Intelligent Control for Drones Using Artificial Neural Network

Just how many times have you heard of weather problems causing aircraft accidents? Countless times. Whether it is an unexpected abrupt change in the speed and direction of the wind, or it’s an unexpected rain, perhaps a violent storm or maybe some unwelcome and unpredictable clouds in the way of an aircraft, may be a wet runway. Aircraft pilots are equipped with the skills and Standard Operating Procedures (SOPs) to follow immediately in case an unexpected, serious environmental hazard is encountered. Aircraft are designed to endure short term, abrupt changes in environment and flight timings are often altered according to weather conditions.

Under abrupt weather conditions, or under fatigue damage; which is the damage caused by overuse over time, or under any other circumstances which might as well be unknown to an engineer, an aircraft might indicate aberrations; it might sound strange, respond to commands differently, have problems with its maneuverability etc. Just like a human body, an aircraft shows symptoms of disease. These symptoms have to be acknowledged, investigated, their source or cause has to be identified and fixed or reset. Not doing so could cause serious damage or accidents.

Due to the serious nature of the matter, research to improve an aircraft’s performance against environmental problems is a hot topic of electronic and aviation engineering; instrumentation engineering is a vital part of this research as the aberrations are noted and fixed using automated or conveniently guided instruments. The research titled, Intelligent Control For Unmanned Aerial Systems With System Uncertainties And Disturbances Using Artificial Neural Network proposed by Mohammad Jafari and Hao Xu of Department of Electrical and Biomedical Engineering in the University of Nevada aims to propose an innovative method to overcome the crushing challenge of stabilizing an Unmanned Aircraft System (UAV) under changing environment conditions.

The researchers propose an intelligent control method to overcome the problems. How does this intelligent method work? It is based on an artificially created, adaptive neural network that is called Radial Basis Function (RBF). The working of RBF is as follows:

1. The first step is to identify any changes in the environment variables of the unmanned aircraft; this, conveniently, is done using a neural network based identifier.
2. With the problem or aberration identified, a processor would easily reset the aircraft system to its optimized values using a neural network based controller.

That, as far as the working of the solution is considered, is it. But that’s not all an innovative solution could bring to the table could it?

System Coordinates.

Well, the learning capability of the proposed intelligent controller makes it a promising approach to take system uncertainties, noises and/or disturbances into account. The performance of the proposed intelligent controller is validated based on the computer based simulation results with system uncertainties and disturbances, such as wind gusts disturbance. While the simulations were done for a small quad copter, the algorithm is promising enough to be used for military grade drones with just a few alterations.

Citation: Jafari, Mohammad and Xu, Hao, Intelligent Control for Unmanned Aerial Systems with System Uncertainties and Disturbances Using Artificial Neural Network http://www.mdpi.com/2504-446X/2/3/30 10.3390/drones2030030