Adaptation Mechanism of Asymmetrical Potential Field Improving Precision of Position Tracking in the Case of Nonholonomic UAVs

Cezary Kownacki , Leszek Ambroziak

Abstract

SummaryPosition-tracking problems in the structures of rigid formations of nonholonomic mobile robots, such as fixed-wing unmanned aerial vehicle (UAVs), must reconcile tracking precision and flight stability, which usually exclude each other due to nonholonomic motion constraints. Therefore, a position-tracking control that is based on distance and position displacement, defined as inputs of control loops, requires the application of dead zones around target positions, which are the points of instability. For this reason, the control becomes sensitive to any external disturbance causing oscillations of control signals and so it becomes difficult to maintain a zero value of position displacement over a long time horizon. Thus, we propose an approach based on the adaptive mechanism of an asymmetrical local artificial potential field, which is defined by a local frame of reference whose origin is located in the tracked position of a UAV in the formation frame. It couples controls of both airspeed and heading angle into a nonlinear potential function of relative position and orientation with respect to the tracked position and adapts it according to heading rate of the leader. The function splits the area around the tracked position longitudinally into two zones of acceleration and deceleration; therefore, velocity vectors are longer (higher airspeed) only when a UAV is behind the tracked position and shorter (lower airspeed) when it is ahead. The area is laterally symmetrical, and orientations of velocity vectors align asymptotically to the longitudinal direction accordingly with the decrease in the lateral error. Finally, velocity vectors are rotated proportionally to the heading rate of the leader, which improves the tracking precision during turns. If we assumed that a UAV’s tracked position is in motion, it could easily be proven that the position control based on the adaptive asymmetrical potential function becomes asymptotically stable in the tracked position. Numerical simulation verifies this thesis and presents more precise and stable position tracking due to the adaptation mechanism.
Author Cezary Kownacki (FME / DACR)
Cezary Kownacki,,
- Department of Automatic Control and Robotics
, Leszek Ambroziak (FME / DACR)
Leszek Ambroziak,,
- Department of Automatic Control and Robotics
Journal seriesRobotica, ISSN 0263-5747, (N/A 70 pkt)
Issue year2019
Vol37
No10
Pages1823-1834
Publication size in sheets0.55
ASJC Classification1706 Computer Science Applications; 2600 General Mathematics; 1712 Software; 2207 Control and Systems Engineering
DOIDOI:10.1017/S0263574719000286
Internal identifierROC 19-20
Languageen angielski
Score (nominal)70
Score sourcejournalList
ScoreMinisterial score = 70.0, 11-03-2020, ArticleFromJournal
Publication indicators Scopus SNIP (Source Normalised Impact per Paper): 2017 = 0.838; WoS Impact Factor: 2018 = 1.184 (2) - 2018=1.494 (5)
Citation count*
Cite
Share Share

Get link to the record


* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back
Confirmation
Are you sure?