Quadrotor Thrust Vectoring Control with Time Optimal Trajectory Planning in Constant Wind Fields

Research output: Contribution to journalArticleScientificpeer-review

7 Citations (Scopus)
17 Downloads (Pure)

Abstract

This paper proposes a trajectory planning and control strategy to optimally visit a given set of waypoints in the presence of wind. First, aerodynamic properties of quadrotors which affect trajectory planning and tracking performance are investigated. Blade flapping, induced and parasitic drag are derived and an extended method to identify all coefficients from flight test data is developed. Then, a three-step approach is suggested to optimize the trajectory. These steps reduce the size of the optimization problem and thereby increase computational efficiency while still guaranteeing near optimal results. The trajectories are optimized for minimal aerodynamic drag and minimal jerk. The derived smooth trajectory generation is compared with traditional trajectory planning consisting of discrete point to point tracking followed by low-pass filtering. The new trajectories yield a clear reduction in maximal needed thrust and in Euler angle aggressiveness. A thrust vectoring controller is designed, which exploits the a priori trajectory information and identified aerodynamic properties. Its performance is compared to a standard PID controller and results show a reduction in tracking delay and an increase in thrust and attitude angle margins, which overall enable faster flight.


Read More: http://www.worldscientific.com/doi/abs/10.1142/S2301385018500024?src=recsys
Original languageEnglish
Pages (from-to)1-23
Number of pages23
JournalUnmanned Systems
Volume6
Issue number1
DOIs
Publication statusPublished - 14 Nov 2017

Keywords

  • Quadrotor
  • control
  • thrust vectoring
  • wind
  • optimal trajectory
  • drag effects
  • minimum jerk
  • waypoint sequencing

Fingerprint

Dive into the research topics of 'Quadrotor Thrust Vectoring Control with Time Optimal Trajectory Planning in Constant Wind Fields'. Together they form a unique fingerprint.

Cite this