OkayPlan: Obstacle Kinematics Augmented Dynamic Real-time Path Planning via Particle Swarm Optimization

TL;DR

OkayPlan addresses real-time global path planning for USVs in dynamic marine environments by embedding obstacle kinematics into the optimization objective (OKAOP) and using dynamic prioritized initialization (DPI) along with a relaxation-based autonomous hyperparameter tuner. The approach achieves high-frequency planning (125 Hz) and favorable safety and length outcomes compared with baselines, demonstrated across simple, complex, and VRX marine simulators. Key innovations include OKAOP’s handling of obstacle motion, DPI’s adaptive particle initialization, and a relaxation strategy enabling robust hyperparameter evolution. The results indicate substantial practical impact for safer, energy-efficient USV navigation in dynamic settings, with promising avenues for multi-agent extensions and real-world deployment.

Abstract

Existing Global Path Planning (GPP) algorithms predominantly presume planning in static environments. This assumption immensely limits their applications to Unmanned Surface Vehicles (USVs) that typically navigate in dynamic environments. To address this limitation, we present OkayPlan, a GPP algorithm capable of generating safe and short paths in dynamic scenarios at a real-time executing speed (125 Hz on a desktop-class computer). Specifically, we approach the challenge of dynamic obstacle avoidance by formulating the path planning problem as an Obstacle Kinematics Augmented Optimization Problem (OKAOP), which can be efficiently resolved through a PSO-based optimizer at a real-time speed. Meanwhile, a Dynamic Prioritized Initialization (DPI) mechanism that adaptively initializes potential solutions for the optimization problem is established to further ameliorate the solution quality. Additionally, a relaxation strategy that facilitates the autonomous tuning of OkayPlan's hyperparameters in dynamic environments is devised. Comprehensive experiments, including comparative evaluations, ablation studies, and \textcolor{black}{applications to 3D physical simulation platforms}, have been conducted to substantiate the efficacy of our approach. Results indicate that OkayPlan outstrips existing methods in terms of path safety, length optimality, and computational efficiency, establishing it as a potent GPP technique for dynamic environments. The video and code associated with this paper are accessible at https://github.com/XinJingHao/OkayPlan.

Figures (17)

• Figure 1: An overview of OkayPlan.
• Figure 2: Problem formulation of global path planning.
• Figure 3: The evolution from PSO to SEPSO.
• Figure 4: Prioritized Initialization Mechanism.
• Figure 5: A comparison of aggressive and conservative path planning.