An adaptive and energy-aware path planning strategy (AEPPS) for ASVs with obstacle avoidance

Liu, L, Johnson, P ORCID: 0000-0003-2379-9700 and Reina, DG An adaptive and energy-aware path planning strategy (AEPPS) for ASVs with obstacle avoidance. In: The 2nd IEEE International Conference on Machine Intelligence and Smart Innovation (ICMISI 2025), 10th May - 12th May 2025, Alexandria, Egypt. (Accepted)

Preview

Text An adaptive and energy-aware path planning strategy (AEPPS) for ASVs with obstacle avoidance (1).pdf - Accepted Version Download (740kB) | Preview

Abstract

The proliferation of debris poses an increasing threat to aquatic ecosystems, necessitating scalable autonomous solutions for effective environmental remediation. This research proposes a novel hierarchical path planning framework for Autonomous Surface Vehicles (ASVs) that significantly improves energy efficiency and navigational capabilities while ensuring real-time operability in resource-constrained Autonomous Surface Vehicles (ASVs). The proposed AEPPS architecture integrates three complementary algorithms across distinct temporal scales: an energy-aware A* algorithm for global route optimization (1Hz), a biased Rapidly-Exploring Random Tree (RRT) for local obstacle negotiation (5Hz), and a Model Predictive Control (MPC) framework for trajectory refinement (10Hz). Key methodological innovations include hydrodynamic force-informed heuristics, environmental-biased sampling distributions, and adaptive prediction horizons responsive to dynamic aquatic conditions. Rigorous validation through ROS2/Gazebo simulations demonstrates statistically significant performance improvements (p<0.001) across diverse scenarios, achieving a 31.4% reduction in energy consumption and 96.4% obstacle avoidance success rate compared to baseline approaches. The system maintains resilient performance across varying obstacle densities (10-30%), current velocities (0-2 m/s), and wave patterns (up to 1.5m), with minimal degradation (4.7%) under highly dynamic conditions. Computational analysis confirms feasibility for deployment on embedded ASV systems with stringent power constraints.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Uncontrolled Keywords:	ASVs, Path planning strategies, RRT, A*, MPC, ROS2
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Engineering
Date of acceptance:	25 April 2025
Date of first compliant Open Access:	1 July 2025
Date Deposited:	01 Jul 2025 08:56
Last Modified:	03 Jul 2025 12:43
URI:	https://researchonline.ljmu.ac.uk/id/eprint/26622

View Item