Researchers Shiqian Liu, Azlan Mohd Zain, and Le-le Mao from the School of Electrical Engineering at the University of Jinan in China have developed an improved optimization algorithm for real-time path planning, which could have significant implications for the energy industry, particularly in the management of autonomous systems and drones used for inspections and maintenance.
The team’s work focuses on enhancing the Wind Driven Optimization (WDO) algorithm, a method inspired by the physics of wind that helps robots navigate complex environments. The researchers identified that while WDO shows promise, it struggles with real-time adaptability in dynamic settings. To address this, they introduced a new variant called Multi-hierarchical adaptive wind driven optimization (MAWDO), which improves adaptability and robustness in changing environments.
MAWDO introduces a hierarchical-guidance mechanism that divides the population of potential solutions into multiple groups, each guided by individual, regional, and global leaders. This approach balances exploration and exploitation, mitigating instability and premature convergence. The researchers tested MAWDO on sixteen benchmark functions and found that it achieved superior optimization accuracy, convergence stability, and adaptability compared to other state-of-the-art metaheuristics.
In dynamic path planning scenarios, MAWDO demonstrated significant improvements. It shortened the path length to 469.28 pixels, outperforming other algorithms like Multi-strategy ensemble wind driven optimization (MEWDO), Adaptive wind driven optimization (AWDO), and the original WDO by 3.51%, 11.63%, and 14.93%, respectively. Additionally, MAWDO achieved the smallest optimality gap (1.01) and smoothness (0.71), leading to smoother, shorter, and collision-free trajectories. These results confirm the effectiveness of MAWDO for real-time path planning in complex environments.
For the energy sector, this research could enhance the efficiency and reliability of autonomous systems used for tasks such as inspecting wind turbines, solar panels, and other energy infrastructure. By enabling smoother and more efficient trajectories, MAWDO could reduce the time and energy required for these inspections, ultimately lowering operational costs and improving safety.
The research was published in the journal Applied Soft Computing, a reputable source for studies on computational intelligence and its applications.
This article is based on research available at arXiv.