Year: 2025
Type: article
Abstract: Path planning algorithms for mobile robots and autonomous systems have advanced considerably, yet challenges remain in navigating complex environments while satisfying non-holonomic constraints and achieving precise target orientation. Phase portraits are traditionally used to analyse dynamical systems via equilibrium points and system trajectories, and can be a powerful framework for addressing these challenges. In this work, we propose a novel orientation-aware path planning algorithm that uses phase portrait dynamics by treating both obstacles and target poses as equilibrium points within the environment. Unlike conventional approaches, our method explicitly incorporates non-holonomic constraints and target orientation requirements, resulting in smooth, feasible trajectories with high final pose accuracy. Simulation results across 28 diverse scenarios show that our method achieves zero final orientation error with path lengths comparable to Hybrid A*, and planning times reduced by 52% on the indoor map and 84% on the playpen map relative to Hybrid A*. These results highlight the potential of phase portrait-based planning as an effective and efficient method for real-time autonomous navigation.
Source: Inventions
Authors: Abdurrahman Yilmaz, Hasan Kivrak
Institutions: Istanbul Technical University, Northumbria University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,172
OpenAlex ID: W4412995358