Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras

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

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

3 Citations (Scopus)

Abstract

A direct trajectory optimization method that uses neural network approximation methods is presented. Neural networks are trained to approximate objective functions and vehicle dynamics. The neural network method reduces computational requirements by removing the need for collocation and providing fast computation of gradients. The method has been shown to significantly reduce computational costs while resulting in trajectories comparable to direct collocation and pseudospectral methods. Since the neural networks readily provide accurate computation of gradients, it removes the need for formulating analytical gradients, so the method is more easily extended to different types of applications with different objective functions and constraints. In previous work, the method was applied towards trajectory optimization of single or multiple UAVs with fixed, downward facing cameras in order to maximize surveillance of a ground target. In this paper the method is extended to optimize trajectories for a UAV equipped with a gimbaled camera, and more complex constraints on the vehicle trajectory are investigated.

Original languageEnglish (US)
Title of host publicationAIAA Guidance, Navigation, and Control Conference
DOIs
StatePublished - Dec 1 2010
EventAIAA Guidance, Navigation, and Control Conference - Toronto, ON, Canada
Duration: Aug 2 2010Aug 5 2010

Publication series

NameAIAA Guidance, Navigation, and Control Conference

Other

OtherAIAA Guidance, Navigation, and Control Conference
CountryCanada
CityToronto, ON
Period8/2/108/5/10

Fingerprint

Unmanned aerial vehicles (UAV)
Cameras
Trajectories
Neural networks
Computational methods
Costs

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Control and Systems Engineering

Cite this

Schmidt, E. M., Horn, J. F., & Geiger, B. R. (2010). Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras. In AIAA Guidance, Navigation, and Control Conference [AIAA 2010-7737] (AIAA Guidance, Navigation, and Control Conference). https://doi.org/10.2514/6.2010-7737
Schmidt, Eric M. ; Horn, Joseph Francis ; Geiger, Brian R. / Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras. AIAA Guidance, Navigation, and Control Conference. 2010. (AIAA Guidance, Navigation, and Control Conference).
@inproceedings{6df5c953c14c43238fa20c56d92cae25,
title = "Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras",
abstract = "A direct trajectory optimization method that uses neural network approximation methods is presented. Neural networks are trained to approximate objective functions and vehicle dynamics. The neural network method reduces computational requirements by removing the need for collocation and providing fast computation of gradients. The method has been shown to significantly reduce computational costs while resulting in trajectories comparable to direct collocation and pseudospectral methods. Since the neural networks readily provide accurate computation of gradients, it removes the need for formulating analytical gradients, so the method is more easily extended to different types of applications with different objective functions and constraints. In previous work, the method was applied towards trajectory optimization of single or multiple UAVs with fixed, downward facing cameras in order to maximize surveillance of a ground target. In this paper the method is extended to optimize trajectories for a UAV equipped with a gimbaled camera, and more complex constraints on the vehicle trajectory are investigated.",
author = "Schmidt, {Eric M.} and Horn, {Joseph Francis} and Geiger, {Brian R.}",
year = "2010",
month = "12",
day = "1",
doi = "10.2514/6.2010-7737",
language = "English (US)",
isbn = "9781600869624",
series = "AIAA Guidance, Navigation, and Control Conference",
booktitle = "AIAA Guidance, Navigation, and Control Conference",

}

Schmidt, EM, Horn, JF & Geiger, BR 2010, Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras. in AIAA Guidance, Navigation, and Control Conference., AIAA 2010-7737, AIAA Guidance, Navigation, and Control Conference, AIAA Guidance, Navigation, and Control Conference, Toronto, ON, Canada, 8/2/10. https://doi.org/10.2514/6.2010-7737

Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras. / Schmidt, Eric M.; Horn, Joseph Francis; Geiger, Brian R.

AIAA Guidance, Navigation, and Control Conference. 2010. AIAA 2010-7737 (AIAA Guidance, Navigation, and Control Conference).

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

TY - GEN

T1 - Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras

AU - Schmidt, Eric M.

AU - Horn, Joseph Francis

AU - Geiger, Brian R.

PY - 2010/12/1

Y1 - 2010/12/1

N2 - A direct trajectory optimization method that uses neural network approximation methods is presented. Neural networks are trained to approximate objective functions and vehicle dynamics. The neural network method reduces computational requirements by removing the need for collocation and providing fast computation of gradients. The method has been shown to significantly reduce computational costs while resulting in trajectories comparable to direct collocation and pseudospectral methods. Since the neural networks readily provide accurate computation of gradients, it removes the need for formulating analytical gradients, so the method is more easily extended to different types of applications with different objective functions and constraints. In previous work, the method was applied towards trajectory optimization of single or multiple UAVs with fixed, downward facing cameras in order to maximize surveillance of a ground target. In this paper the method is extended to optimize trajectories for a UAV equipped with a gimbaled camera, and more complex constraints on the vehicle trajectory are investigated.

AB - A direct trajectory optimization method that uses neural network approximation methods is presented. Neural networks are trained to approximate objective functions and vehicle dynamics. The neural network method reduces computational requirements by removing the need for collocation and providing fast computation of gradients. The method has been shown to significantly reduce computational costs while resulting in trajectories comparable to direct collocation and pseudospectral methods. Since the neural networks readily provide accurate computation of gradients, it removes the need for formulating analytical gradients, so the method is more easily extended to different types of applications with different objective functions and constraints. In previous work, the method was applied towards trajectory optimization of single or multiple UAVs with fixed, downward facing cameras in order to maximize surveillance of a ground target. In this paper the method is extended to optimize trajectories for a UAV equipped with a gimbaled camera, and more complex constraints on the vehicle trajectory are investigated.

UR - http://www.scopus.com/inward/record.url?scp=84860377084&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860377084&partnerID=8YFLogxK

U2 - 10.2514/6.2010-7737

DO - 10.2514/6.2010-7737

M3 - Conference contribution

AN - SCOPUS:84860377084

SN - 9781600869624

T3 - AIAA Guidance, Navigation, and Control Conference

BT - AIAA Guidance, Navigation, and Control Conference

ER -

Schmidt EM, Horn JF, Geiger BR. Use of neural network approximation for trajectory optimization of unmanned aerial vehicles with gimbaled cameras. In AIAA Guidance, Navigation, and Control Conference. 2010. AIAA 2010-7737. (AIAA Guidance, Navigation, and Control Conference). https://doi.org/10.2514/6.2010-7737