Use of neural network approximation in multiple-unmanned aerial vehicle trajectory optimization

Brian R. Geiger, Eric M. Schmidt, Joseph F. Horn

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Scopus citations

Abstract

A direct trajectory optimization method that uses neural network approximations is presented. Neural networks are trained to model objective functions, vehicle dynamics, and non-linear constraints. The neural network method reduces computational requirements by removing the need for collocation and providing fast analytical computation of gradients. The method was shown to significantly reduce computational costs while resulting in trajectories comparable to direct collocation and pseudospectral methods. The method is applied to a multi-aircraft surveillance mission to demonstrate the scalability of the method when adding additional aircraft or objective functions to the optimization problem. Cases are presented in which neural networks can be reused for different purposes without additional training.

Original languageEnglish (US)
Title of host publicationAIAA Guidance, Navigation, and Control Conference and Exhibit
StatePublished - Dec 1 2009
EventAIAA Guidance, Navigation, and Control Conference and Exhibit - Chicago, IL, United States
Duration: Aug 10 2009Aug 13 2009

Publication series

NameAIAA Guidance, Navigation, and Control Conference and Exhibit

Other

OtherAIAA Guidance, Navigation, and Control Conference and Exhibit
Country/TerritoryUnited States
CityChicago, IL
Period8/10/098/13/09

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Use of neural network approximation in multiple-unmanned aerial vehicle trajectory optimization'. Together they form a unique fingerprint.

Cite this