AI Inspired ATC, Based on ANN and Using NLP

2023-01-0985

03/07/2023

Event
2023 AeroTech
Authors Abstract
Content
An Air Traffic Controller(ATC) is a person responsible for the proper Take-Off and Landing of an Aircraft from the runway, and for relaying continuous vital information back and forth from Pilots. The proposed ATC will automate this entire process to reduce human-generated errors and save costs. The entire system will be made using Artificial Intelligence and will use Natural Language Processing and Artificial Neural Networks to create a human-like, but a better-prepared system. The model needed to create the ATC, can be trained on already available crucial flight data. The data must include flight take-off and landing time, along with altered time based on weather, climate and other physical factors. The back-end system of the ATC, can be then made to work on this trained model, and produce correct and calculated flight path and timings for the take-off and Landing. The system will do an automatic Pre-flight checkup, based on weather and other clear-sky conditions, such as birds and overhead flights. If there are no problematic conditions, a flight can be allowed to take-off. Similarly, a flight can be allowed to land, based on a clear runaway and good weather conditions. Also, the system will use Artificial Neural Networks, to pan out an optimized flight path for the aircraft to follow, so as to reach a particular destination by avoiding extra air traffic, and saving fuel.The AI-inspired ATC will also be responsible, to tackle problems faced by pilots based on their requests, voice & mood conditions, which will be processed using a customized NLP Component. Implementing the proposed AI-inspired Air Traffic Controller can significantly reduce errors, save costs, and reduce the overhead of extra time in panic situations.
Meta TagsDetails
DOI
https://doi.org/10.4271/2023-01-0985
Pages
13
Citation
Aman, E., Jana, S., Athikary, K., and Suryanarayana, R., "AI Inspired ATC, Based on ANN and Using NLP," SAE Technical Paper 2023-01-0985, 2023, https://doi.org/10.4271/2023-01-0985.
Additional Details
Publisher
Published
Mar 7, 2023
Product Code
2023-01-0985
Content Type
Technical Paper
Language
English