Robust Predictive Motion Planning by Learning Obstacle Uncertainty
(2025) In IEEE Transactions on Control Systems Technology- Abstract
Safe motion planning for robotic systems in dynamic environments is nontrivial in the presence of uncertain obstacles, where estimation of obstacle uncertainties is crucial in predicting future motions of dynamic obstacles. The worst case characterization gives a conservative uncertainty prediction and may result in infeasible motion planning for the ego robotic system. In this article, an efficient, robust, and safe motion-planning algorithm is developed by learning the obstacle uncertainties online. More specifically, the unknown yet intended control set of obstacles is efficiently computed by solving a linear programming (LP) problem. The learned control set is used to compute forward reachable sets (FRSs) of obstacles that are less... (More)
Safe motion planning for robotic systems in dynamic environments is nontrivial in the presence of uncertain obstacles, where estimation of obstacle uncertainties is crucial in predicting future motions of dynamic obstacles. The worst case characterization gives a conservative uncertainty prediction and may result in infeasible motion planning for the ego robotic system. In this article, an efficient, robust, and safe motion-planning algorithm is developed by learning the obstacle uncertainties online. More specifically, the unknown yet intended control set of obstacles is efficiently computed by solving a linear programming (LP) problem. The learned control set is used to compute forward reachable sets (FRSs) of obstacles that are less conservative than the worst case prediction. Based on the forward prediction, a robust model predictive controller is designed to compute a safe reference trajectory for the ego robotic system that remains outside the reachable sets of obstacles over the prediction horizon. The method is applied to a car-like mobile robot in both simulations and hardware experiments to demonstrate its effectiveness.
(Less)
- author
- Zhou, Jian ; Gao, Yulong ; Johansson, Ola ; Olofsson, Björn LU and Frisk, Erik
- organization
- publishing date
- 2025
- type
- Contribution to journal
- publication status
- epub
- subject
- keywords
- Predictive control, robust motion planning, safe autonomy, uncertainty quantification
- in
- IEEE Transactions on Control Systems Technology
- pages
- 15 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- external identifiers
-
- scopus:85217523432
- ISSN
- 1063-6536
- DOI
- 10.1109/TCST.2025.3533378
- project
- ELLIIT B14: Autonomous Force-Aware Swift Motion Control
- RobotLab LTH
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 1993-2012 IEEE.
- id
- 487e6d13-28ea-4b5d-9c4a-c14b97cf42ce
- alternative location
- https://arxiv.org/abs/2403.06222
- date added to LUP
- 2025-04-08 17:11:02
- date last changed
- 2025-04-09 10:10:36
@article{487e6d13-28ea-4b5d-9c4a-c14b97cf42ce,
abstract = {{<p>Safe motion planning for robotic systems in dynamic environments is nontrivial in the presence of uncertain obstacles, where estimation of obstacle uncertainties is crucial in predicting future motions of dynamic obstacles. The worst case characterization gives a conservative uncertainty prediction and may result in infeasible motion planning for the ego robotic system. In this article, an efficient, robust, and safe motion-planning algorithm is developed by learning the obstacle uncertainties online. More specifically, the unknown yet intended control set of obstacles is efficiently computed by solving a linear programming (LP) problem. The learned control set is used to compute forward reachable sets (FRSs) of obstacles that are less conservative than the worst case prediction. Based on the forward prediction, a robust model predictive controller is designed to compute a safe reference trajectory for the ego robotic system that remains outside the reachable sets of obstacles over the prediction horizon. The method is applied to a car-like mobile robot in both simulations and hardware experiments to demonstrate its effectiveness.</p>}},
author = {{Zhou, Jian and Gao, Yulong and Johansson, Ola and Olofsson, Björn and Frisk, Erik}},
issn = {{1063-6536}},
keywords = {{Predictive control; robust motion planning; safe autonomy; uncertainty quantification}},
language = {{eng}},
publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
series = {{IEEE Transactions on Control Systems Technology}},
title = {{Robust Predictive Motion Planning by Learning Obstacle Uncertainty}},
url = {{http://dx.doi.org/10.1109/TCST.2025.3533378}},
doi = {{10.1109/TCST.2025.3533378}},
year = {{2025}},
}