SMART-OC: A Real-time Time-risk Optimal Replanning Algorithm for Dynamic Obstacles and Spatio-temporally Varying Currents

TL;DR

The paper introduces SMART-OC, a real-time replanning algorithm for USVs navigating complex marine environments with dynamic obstacles and currents, optimizing safety and efficiency by balancing obstacle risk and time cost.

Contribution

It presents a novel real-time replanning algorithm that integrates obstacle risk and ocean currents for safe, efficient USV navigation in dynamic environments.

Findings

SMART-OC achieves fast replanning in simulations.

USVs successfully avoid dynamic obstacles.

USVs exploit currents to reach goals efficiently.

Abstract

Typical marine environments are highly complex with spatio-temporally varying currents and dynamic obstacles, presenting significant challenges to Unmanned Surface Vehicles (USVs) for safe and efficient navigation. Thus, the USVs need to continuously adapt their paths with real-time information to avoid collisions and follow the path of least resistance to the goal via exploiting ocean currents. In this regard, we introduce a novel algorithm, called Self-Morphing Adaptive Replanning Tree for dynamic Obstacles and Currents (SMART-OC), that facilitates real-time time-risk optimal replanning in dynamic environments. SMART-OC integrates the obstacle risks along a path with the time cost to reach the goal to find the time-risk optimal path. The effectiveness of SMART-OC is validated by simulation experiments, which demonstrate that the USV performs fast replannings to avoid dynamic obstacles and exploit ocean currents to successfully reach the goal.