, existing methods have limitations. The development of neural networks can effectively solve this problem. Neural networks have complex structures that can adapt to various data and learn relevant knowledge from the data. The method of neural networks can not only learn the features between data, but also the personal factors hidden when different commanders make decisions. The universal approximation theorem points out that if there are enough neurons, the multi-layer perceptron can theoretically approximate the unknown function. Supported by this theory, this report discusses several trajectory prediction algorithms for aircrafts. |
