Vision-Based Reactive Planning for Aggressive Target Tracking While Avoiding Collisions and Occlusions
Bryan Penin, Paolo Robuffo Giordano, François Chaumette
Abstract
In this letter, we investigate the online generation of optimal trajectories for target tracking with a quadrotor while satisfying a set of image-based and actuation constraints. We consider a quadrotor equipped with a camera (either down or front-looking) with limited field of view. The aim is to follow in a smooth but reactive way a moving target while avoiding obstacles in the environment and occlusions in the image space. We propose vision-based approaches based on multiobjective optimization, especially with the occlusion constraint formulation. We design an online replanning strategy inspired from model predictive control that successively solves a nonlinear optimization problem. The problem is formulated as a nonlinear program (NLP) using differential flatness and finite parametrization with B-Splines. This allows a resolution by sequential quadratic programming (SQP) at a rate of 30 Hz. The robustness and reactivity of the replanning algorithm are demonstrated through realistic simulation results. Experiments validating the performance with a real quadrotor are also presented.
BibTeX
@inproceedings{ral2018_visionbasedreact,
title = {Vision-Based Reactive Planning for Aggressive Target Tracking While Avoiding Collisions and Occlusions},
author = {Bryan Penin and Paolo Robuffo Giordano and François Chaumette},
booktitle = {RA-L 2018},
year = {2018}
}